dan-latn-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed10/checkpoint-36000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0070163715335783,
  "eval_steps": 3000,
  "global_step": 36000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.823197078704834,
      "epoch": 0.000417641162712997,
      "grad_norm": 13.4375,
      "learning_rate": 2e-06,
      "loss": 14.3319,
      "mean_token_accuracy": 0.0,
      "num_tokens": 8702.0,
      "step": 5
    },
    {
      "entropy": 4.829868984222412,
      "epoch": 0.000835282325425994,
      "grad_norm": 15.5,
      "learning_rate": 4.5e-06,
      "loss": 14.2442,
      "mean_token_accuracy": 0.0,
      "num_tokens": 17355.0,
      "step": 10
    },
    {
      "entropy": 4.871206903457642,
      "epoch": 0.001252923488138991,
      "grad_norm": 14.625,
      "learning_rate": 7e-06,
      "loss": 14.055,
      "mean_token_accuracy": 0.0,
      "num_tokens": 25975.0,
      "step": 15
    },
    {
      "entropy": 5.015691137313842,
      "epoch": 0.001670564650851988,
      "grad_norm": 20.5,
      "learning_rate": 9.5e-06,
      "loss": 13.6199,
      "mean_token_accuracy": 0.0,
      "num_tokens": 35099.0,
      "step": 20
    },
    {
      "entropy": 6.097980880737305,
      "epoch": 0.002088205813564985,
      "grad_norm": 21.75,
      "learning_rate": 1.2e-05,
      "loss": 12.4018,
      "mean_token_accuracy": 0.0,
      "num_tokens": 44526.0,
      "step": 25
    },
    {
      "entropy": 9.22457389831543,
      "epoch": 0.002505846976277982,
      "grad_norm": 5.375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 11.1046,
      "mean_token_accuracy": 0.0,
      "num_tokens": 53856.0,
      "step": 30
    },
    {
      "entropy": 10.673493671417237,
      "epoch": 0.002923488138990979,
      "grad_norm": 3.15625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.6655,
      "mean_token_accuracy": 0.0008215151203330607,
      "num_tokens": 63583.0,
      "step": 35
    },
    {
      "entropy": 10.704101467132569,
      "epoch": 0.003341129301703976,
      "grad_norm": 2.84375,
      "learning_rate": 1.95e-05,
      "loss": 10.3643,
      "mean_token_accuracy": 0.038832120783627036,
      "num_tokens": 71894.0,
      "step": 40
    },
    {
      "entropy": 10.683347702026367,
      "epoch": 0.003758770464416973,
      "grad_norm": 2.859375,
      "learning_rate": 2.2e-05,
      "loss": 10.049,
      "mean_token_accuracy": 0.043674240633845327,
      "num_tokens": 81234.0,
      "step": 45
    },
    {
      "entropy": 10.607063961029052,
      "epoch": 0.00417641162712997,
      "grad_norm": 2.21875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.7618,
      "mean_token_accuracy": 0.04232829064130783,
      "num_tokens": 90759.0,
      "step": 50
    },
    {
      "entropy": 10.514880847930907,
      "epoch": 0.004594052789842967,
      "grad_norm": 2.140625,
      "learning_rate": 2.7e-05,
      "loss": 9.612,
      "mean_token_accuracy": 0.04196392707526684,
      "num_tokens": 100765.0,
      "step": 55
    },
    {
      "entropy": 10.509155654907227,
      "epoch": 0.005011693952555964,
      "grad_norm": 1.96875,
      "learning_rate": 2.95e-05,
      "loss": 9.4891,
      "mean_token_accuracy": 0.04262847602367401,
      "num_tokens": 110317.0,
      "step": 60
    },
    {
      "entropy": 10.529613208770751,
      "epoch": 0.005429335115268961,
      "grad_norm": 1.9921875,
      "learning_rate": 3.2e-05,
      "loss": 9.3861,
      "mean_token_accuracy": 0.04258695356547833,
      "num_tokens": 119665.0,
      "step": 65
    },
    {
      "entropy": 10.530327224731446,
      "epoch": 0.005846976277981958,
      "grad_norm": 1.984375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.2863,
      "mean_token_accuracy": 0.04265623725950718,
      "num_tokens": 128083.0,
      "step": 70
    },
    {
      "entropy": 10.505690383911134,
      "epoch": 0.006264617440694955,
      "grad_norm": 1.9609375,
      "learning_rate": 3.7e-05,
      "loss": 9.2254,
      "mean_token_accuracy": 0.04184907637536526,
      "num_tokens": 137168.0,
      "step": 75
    },
    {
      "entropy": 10.455103874206543,
      "epoch": 0.006682258603407952,
      "grad_norm": 1.9765625,
      "learning_rate": 3.95e-05,
      "loss": 9.0861,
      "mean_token_accuracy": 0.04463305622339249,
      "num_tokens": 146023.0,
      "step": 80
    },
    {
      "entropy": 10.415075778961182,
      "epoch": 0.007099899766120949,
      "grad_norm": 1.78125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.074,
      "mean_token_accuracy": 0.0408624455332756,
      "num_tokens": 154392.0,
      "step": 85
    },
    {
      "entropy": 10.39247169494629,
      "epoch": 0.007517540928833946,
      "grad_norm": 1.8125,
      "learning_rate": 4.45e-05,
      "loss": 8.907,
      "mean_token_accuracy": 0.04568507187068462,
      "num_tokens": 163673.0,
      "step": 90
    },
    {
      "entropy": 10.380887985229492,
      "epoch": 0.007935182091546943,
      "grad_norm": 1.8671875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.8612,
      "mean_token_accuracy": 0.04349003247916698,
      "num_tokens": 172879.0,
      "step": 95
    },
    {
      "entropy": 10.331772422790527,
      "epoch": 0.00835282325425994,
      "grad_norm": 1.734375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.738,
      "mean_token_accuracy": 0.04735205471515656,
      "num_tokens": 183378.0,
      "step": 100
    },
    {
      "entropy": 10.24242639541626,
      "epoch": 0.008770464416972937,
      "grad_norm": 1.6953125,
      "learning_rate": 5.2e-05,
      "loss": 8.6205,
      "mean_token_accuracy": 0.05316540449857712,
      "num_tokens": 192255.0,
      "step": 105
    },
    {
      "entropy": 10.131509399414062,
      "epoch": 0.009188105579685933,
      "grad_norm": 1.5390625,
      "learning_rate": 5.45e-05,
      "loss": 8.4857,
      "mean_token_accuracy": 0.055044937133789065,
      "num_tokens": 202118.0,
      "step": 110
    },
    {
      "entropy": 9.957079792022705,
      "epoch": 0.00960574674239893,
      "grad_norm": 1.5234375,
      "learning_rate": 5.7e-05,
      "loss": 8.3159,
      "mean_token_accuracy": 0.05693938583135605,
      "num_tokens": 210536.0,
      "step": 115
    },
    {
      "entropy": 9.808850955963134,
      "epoch": 0.010023387905111928,
      "grad_norm": 1.40625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.1694,
      "mean_token_accuracy": 0.05625454857945442,
      "num_tokens": 219577.0,
      "step": 120
    },
    {
      "entropy": 9.689696979522704,
      "epoch": 0.010441029067824925,
      "grad_norm": 1.5234375,
      "learning_rate": 6.2e-05,
      "loss": 8.0917,
      "mean_token_accuracy": 0.05639561228454113,
      "num_tokens": 227676.0,
      "step": 125
    },
    {
      "entropy": 9.504155826568603,
      "epoch": 0.010858670230537921,
      "grad_norm": 1.25,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.9152,
      "mean_token_accuracy": 0.06222889423370361,
      "num_tokens": 236692.0,
      "step": 130
    },
    {
      "entropy": 9.237131118774414,
      "epoch": 0.011276311393250919,
      "grad_norm": 1.0625,
      "learning_rate": 6.7e-05,
      "loss": 7.901,
      "mean_token_accuracy": 0.053731896728277204,
      "num_tokens": 246143.0,
      "step": 135
    },
    {
      "entropy": 9.096350193023682,
      "epoch": 0.011693952555963916,
      "grad_norm": 1.0078125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.7313,
      "mean_token_accuracy": 0.061722924932837486,
      "num_tokens": 255263.0,
      "step": 140
    },
    {
      "entropy": 8.774121379852295,
      "epoch": 0.012111593718676913,
      "grad_norm": 1.109375,
      "learning_rate": 7.2e-05,
      "loss": 7.6964,
      "mean_token_accuracy": 0.058184793964028356,
      "num_tokens": 264219.0,
      "step": 145
    },
    {
      "entropy": 8.612203598022461,
      "epoch": 0.01252923488138991,
      "grad_norm": 0.77734375,
      "learning_rate": 7.45e-05,
      "loss": 7.6437,
      "mean_token_accuracy": 0.05833452381193638,
      "num_tokens": 274819.0,
      "step": 150
    },
    {
      "entropy": 8.410180759429931,
      "epoch": 0.012946876044102907,
      "grad_norm": 0.96484375,
      "learning_rate": 7.7e-05,
      "loss": 7.5712,
      "mean_token_accuracy": 0.05876019336283207,
      "num_tokens": 284382.0,
      "step": 155
    },
    {
      "entropy": 8.265682983398438,
      "epoch": 0.013364517206815904,
      "grad_norm": 1.0546875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.5376,
      "mean_token_accuracy": 0.06281311847269536,
      "num_tokens": 294457.0,
      "step": 160
    },
    {
      "entropy": 8.116789722442627,
      "epoch": 0.013782158369528901,
      "grad_norm": 0.91796875,
      "learning_rate": 8.2e-05,
      "loss": 7.4582,
      "mean_token_accuracy": 0.058620238304138185,
      "num_tokens": 303788.0,
      "step": 165
    },
    {
      "entropy": 8.073031854629516,
      "epoch": 0.014199799532241897,
      "grad_norm": 1.2578125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.4559,
      "mean_token_accuracy": 0.0672534491866827,
      "num_tokens": 312815.0,
      "step": 170
    },
    {
      "entropy": 8.08813018798828,
      "epoch": 0.014617440694954895,
      "grad_norm": 1.9140625,
      "learning_rate": 8.7e-05,
      "loss": 7.4682,
      "mean_token_accuracy": 0.0674412775784731,
      "num_tokens": 323599.0,
      "step": 175
    },
    {
      "entropy": 7.976608514785767,
      "epoch": 0.015035081857667892,
      "grad_norm": 1.109375,
      "learning_rate": 8.95e-05,
      "loss": 7.4279,
      "mean_token_accuracy": 0.0645599164068699,
      "num_tokens": 332788.0,
      "step": 180
    },
    {
      "entropy": 7.931798505783081,
      "epoch": 0.01545272302038089,
      "grad_norm": 0.984375,
      "learning_rate": 9.2e-05,
      "loss": 7.4383,
      "mean_token_accuracy": 0.06663466952741146,
      "num_tokens": 342528.0,
      "step": 185
    },
    {
      "entropy": 7.9960370540618895,
      "epoch": 0.015870364183093885,
      "grad_norm": 1.2421875,
      "learning_rate": 9.45e-05,
      "loss": 7.3634,
      "mean_token_accuracy": 0.07428645715117455,
      "num_tokens": 350404.0,
      "step": 190
    },
    {
      "entropy": 7.821600151062012,
      "epoch": 0.016288005345806884,
      "grad_norm": 1.2265625,
      "learning_rate": 9.7e-05,
      "loss": 7.2994,
      "mean_token_accuracy": 0.06746574863791466,
      "num_tokens": 360806.0,
      "step": 195
    },
    {
      "entropy": 7.909206867218018,
      "epoch": 0.01670564650851988,
      "grad_norm": 1.5,
      "learning_rate": 9.95e-05,
      "loss": 7.3168,
      "mean_token_accuracy": 0.06789615079760551,
      "num_tokens": 370236.0,
      "step": 200
    },
    {
      "entropy": 7.853248739242554,
      "epoch": 0.017123287671232876,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000102,
      "loss": 7.2841,
      "mean_token_accuracy": 0.07116037532687187,
      "num_tokens": 380409.0,
      "step": 205
    },
    {
      "entropy": 7.860307741165161,
      "epoch": 0.017540928833945875,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.2027,
      "mean_token_accuracy": 0.0794663280248642,
      "num_tokens": 389797.0,
      "step": 210
    },
    {
      "entropy": 7.719875431060791,
      "epoch": 0.01795856999665887,
      "grad_norm": 1.515625,
      "learning_rate": 0.000107,
      "loss": 7.1302,
      "mean_token_accuracy": 0.07808805145323276,
      "num_tokens": 398867.0,
      "step": 215
    },
    {
      "entropy": 7.7590044975280765,
      "epoch": 0.018376211159371866,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0001095,
      "loss": 7.3234,
      "mean_token_accuracy": 0.07359835654497146,
      "num_tokens": 409921.0,
      "step": 220
    },
    {
      "entropy": 7.673631000518799,
      "epoch": 0.018793852322084866,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000112,
      "loss": 7.1328,
      "mean_token_accuracy": 0.08627882376313209,
      "num_tokens": 418962.0,
      "step": 225
    },
    {
      "entropy": 7.64231309890747,
      "epoch": 0.01921149348479786,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001145,
      "loss": 7.1542,
      "mean_token_accuracy": 0.07698500528931618,
      "num_tokens": 428111.0,
      "step": 230
    },
    {
      "entropy": 7.689408874511718,
      "epoch": 0.01962913464751086,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.1785,
      "mean_token_accuracy": 0.0806326750665903,
      "num_tokens": 437370.0,
      "step": 235
    },
    {
      "entropy": 7.6315025806427,
      "epoch": 0.020046775810223856,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.0577,
      "mean_token_accuracy": 0.08123884871602058,
      "num_tokens": 446687.0,
      "step": 240
    },
    {
      "entropy": 7.64709963798523,
      "epoch": 0.020464416972936852,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000122,
      "loss": 7.2263,
      "mean_token_accuracy": 0.07417257204651832,
      "num_tokens": 455306.0,
      "step": 245
    },
    {
      "entropy": 7.604256057739258,
      "epoch": 0.02088205813564985,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001245,
      "loss": 7.1418,
      "mean_token_accuracy": 0.07368349954485894,
      "num_tokens": 465491.0,
      "step": 250
    },
    {
      "entropy": 7.6127721786499025,
      "epoch": 0.021299699298362847,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000127,
      "loss": 7.1261,
      "mean_token_accuracy": 0.08081023842096328,
      "num_tokens": 475782.0,
      "step": 255
    },
    {
      "entropy": 7.605498552322388,
      "epoch": 0.021717340461075842,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001295,
      "loss": 7.1321,
      "mean_token_accuracy": 0.0829504206776619,
      "num_tokens": 484901.0,
      "step": 260
    },
    {
      "entropy": 7.480518579483032,
      "epoch": 0.02213498162378884,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000132,
      "loss": 7.0694,
      "mean_token_accuracy": 0.07500050067901612,
      "num_tokens": 493742.0,
      "step": 265
    },
    {
      "entropy": 7.610927629470825,
      "epoch": 0.022552622786501837,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.086,
      "mean_token_accuracy": 0.07933647558093071,
      "num_tokens": 502959.0,
      "step": 270
    },
    {
      "entropy": 7.507209444046021,
      "epoch": 0.022970263949214833,
      "grad_norm": 1.140625,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.0187,
      "mean_token_accuracy": 0.08608514666557313,
      "num_tokens": 512003.0,
      "step": 275
    },
    {
      "entropy": 7.5971019744873045,
      "epoch": 0.023387905111927832,
      "grad_norm": 1.125,
      "learning_rate": 0.0001395,
      "loss": 7.1144,
      "mean_token_accuracy": 0.08404356315732002,
      "num_tokens": 520680.0,
      "step": 280
    },
    {
      "entropy": 7.484663248062134,
      "epoch": 0.023805546274640828,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.0336,
      "mean_token_accuracy": 0.08570652380585671,
      "num_tokens": 529001.0,
      "step": 285
    },
    {
      "entropy": 7.444897890090942,
      "epoch": 0.024223187437353827,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001445,
      "loss": 7.0136,
      "mean_token_accuracy": 0.09252469018101692,
      "num_tokens": 537309.0,
      "step": 290
    },
    {
      "entropy": 7.436733818054199,
      "epoch": 0.024640828600066823,
      "grad_norm": 1.21875,
      "learning_rate": 0.000147,
      "loss": 6.9432,
      "mean_token_accuracy": 0.08575214147567749,
      "num_tokens": 546436.0,
      "step": 295
    },
    {
      "entropy": 7.373424291610718,
      "epoch": 0.02505846976277982,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001495,
      "loss": 7.0205,
      "mean_token_accuracy": 0.08929547369480133,
      "num_tokens": 555634.0,
      "step": 300
    },
    {
      "entropy": 7.337024974822998,
      "epoch": 0.025476110925492818,
      "grad_norm": 1.4375,
      "learning_rate": 0.000152,
      "loss": 6.8664,
      "mean_token_accuracy": 0.08913953229784966,
      "num_tokens": 564781.0,
      "step": 305
    },
    {
      "entropy": 7.392226839065552,
      "epoch": 0.025893752088205813,
      "grad_norm": 1.125,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.0876,
      "mean_token_accuracy": 0.08700757548213005,
      "num_tokens": 575130.0,
      "step": 310
    },
    {
      "entropy": 7.466950368881226,
      "epoch": 0.02631139325091881,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000157,
      "loss": 7.0072,
      "mean_token_accuracy": 0.08388047441840171,
      "num_tokens": 584030.0,
      "step": 315
    },
    {
      "entropy": 7.381117963790894,
      "epoch": 0.026729034413631808,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0001595,
      "loss": 6.9283,
      "mean_token_accuracy": 0.08811081871390343,
      "num_tokens": 593079.0,
      "step": 320
    },
    {
      "entropy": 7.281032371520996,
      "epoch": 0.027146675576344804,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000162,
      "loss": 6.8652,
      "mean_token_accuracy": 0.10489550679922104,
      "num_tokens": 601830.0,
      "step": 325
    },
    {
      "entropy": 7.323091983795166,
      "epoch": 0.027564316739057803,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.8333,
      "mean_token_accuracy": 0.09269191026687622,
      "num_tokens": 610296.0,
      "step": 330
    },
    {
      "entropy": 7.244619131088257,
      "epoch": 0.0279819579017708,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8653,
      "mean_token_accuracy": 0.08757187873125076,
      "num_tokens": 619762.0,
      "step": 335
    },
    {
      "entropy": 7.36158595085144,
      "epoch": 0.028399599064483794,
      "grad_norm": 1.046875,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.8867,
      "mean_token_accuracy": 0.08732209205627442,
      "num_tokens": 629271.0,
      "step": 340
    },
    {
      "entropy": 7.3284117698669435,
      "epoch": 0.028817240227196794,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.9949,
      "mean_token_accuracy": 0.08122688755393029,
      "num_tokens": 638425.0,
      "step": 345
    },
    {
      "entropy": 7.380646419525147,
      "epoch": 0.02923488138990979,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.9775,
      "mean_token_accuracy": 0.08487723991274834,
      "num_tokens": 647322.0,
      "step": 350
    },
    {
      "entropy": 7.31452112197876,
      "epoch": 0.029652522552622785,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000177,
      "loss": 6.9627,
      "mean_token_accuracy": 0.08671952188014984,
      "num_tokens": 656511.0,
      "step": 355
    },
    {
      "entropy": 7.3450275421142575,
      "epoch": 0.030070163715335784,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001795,
      "loss": 6.8999,
      "mean_token_accuracy": 0.08400550261139869,
      "num_tokens": 667039.0,
      "step": 360
    },
    {
      "entropy": 7.224361610412598,
      "epoch": 0.03048780487804878,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000182,
      "loss": 6.7735,
      "mean_token_accuracy": 0.0921957716345787,
      "num_tokens": 676065.0,
      "step": 365
    },
    {
      "entropy": 7.357305717468262,
      "epoch": 0.03090544604076178,
      "grad_norm": 1.125,
      "learning_rate": 0.0001845,
      "loss": 6.9248,
      "mean_token_accuracy": 0.08780089169740676,
      "num_tokens": 684344.0,
      "step": 370
    },
    {
      "entropy": 7.180531406402588,
      "epoch": 0.03132308720347477,
      "grad_norm": 1.109375,
      "learning_rate": 0.000187,
      "loss": 6.8688,
      "mean_token_accuracy": 0.08930155485868455,
      "num_tokens": 692839.0,
      "step": 375
    },
    {
      "entropy": 7.284558629989624,
      "epoch": 0.03174072836618777,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001895,
      "loss": 6.8487,
      "mean_token_accuracy": 0.09052183702588082,
      "num_tokens": 702689.0,
      "step": 380
    },
    {
      "entropy": 7.155199098587036,
      "epoch": 0.03215836952890077,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000192,
      "loss": 6.8362,
      "mean_token_accuracy": 0.08517579063773155,
      "num_tokens": 711579.0,
      "step": 385
    },
    {
      "entropy": 7.187967967987061,
      "epoch": 0.03257601069161377,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001945,
      "loss": 6.7925,
      "mean_token_accuracy": 0.09016783386468888,
      "num_tokens": 720469.0,
      "step": 390
    },
    {
      "entropy": 7.122943782806397,
      "epoch": 0.03299365185432676,
      "grad_norm": 1.25,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.7687,
      "mean_token_accuracy": 0.08865757063031196,
      "num_tokens": 730326.0,
      "step": 395
    },
    {
      "entropy": 7.173013639450073,
      "epoch": 0.03341129301703976,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.7587,
      "mean_token_accuracy": 0.09741819649934769,
      "num_tokens": 738992.0,
      "step": 400
    },
    {
      "entropy": 7.070012474060059,
      "epoch": 0.03382893417975276,
      "grad_norm": 1.015625,
      "learning_rate": 0.000202,
      "loss": 6.8013,
      "mean_token_accuracy": 0.09208141937851906,
      "num_tokens": 748387.0,
      "step": 405
    },
    {
      "entropy": 7.231819820404053,
      "epoch": 0.03424657534246575,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.7421,
      "mean_token_accuracy": 0.09340555220842361,
      "num_tokens": 757712.0,
      "step": 410
    },
    {
      "entropy": 7.06294436454773,
      "epoch": 0.03466421650517875,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000207,
      "loss": 6.7005,
      "mean_token_accuracy": 0.09443574398756027,
      "num_tokens": 766626.0,
      "step": 415
    },
    {
      "entropy": 7.136194896697998,
      "epoch": 0.03508185766789175,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002095,
      "loss": 6.6657,
      "mean_token_accuracy": 0.09597866535186768,
      "num_tokens": 775942.0,
      "step": 420
    },
    {
      "entropy": 7.111518144607544,
      "epoch": 0.03549949883060474,
      "grad_norm": 1.21875,
      "learning_rate": 0.000212,
      "loss": 6.7432,
      "mean_token_accuracy": 0.09417268559336663,
      "num_tokens": 784680.0,
      "step": 425
    },
    {
      "entropy": 7.168673133850097,
      "epoch": 0.03591713999331774,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002145,
      "loss": 6.8297,
      "mean_token_accuracy": 0.09111174866557122,
      "num_tokens": 794107.0,
      "step": 430
    },
    {
      "entropy": 7.135743093490601,
      "epoch": 0.03633478115603074,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.79,
      "mean_token_accuracy": 0.09644902572035789,
      "num_tokens": 803584.0,
      "step": 435
    },
    {
      "entropy": 7.1497382640838625,
      "epoch": 0.03675242231874373,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002195,
      "loss": 6.7676,
      "mean_token_accuracy": 0.092482028901577,
      "num_tokens": 812460.0,
      "step": 440
    },
    {
      "entropy": 7.023286437988281,
      "epoch": 0.03717006348145673,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000222,
      "loss": 6.7414,
      "mean_token_accuracy": 0.09485919401049614,
      "num_tokens": 821946.0,
      "step": 445
    },
    {
      "entropy": 7.057104301452637,
      "epoch": 0.03758770464416973,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002245,
      "loss": 6.6867,
      "mean_token_accuracy": 0.08957326710224152,
      "num_tokens": 831781.0,
      "step": 450
    },
    {
      "entropy": 7.06196608543396,
      "epoch": 0.03800534580688272,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.6303,
      "mean_token_accuracy": 0.09696827605366706,
      "num_tokens": 840903.0,
      "step": 455
    },
    {
      "entropy": 6.991696548461914,
      "epoch": 0.03842298696959572,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.7556,
      "mean_token_accuracy": 0.09650159776210784,
      "num_tokens": 850545.0,
      "step": 460
    },
    {
      "entropy": 7.060777759552002,
      "epoch": 0.03884062813230872,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.7266,
      "mean_token_accuracy": 0.09582033678889275,
      "num_tokens": 860606.0,
      "step": 465
    },
    {
      "entropy": 7.0842451572418215,
      "epoch": 0.03925826929502172,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.7709,
      "mean_token_accuracy": 0.09876812249422073,
      "num_tokens": 868767.0,
      "step": 470
    },
    {
      "entropy": 6.999656438827515,
      "epoch": 0.03967591045773471,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000237,
      "loss": 6.6226,
      "mean_token_accuracy": 0.09780478551983833,
      "num_tokens": 877820.0,
      "step": 475
    },
    {
      "entropy": 7.04571213722229,
      "epoch": 0.04009355162044771,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002395,
      "loss": 6.7835,
      "mean_token_accuracy": 0.08980523273348809,
      "num_tokens": 887035.0,
      "step": 480
    },
    {
      "entropy": 7.02214469909668,
      "epoch": 0.04051119278316071,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000242,
      "loss": 6.7462,
      "mean_token_accuracy": 0.0942918598651886,
      "num_tokens": 896249.0,
      "step": 485
    },
    {
      "entropy": 7.005578708648682,
      "epoch": 0.040928833945873704,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002445,
      "loss": 6.6631,
      "mean_token_accuracy": 0.088001199811697,
      "num_tokens": 905307.0,
      "step": 490
    },
    {
      "entropy": 6.922316265106201,
      "epoch": 0.0413464751085867,
      "grad_norm": 1.109375,
      "learning_rate": 0.000247,
      "loss": 6.6487,
      "mean_token_accuracy": 0.0977574922144413,
      "num_tokens": 914587.0,
      "step": 495
    },
    {
      "entropy": 6.8909790992736815,
      "epoch": 0.0417641162712997,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002495,
      "loss": 6.5553,
      "mean_token_accuracy": 0.10295188128948211,
      "num_tokens": 923361.0,
      "step": 500
    },
    {
      "entropy": 6.975408411026001,
      "epoch": 0.042181757434012694,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000252,
      "loss": 6.6912,
      "mean_token_accuracy": 0.09353699833154679,
      "num_tokens": 931698.0,
      "step": 505
    },
    {
      "entropy": 6.927935075759888,
      "epoch": 0.04259939859672569,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002545,
      "loss": 6.6488,
      "mean_token_accuracy": 0.09546012356877327,
      "num_tokens": 940141.0,
      "step": 510
    },
    {
      "entropy": 6.855420064926148,
      "epoch": 0.04301703975943869,
      "grad_norm": 1.21875,
      "learning_rate": 0.000257,
      "loss": 6.6061,
      "mean_token_accuracy": 0.09668388590216637,
      "num_tokens": 948926.0,
      "step": 515
    },
    {
      "entropy": 6.977548837661743,
      "epoch": 0.043434680922151685,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002595,
      "loss": 6.6822,
      "mean_token_accuracy": 0.09463226422667503,
      "num_tokens": 957995.0,
      "step": 520
    },
    {
      "entropy": 6.9774463176727295,
      "epoch": 0.043852322084864684,
      "grad_norm": 1.1875,
      "learning_rate": 0.000262,
      "loss": 6.6742,
      "mean_token_accuracy": 0.09843497723340988,
      "num_tokens": 968004.0,
      "step": 525
    },
    {
      "entropy": 6.957279062271118,
      "epoch": 0.04426996324757768,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.7089,
      "mean_token_accuracy": 0.09921421632170677,
      "num_tokens": 977931.0,
      "step": 530
    },
    {
      "entropy": 6.856095695495606,
      "epoch": 0.044687604410290675,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6626,
      "mean_token_accuracy": 0.09421706050634385,
      "num_tokens": 986904.0,
      "step": 535
    },
    {
      "entropy": 6.8645508766174315,
      "epoch": 0.045105245573003674,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.6068,
      "mean_token_accuracy": 0.09787782728672027,
      "num_tokens": 997482.0,
      "step": 540
    },
    {
      "entropy": 6.911374950408936,
      "epoch": 0.045522886735716674,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6184,
      "mean_token_accuracy": 0.10066768899559975,
      "num_tokens": 1007196.0,
      "step": 545
    },
    {
      "entropy": 6.884297323226929,
      "epoch": 0.045940527898429666,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002745,
      "loss": 6.7243,
      "mean_token_accuracy": 0.09924571886658669,
      "num_tokens": 1016435.0,
      "step": 550
    },
    {
      "entropy": 6.889643764495849,
      "epoch": 0.046358169061142665,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000277,
      "loss": 6.5819,
      "mean_token_accuracy": 0.10183246359229088,
      "num_tokens": 1026299.0,
      "step": 555
    },
    {
      "entropy": 6.843575668334961,
      "epoch": 0.046775810223855664,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002795,
      "loss": 6.5463,
      "mean_token_accuracy": 0.10173309668898582,
      "num_tokens": 1034672.0,
      "step": 560
    },
    {
      "entropy": 6.805506658554077,
      "epoch": 0.04719345138656866,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.5295,
      "mean_token_accuracy": 0.11204413995146752,
      "num_tokens": 1042786.0,
      "step": 565
    },
    {
      "entropy": 6.775815105438232,
      "epoch": 0.047611092549281656,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002845,
      "loss": 6.5225,
      "mean_token_accuracy": 0.10164727568626404,
      "num_tokens": 1052709.0,
      "step": 570
    },
    {
      "entropy": 6.816510200500488,
      "epoch": 0.048028733711994655,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000287,
      "loss": 6.6023,
      "mean_token_accuracy": 0.0973842978477478,
      "num_tokens": 1061880.0,
      "step": 575
    },
    {
      "entropy": 6.891012907028198,
      "epoch": 0.048446374874707654,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002895,
      "loss": 6.5677,
      "mean_token_accuracy": 0.09752165898680687,
      "num_tokens": 1071234.0,
      "step": 580
    },
    {
      "entropy": 6.905015516281128,
      "epoch": 0.048864016037420646,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000292,
      "loss": 6.7021,
      "mean_token_accuracy": 0.0992009960114956,
      "num_tokens": 1081252.0,
      "step": 585
    },
    {
      "entropy": 6.769313478469849,
      "epoch": 0.049281657200133645,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002945,
      "loss": 6.5571,
      "mean_token_accuracy": 0.10143707543611527,
      "num_tokens": 1090642.0,
      "step": 590
    },
    {
      "entropy": 6.751708698272705,
      "epoch": 0.049699298362846644,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000297,
      "loss": 6.5311,
      "mean_token_accuracy": 0.10075362771749496,
      "num_tokens": 1099918.0,
      "step": 595
    },
    {
      "entropy": 6.845557498931885,
      "epoch": 0.05011693952555964,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002995,
      "loss": 6.5647,
      "mean_token_accuracy": 0.09541338533163071,
      "num_tokens": 1109521.0,
      "step": 600
    },
    {
      "entropy": 6.70610933303833,
      "epoch": 0.050534580688272636,
      "grad_norm": 1.1875,
      "learning_rate": 0.000302,
      "loss": 6.4294,
      "mean_token_accuracy": 0.11414572820067406,
      "num_tokens": 1118140.0,
      "step": 605
    },
    {
      "entropy": 6.723880195617676,
      "epoch": 0.050952221850985635,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003045,
      "loss": 6.5889,
      "mean_token_accuracy": 0.1042640283703804,
      "num_tokens": 1127591.0,
      "step": 610
    },
    {
      "entropy": 6.8105981826782225,
      "epoch": 0.05136986301369863,
      "grad_norm": 1.3125,
      "learning_rate": 0.000307,
      "loss": 6.5607,
      "mean_token_accuracy": 0.09807760342955589,
      "num_tokens": 1136223.0,
      "step": 615
    },
    {
      "entropy": 6.805627489089966,
      "epoch": 0.051787504176411626,
      "grad_norm": 1.375,
      "learning_rate": 0.0003095,
      "loss": 6.4641,
      "mean_token_accuracy": 0.10856227055191994,
      "num_tokens": 1146367.0,
      "step": 620
    },
    {
      "entropy": 6.634521865844727,
      "epoch": 0.052205145339124626,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000312,
      "loss": 6.4683,
      "mean_token_accuracy": 0.10519542768597603,
      "num_tokens": 1156230.0,
      "step": 625
    },
    {
      "entropy": 6.817045211791992,
      "epoch": 0.05262278650183762,
      "grad_norm": 1.25,
      "learning_rate": 0.0003145,
      "loss": 6.58,
      "mean_token_accuracy": 0.10044590011239052,
      "num_tokens": 1166106.0,
      "step": 630
    },
    {
      "entropy": 6.7388416767120365,
      "epoch": 0.05304042766455062,
      "grad_norm": 1.203125,
      "learning_rate": 0.000317,
      "loss": 6.5653,
      "mean_token_accuracy": 0.10486629605293274,
      "num_tokens": 1174887.0,
      "step": 635
    },
    {
      "entropy": 6.71886944770813,
      "epoch": 0.053458068827263616,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003195,
      "loss": 6.5369,
      "mean_token_accuracy": 0.11368259638547898,
      "num_tokens": 1184027.0,
      "step": 640
    },
    {
      "entropy": 6.694011211395264,
      "epoch": 0.053875709989976615,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000322,
      "loss": 6.4874,
      "mean_token_accuracy": 0.10772749111056328,
      "num_tokens": 1194632.0,
      "step": 645
    },
    {
      "entropy": 6.656815147399902,
      "epoch": 0.05429335115268961,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.4208,
      "mean_token_accuracy": 0.11698552221059799,
      "num_tokens": 1204601.0,
      "step": 650
    },
    {
      "entropy": 6.726843070983887,
      "epoch": 0.05471099231540261,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.6166,
      "mean_token_accuracy": 0.10057807937264443,
      "num_tokens": 1213288.0,
      "step": 655
    },
    {
      "entropy": 6.657552337646484,
      "epoch": 0.055128633478115606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5653,
      "mean_token_accuracy": 0.10514772906899453,
      "num_tokens": 1222768.0,
      "step": 660
    },
    {
      "entropy": 6.744826745986939,
      "epoch": 0.0555462746408286,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5451,
      "mean_token_accuracy": 0.10463840067386627,
      "num_tokens": 1232421.0,
      "step": 665
    },
    {
      "entropy": 6.6112964153289795,
      "epoch": 0.0559639158035416,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.4243,
      "mean_token_accuracy": 0.11523082107305527,
      "num_tokens": 1241230.0,
      "step": 670
    },
    {
      "entropy": 6.793651056289673,
      "epoch": 0.056381556966254597,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000337,
      "loss": 6.5711,
      "mean_token_accuracy": 0.10192721858620643,
      "num_tokens": 1250879.0,
      "step": 675
    },
    {
      "entropy": 6.692268371582031,
      "epoch": 0.05679919812896759,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003395,
      "loss": 6.4367,
      "mean_token_accuracy": 0.11580214574933052,
      "num_tokens": 1259073.0,
      "step": 680
    },
    {
      "entropy": 6.604025888442993,
      "epoch": 0.05721683929168059,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000342,
      "loss": 6.4538,
      "mean_token_accuracy": 0.10393711403012276,
      "num_tokens": 1269174.0,
      "step": 685
    },
    {
      "entropy": 6.712643575668335,
      "epoch": 0.05763448045439359,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.6139,
      "mean_token_accuracy": 0.10357280150055885,
      "num_tokens": 1279366.0,
      "step": 690
    },
    {
      "entropy": 6.651198530197144,
      "epoch": 0.05805212161710658,
      "grad_norm": 1.234375,
      "learning_rate": 0.000347,
      "loss": 6.4171,
      "mean_token_accuracy": 0.10862385183572769,
      "num_tokens": 1288879.0,
      "step": 695
    },
    {
      "entropy": 6.577065515518188,
      "epoch": 0.05846976277981958,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003495,
      "loss": 6.3147,
      "mean_token_accuracy": 0.11318397149443626,
      "num_tokens": 1298098.0,
      "step": 700
    },
    {
      "entropy": 6.5543561458587645,
      "epoch": 0.05888740394253258,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000352,
      "loss": 6.485,
      "mean_token_accuracy": 0.1087345264852047,
      "num_tokens": 1306960.0,
      "step": 705
    },
    {
      "entropy": 6.727486610412598,
      "epoch": 0.05930504510524557,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003545,
      "loss": 6.5461,
      "mean_token_accuracy": 0.10284490808844567,
      "num_tokens": 1317252.0,
      "step": 710
    },
    {
      "entropy": 6.565056228637696,
      "epoch": 0.05972268626795857,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000357,
      "loss": 6.4426,
      "mean_token_accuracy": 0.11380289196968078,
      "num_tokens": 1326712.0,
      "step": 715
    },
    {
      "entropy": 6.661278581619262,
      "epoch": 0.06014032743067157,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003595,
      "loss": 6.4932,
      "mean_token_accuracy": 0.10850499421358109,
      "num_tokens": 1336205.0,
      "step": 720
    },
    {
      "entropy": 6.617316246032715,
      "epoch": 0.06055796859338457,
      "grad_norm": 1.171875,
      "learning_rate": 0.000362,
      "loss": 6.4705,
      "mean_token_accuracy": 0.10904657989740371,
      "num_tokens": 1345347.0,
      "step": 725
    },
    {
      "entropy": 6.533994197845459,
      "epoch": 0.06097560975609756,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003645,
      "loss": 6.4027,
      "mean_token_accuracy": 0.10399781167507172,
      "num_tokens": 1354267.0,
      "step": 730
    },
    {
      "entropy": 6.5597916603088375,
      "epoch": 0.06139325091881056,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000367,
      "loss": 6.4472,
      "mean_token_accuracy": 0.10820634737610817,
      "num_tokens": 1363949.0,
      "step": 735
    },
    {
      "entropy": 6.6761926174163815,
      "epoch": 0.06181089208152356,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003695,
      "loss": 6.4872,
      "mean_token_accuracy": 0.1058058723807335,
      "num_tokens": 1374405.0,
      "step": 740
    },
    {
      "entropy": 6.449610376358033,
      "epoch": 0.06222853324423655,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000372,
      "loss": 6.3896,
      "mean_token_accuracy": 0.11227544099092483,
      "num_tokens": 1384156.0,
      "step": 745
    },
    {
      "entropy": 6.624826717376709,
      "epoch": 0.06264617440694954,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003745,
      "loss": 6.394,
      "mean_token_accuracy": 0.11085910871624946,
      "num_tokens": 1393896.0,
      "step": 750
    },
    {
      "entropy": 6.4982311725616455,
      "epoch": 0.06306381556966255,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000377,
      "loss": 6.2944,
      "mean_token_accuracy": 0.11603493541479111,
      "num_tokens": 1403842.0,
      "step": 755
    },
    {
      "entropy": 6.578821229934692,
      "epoch": 0.06348145673237554,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003795,
      "loss": 6.4013,
      "mean_token_accuracy": 0.11297919154167176,
      "num_tokens": 1412644.0,
      "step": 760
    },
    {
      "entropy": 6.474055290222168,
      "epoch": 0.06389909789508853,
      "grad_norm": 1.09375,
      "learning_rate": 0.000382,
      "loss": 6.3473,
      "mean_token_accuracy": 0.12095814496278763,
      "num_tokens": 1421815.0,
      "step": 765
    },
    {
      "entropy": 6.541539335250855,
      "epoch": 0.06431673905780154,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003845,
      "loss": 6.4752,
      "mean_token_accuracy": 0.10714265555143357,
      "num_tokens": 1431572.0,
      "step": 770
    },
    {
      "entropy": 6.482412481307984,
      "epoch": 0.06473438022051453,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.3723,
      "mean_token_accuracy": 0.11449380144476891,
      "num_tokens": 1440609.0,
      "step": 775
    },
    {
      "entropy": 6.595385599136352,
      "epoch": 0.06515202138322754,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.4756,
      "mean_token_accuracy": 0.10856133922934533,
      "num_tokens": 1451193.0,
      "step": 780
    },
    {
      "entropy": 6.449068069458008,
      "epoch": 0.06556966254594053,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.2807,
      "mean_token_accuracy": 0.11979811117053032,
      "num_tokens": 1459772.0,
      "step": 785
    },
    {
      "entropy": 6.479738807678222,
      "epoch": 0.06598730370865352,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.395,
      "mean_token_accuracy": 0.10759104117751121,
      "num_tokens": 1469251.0,
      "step": 790
    },
    {
      "entropy": 6.606679821014405,
      "epoch": 0.06640494487136653,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.5059,
      "mean_token_accuracy": 0.11142713129520417,
      "num_tokens": 1478090.0,
      "step": 795
    },
    {
      "entropy": 6.525738525390625,
      "epoch": 0.06682258603407952,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003995,
      "loss": 6.3508,
      "mean_token_accuracy": 0.11475891843438149,
      "num_tokens": 1488231.0,
      "step": 800
    },
    {
      "entropy": 6.5654744625091555,
      "epoch": 0.06724022719679251,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000402,
      "loss": 6.4112,
      "mean_token_accuracy": 0.11717421114444733,
      "num_tokens": 1496904.0,
      "step": 805
    },
    {
      "entropy": 6.42963719367981,
      "epoch": 0.06765786835950552,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004045,
      "loss": 6.3443,
      "mean_token_accuracy": 0.1201884925365448,
      "num_tokens": 1506006.0,
      "step": 810
    },
    {
      "entropy": 6.449509525299073,
      "epoch": 0.06807550952221851,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.2919,
      "mean_token_accuracy": 0.11583777889609337,
      "num_tokens": 1515826.0,
      "step": 815
    },
    {
      "entropy": 6.541458606719971,
      "epoch": 0.0684931506849315,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004095,
      "loss": 6.3781,
      "mean_token_accuracy": 0.11131022721529008,
      "num_tokens": 1525001.0,
      "step": 820
    },
    {
      "entropy": 6.363485288619995,
      "epoch": 0.06891079184764451,
      "grad_norm": 1.140625,
      "learning_rate": 0.000412,
      "loss": 6.3647,
      "mean_token_accuracy": 0.11138168722391129,
      "num_tokens": 1533647.0,
      "step": 825
    },
    {
      "entropy": 6.46032567024231,
      "epoch": 0.0693284330103575,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004145,
      "loss": 6.3677,
      "mean_token_accuracy": 0.11274689063429832,
      "num_tokens": 1542461.0,
      "step": 830
    },
    {
      "entropy": 6.431656455993652,
      "epoch": 0.0697460741730705,
      "grad_norm": 1.03125,
      "learning_rate": 0.000417,
      "loss": 6.2496,
      "mean_token_accuracy": 0.11651496887207032,
      "num_tokens": 1552249.0,
      "step": 835
    },
    {
      "entropy": 6.488580369949341,
      "epoch": 0.0701637153357835,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004195,
      "loss": 6.4204,
      "mean_token_accuracy": 0.11872618049383163,
      "num_tokens": 1561570.0,
      "step": 840
    },
    {
      "entropy": 6.569738483428955,
      "epoch": 0.07058135649849649,
      "grad_norm": 1.078125,
      "learning_rate": 0.000422,
      "loss": 6.4492,
      "mean_token_accuracy": 0.1142809510231018,
      "num_tokens": 1572263.0,
      "step": 845
    },
    {
      "entropy": 6.341160726547241,
      "epoch": 0.07099899766120948,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004245,
      "loss": 6.2765,
      "mean_token_accuracy": 0.11415041461586953,
      "num_tokens": 1580828.0,
      "step": 850
    },
    {
      "entropy": 6.458345603942871,
      "epoch": 0.07141663882392249,
      "grad_norm": 1.046875,
      "learning_rate": 0.000427,
      "loss": 6.269,
      "mean_token_accuracy": 0.11674099043011665,
      "num_tokens": 1589858.0,
      "step": 855
    },
    {
      "entropy": 6.500254964828491,
      "epoch": 0.07183427998663548,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004295,
      "loss": 6.3943,
      "mean_token_accuracy": 0.11778875812888145,
      "num_tokens": 1598822.0,
      "step": 860
    },
    {
      "entropy": 6.364452123641968,
      "epoch": 0.07225192114934847,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000432,
      "loss": 6.2859,
      "mean_token_accuracy": 0.11785781309008599,
      "num_tokens": 1607508.0,
      "step": 865
    },
    {
      "entropy": 6.370275688171387,
      "epoch": 0.07266956231206148,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004345,
      "loss": 6.3014,
      "mean_token_accuracy": 0.12148084938526153,
      "num_tokens": 1616986.0,
      "step": 870
    },
    {
      "entropy": 6.499317264556884,
      "epoch": 0.07308720347477447,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000437,
      "loss": 6.3641,
      "mean_token_accuracy": 0.11438111290335655,
      "num_tokens": 1626940.0,
      "step": 875
    },
    {
      "entropy": 6.419982290267944,
      "epoch": 0.07350484463748747,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004395,
      "loss": 6.3283,
      "mean_token_accuracy": 0.11768875047564506,
      "num_tokens": 1636249.0,
      "step": 880
    },
    {
      "entropy": 6.475973129272461,
      "epoch": 0.07392248580020047,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000442,
      "loss": 6.373,
      "mean_token_accuracy": 0.12241241857409477,
      "num_tokens": 1645943.0,
      "step": 885
    },
    {
      "entropy": 6.43145170211792,
      "epoch": 0.07434012696291346,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004445,
      "loss": 6.4209,
      "mean_token_accuracy": 0.11863628476858139,
      "num_tokens": 1655119.0,
      "step": 890
    },
    {
      "entropy": 6.374294900894165,
      "epoch": 0.07475776812562646,
      "grad_norm": 1.1875,
      "learning_rate": 0.000447,
      "loss": 6.3255,
      "mean_token_accuracy": 0.11957648620009423,
      "num_tokens": 1663635.0,
      "step": 895
    },
    {
      "entropy": 6.394734907150268,
      "epoch": 0.07517540928833946,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.1718,
      "mean_token_accuracy": 0.13117269724607467,
      "num_tokens": 1672360.0,
      "step": 900
    },
    {
      "entropy": 6.220319652557373,
      "epoch": 0.07559305045105245,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.251,
      "mean_token_accuracy": 0.1205068476498127,
      "num_tokens": 1681750.0,
      "step": 905
    },
    {
      "entropy": 6.459862565994262,
      "epoch": 0.07601069161376545,
      "grad_norm": 1.125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.2708,
      "mean_token_accuracy": 0.12147213593125343,
      "num_tokens": 1690434.0,
      "step": 910
    },
    {
      "entropy": 6.326650142669678,
      "epoch": 0.07642833277647845,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.2803,
      "mean_token_accuracy": 0.11960572525858879,
      "num_tokens": 1699982.0,
      "step": 915
    },
    {
      "entropy": 6.262086200714111,
      "epoch": 0.07684597393919144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.2058,
      "mean_token_accuracy": 0.12053424194455147,
      "num_tokens": 1709251.0,
      "step": 920
    },
    {
      "entropy": 6.2985471248626705,
      "epoch": 0.07726361510190444,
      "grad_norm": 1.3125,
      "learning_rate": 0.000462,
      "loss": 6.2876,
      "mean_token_accuracy": 0.11529200449585915,
      "num_tokens": 1718556.0,
      "step": 925
    },
    {
      "entropy": 6.278677320480346,
      "epoch": 0.07768125626461744,
      "grad_norm": 1.0,
      "learning_rate": 0.0004645,
      "loss": 6.2471,
      "mean_token_accuracy": 0.12173248827457428,
      "num_tokens": 1728265.0,
      "step": 930
    },
    {
      "entropy": 6.385457515716553,
      "epoch": 0.07809889742733044,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000467,
      "loss": 6.2542,
      "mean_token_accuracy": 0.12272520884871482,
      "num_tokens": 1737537.0,
      "step": 935
    },
    {
      "entropy": 6.194164800643921,
      "epoch": 0.07851653859004344,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004695,
      "loss": 6.1366,
      "mean_token_accuracy": 0.1311381496489048,
      "num_tokens": 1747357.0,
      "step": 940
    },
    {
      "entropy": 6.393703508377075,
      "epoch": 0.07893417975275643,
      "grad_norm": 1.078125,
      "learning_rate": 0.000472,
      "loss": 6.3576,
      "mean_token_accuracy": 0.11244316771626472,
      "num_tokens": 1757462.0,
      "step": 945
    },
    {
      "entropy": 6.308323192596435,
      "epoch": 0.07935182091546943,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004745,
      "loss": 6.1857,
      "mean_token_accuracy": 0.12801790982484818,
      "num_tokens": 1766127.0,
      "step": 950
    },
    {
      "entropy": 6.3302711009979244,
      "epoch": 0.07976946207818243,
      "grad_norm": 1.09375,
      "learning_rate": 0.000477,
      "loss": 6.2864,
      "mean_token_accuracy": 0.12558708786964418,
      "num_tokens": 1775106.0,
      "step": 955
    },
    {
      "entropy": 6.346054887771606,
      "epoch": 0.08018710324089542,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004795,
      "loss": 6.279,
      "mean_token_accuracy": 0.12811500430107117,
      "num_tokens": 1785626.0,
      "step": 960
    },
    {
      "entropy": 6.297150611877441,
      "epoch": 0.08060474440360842,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000482,
      "loss": 6.3037,
      "mean_token_accuracy": 0.11734841465950012,
      "num_tokens": 1795064.0,
      "step": 965
    },
    {
      "entropy": 6.315353155136108,
      "epoch": 0.08102238556632142,
      "grad_norm": 1.125,
      "learning_rate": 0.0004845,
      "loss": 6.2704,
      "mean_token_accuracy": 0.12119710147380829,
      "num_tokens": 1804429.0,
      "step": 970
    },
    {
      "entropy": 6.390869808197022,
      "epoch": 0.08144002672903441,
      "grad_norm": 1.0625,
      "learning_rate": 0.000487,
      "loss": 6.2921,
      "mean_token_accuracy": 0.12317592203617096,
      "num_tokens": 1813542.0,
      "step": 975
    },
    {
      "entropy": 6.2793159008026125,
      "epoch": 0.08185766789174741,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004895,
      "loss": 6.2056,
      "mean_token_accuracy": 0.12073831930756569,
      "num_tokens": 1824172.0,
      "step": 980
    },
    {
      "entropy": 6.2364945888519285,
      "epoch": 0.08227530905446041,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000492,
      "loss": 6.2292,
      "mean_token_accuracy": 0.12011673152446747,
      "num_tokens": 1832989.0,
      "step": 985
    },
    {
      "entropy": 6.346784162521362,
      "epoch": 0.0826929502171734,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004945,
      "loss": 6.1914,
      "mean_token_accuracy": 0.12780134081840516,
      "num_tokens": 1842345.0,
      "step": 990
    },
    {
      "entropy": 6.243885612487793,
      "epoch": 0.0831105913798864,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497,
      "loss": 6.2414,
      "mean_token_accuracy": 0.12187250629067421,
      "num_tokens": 1851952.0,
      "step": 995
    },
    {
      "entropy": 6.214503812789917,
      "epoch": 0.0835282325425994,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995,
      "loss": 6.2285,
      "mean_token_accuracy": 0.12935626581311227,
      "num_tokens": 1861029.0,
      "step": 1000
    },
    {
      "entropy": 6.397096061706543,
      "epoch": 0.0839458737053124,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999999987395534,
      "loss": 6.3104,
      "mean_token_accuracy": 0.12734212428331376,
      "num_tokens": 1869650.0,
      "step": 1005
    },
    {
      "entropy": 6.451109504699707,
      "epoch": 0.08436351486802539,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999999936189888,
      "loss": 6.3661,
      "mean_token_accuracy": 0.12487528920173645,
      "num_tokens": 1879078.0,
      "step": 1010
    },
    {
      "entropy": 6.1340961933135985,
      "epoch": 0.0847811560307384,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999999845595285,
      "loss": 6.2023,
      "mean_token_accuracy": 0.12111020982265472,
      "num_tokens": 1888596.0,
      "step": 1015
    },
    {
      "entropy": 6.2194208145141605,
      "epoch": 0.08519879719345139,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999999715611727,
      "loss": 6.2084,
      "mean_token_accuracy": 0.13148325234651564,
      "num_tokens": 1897342.0,
      "step": 1020
    },
    {
      "entropy": 6.287281084060669,
      "epoch": 0.08561643835616438,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999999546239215,
      "loss": 6.1531,
      "mean_token_accuracy": 0.13133467957377434,
      "num_tokens": 1905685.0,
      "step": 1025
    },
    {
      "entropy": 6.356633090972901,
      "epoch": 0.08603407951887738,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999999337477753,
      "loss": 6.3405,
      "mean_token_accuracy": 0.12002272456884384,
      "num_tokens": 1915195.0,
      "step": 1030
    },
    {
      "entropy": 6.303962469100952,
      "epoch": 0.08645172068159038,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999999089327344,
      "loss": 6.3341,
      "mean_token_accuracy": 0.11654869839549065,
      "num_tokens": 1925156.0,
      "step": 1035
    },
    {
      "entropy": 6.244997501373291,
      "epoch": 0.08686936184430337,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999998801787993,
      "loss": 6.2161,
      "mean_token_accuracy": 0.12304776981472969,
      "num_tokens": 1934521.0,
      "step": 1040
    },
    {
      "entropy": 6.287100887298584,
      "epoch": 0.08728700300701638,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999998474859704,
      "loss": 6.1296,
      "mean_token_accuracy": 0.12770192176103592,
      "num_tokens": 1944160.0,
      "step": 1045
    },
    {
      "entropy": 6.196858263015747,
      "epoch": 0.08770464416972937,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999998108542483,
      "loss": 6.1243,
      "mean_token_accuracy": 0.13160740509629248,
      "num_tokens": 1953022.0,
      "step": 1050
    },
    {
      "entropy": 6.052315664291382,
      "epoch": 0.08812228533244236,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999997702836339,
      "loss": 6.1009,
      "mean_token_accuracy": 0.1283707305788994,
      "num_tokens": 1961544.0,
      "step": 1055
    },
    {
      "entropy": 6.301678705215454,
      "epoch": 0.08853992649515537,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999997257741274,
      "loss": 6.2221,
      "mean_token_accuracy": 0.12650251016020775,
      "num_tokens": 1970319.0,
      "step": 1060
    },
    {
      "entropy": 6.141233491897583,
      "epoch": 0.08895756765786836,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049999967732573,
      "loss": 6.0681,
      "mean_token_accuracy": 0.13388170525431634,
      "num_tokens": 1979582.0,
      "step": 1065
    },
    {
      "entropy": 6.205691766738892,
      "epoch": 0.08937520882058135,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999996249384425,
      "loss": 6.2144,
      "mean_token_accuracy": 0.1302635855972767,
      "num_tokens": 1988422.0,
      "step": 1070
    },
    {
      "entropy": 6.159866285324097,
      "epoch": 0.08979284998329436,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999995686122657,
      "loss": 6.1003,
      "mean_token_accuracy": 0.13204355984926225,
      "num_tokens": 1998125.0,
      "step": 1075
    },
    {
      "entropy": 6.109376859664917,
      "epoch": 0.09021049114600735,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999995083472006,
      "loss": 6.0894,
      "mean_token_accuracy": 0.13217169865965844,
      "num_tokens": 2006993.0,
      "step": 1080
    },
    {
      "entropy": 6.156049728393555,
      "epoch": 0.09062813230872034,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999994441432481,
      "loss": 6.1508,
      "mean_token_accuracy": 0.13903949335217475,
      "num_tokens": 2016901.0,
      "step": 1085
    },
    {
      "entropy": 6.173201704025269,
      "epoch": 0.09104577347143335,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999993760004097,
      "loss": 6.1903,
      "mean_token_accuracy": 0.1270303227007389,
      "num_tokens": 2026397.0,
      "step": 1090
    },
    {
      "entropy": 6.2164756774902346,
      "epoch": 0.09146341463414634,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999993039186864,
      "loss": 6.1501,
      "mean_token_accuracy": 0.1260007508099079,
      "num_tokens": 2035241.0,
      "step": 1095
    },
    {
      "entropy": 6.195575380325318,
      "epoch": 0.09188105579685933,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999992278980794,
      "loss": 6.128,
      "mean_token_accuracy": 0.12700524181127548,
      "num_tokens": 2044559.0,
      "step": 1100
    },
    {
      "entropy": 6.161809253692627,
      "epoch": 0.09229869695957234,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999991479385901,
      "loss": 6.1697,
      "mean_token_accuracy": 0.13099708110094072,
      "num_tokens": 2055368.0,
      "step": 1105
    },
    {
      "entropy": 6.134675312042236,
      "epoch": 0.09271633812228533,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999990640402198,
      "loss": 6.0976,
      "mean_token_accuracy": 0.1294672280550003,
      "num_tokens": 2064405.0,
      "step": 1110
    },
    {
      "entropy": 6.23328275680542,
      "epoch": 0.09313397928499834,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999989762029701,
      "loss": 6.157,
      "mean_token_accuracy": 0.13345658630132676,
      "num_tokens": 2073341.0,
      "step": 1115
    },
    {
      "entropy": 6.192983865737915,
      "epoch": 0.09355162044771133,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999988844268425,
      "loss": 6.2798,
      "mean_token_accuracy": 0.12386398613452912,
      "num_tokens": 2081701.0,
      "step": 1120
    },
    {
      "entropy": 6.240271043777466,
      "epoch": 0.09396926161042432,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999987887118387,
      "loss": 6.0968,
      "mean_token_accuracy": 0.1338454693555832,
      "num_tokens": 2090662.0,
      "step": 1125
    },
    {
      "entropy": 6.076743459701538,
      "epoch": 0.09438690277313733,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999986890579602,
      "loss": 6.1238,
      "mean_token_accuracy": 0.1278868205845356,
      "num_tokens": 2100351.0,
      "step": 1130
    },
    {
      "entropy": 6.224796867370605,
      "epoch": 0.09480454393585032,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999985854652087,
      "loss": 6.1608,
      "mean_token_accuracy": 0.13114732205867768,
      "num_tokens": 2108776.0,
      "step": 1135
    },
    {
      "entropy": 6.155978584289551,
      "epoch": 0.09522218509856331,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999984779335862,
      "loss": 6.2038,
      "mean_token_accuracy": 0.13971752375364305,
      "num_tokens": 2117235.0,
      "step": 1140
    },
    {
      "entropy": 6.197133541107178,
      "epoch": 0.09563982626127632,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999983664630945,
      "loss": 6.064,
      "mean_token_accuracy": 0.1335673935711384,
      "num_tokens": 2126482.0,
      "step": 1145
    },
    {
      "entropy": 6.223146295547485,
      "epoch": 0.09605746742398931,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999982510537357,
      "loss": 6.1778,
      "mean_token_accuracy": 0.13484406024217604,
      "num_tokens": 2135949.0,
      "step": 1150
    },
    {
      "entropy": 6.113941478729248,
      "epoch": 0.0964751085867023,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999981317055116,
      "loss": 6.144,
      "mean_token_accuracy": 0.12856294363737106,
      "num_tokens": 2144887.0,
      "step": 1155
    },
    {
      "entropy": 6.211896467208862,
      "epoch": 0.09689274974941531,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999980084184244,
      "loss": 6.0844,
      "mean_token_accuracy": 0.12648672610521317,
      "num_tokens": 2153159.0,
      "step": 1160
    },
    {
      "entropy": 6.130585336685181,
      "epoch": 0.0973103909121283,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999978811924762,
      "loss": 6.1537,
      "mean_token_accuracy": 0.13049330040812493,
      "num_tokens": 2162408.0,
      "step": 1165
    },
    {
      "entropy": 6.195689630508423,
      "epoch": 0.09772803207484129,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999977500276693,
      "loss": 6.1135,
      "mean_token_accuracy": 0.12888743430376054,
      "num_tokens": 2171407.0,
      "step": 1170
    },
    {
      "entropy": 6.045321702957153,
      "epoch": 0.0981456732375543,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999976149240058,
      "loss": 5.9866,
      "mean_token_accuracy": 0.13812673091888428,
      "num_tokens": 2180281.0,
      "step": 1175
    },
    {
      "entropy": 6.090527582168579,
      "epoch": 0.09856331440026729,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999974758814884,
      "loss": 6.1043,
      "mean_token_accuracy": 0.13660623878240585,
      "num_tokens": 2188901.0,
      "step": 1180
    },
    {
      "entropy": 6.258664798736572,
      "epoch": 0.09898095556298028,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999973329001193,
      "loss": 6.2479,
      "mean_token_accuracy": 0.12757005020976067,
      "num_tokens": 2199284.0,
      "step": 1185
    },
    {
      "entropy": 6.136487579345703,
      "epoch": 0.09939859672569329,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999971859799009,
      "loss": 6.0703,
      "mean_token_accuracy": 0.13646215051412583,
      "num_tokens": 2208013.0,
      "step": 1190
    },
    {
      "entropy": 6.13501935005188,
      "epoch": 0.09981623788840628,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999970351208361,
      "loss": 6.1115,
      "mean_token_accuracy": 0.13660111427307128,
      "num_tokens": 2217444.0,
      "step": 1195
    },
    {
      "entropy": 6.019062089920044,
      "epoch": 0.10023387905111927,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999968803229275,
      "loss": 6.0133,
      "mean_token_accuracy": 0.13833465054631233,
      "num_tokens": 2226596.0,
      "step": 1200
    },
    {
      "entropy": 6.099384546279907,
      "epoch": 0.10065152021383228,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999967215861775,
      "loss": 6.0734,
      "mean_token_accuracy": 0.12789982706308364,
      "num_tokens": 2235441.0,
      "step": 1205
    },
    {
      "entropy": 6.202890872955322,
      "epoch": 0.10106916137654527,
      "grad_norm": 1.0,
      "learning_rate": 0.000499996558910589,
      "loss": 6.1548,
      "mean_token_accuracy": 0.12868092879652976,
      "num_tokens": 2244783.0,
      "step": 1210
    },
    {
      "entropy": 6.064100122451782,
      "epoch": 0.10148680253925826,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499996392296165,
      "loss": 6.022,
      "mean_token_accuracy": 0.13114323765039443,
      "num_tokens": 2254484.0,
      "step": 1215
    },
    {
      "entropy": 6.059143877029419,
      "epoch": 0.10190444370197127,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999962217429083,
      "loss": 6.0315,
      "mean_token_accuracy": 0.14302524775266648,
      "num_tokens": 2263394.0,
      "step": 1220
    },
    {
      "entropy": 5.966811752319336,
      "epoch": 0.10232208486468426,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999960472508219,
      "loss": 6.0015,
      "mean_token_accuracy": 0.1347162365913391,
      "num_tokens": 2272826.0,
      "step": 1225
    },
    {
      "entropy": 6.1172199726104735,
      "epoch": 0.10273972602739725,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499995868819909,
      "loss": 6.0999,
      "mean_token_accuracy": 0.1349620133638382,
      "num_tokens": 2282169.0,
      "step": 1230
    },
    {
      "entropy": 6.115842723846436,
      "epoch": 0.10315736719011026,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999956864501725,
      "loss": 6.0753,
      "mean_token_accuracy": 0.13311911448836328,
      "num_tokens": 2291432.0,
      "step": 1235
    },
    {
      "entropy": 6.1954936504364015,
      "epoch": 0.10357500835282325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999955001416156,
      "loss": 6.0596,
      "mean_token_accuracy": 0.13753374814987182,
      "num_tokens": 2300576.0,
      "step": 1240
    },
    {
      "entropy": 6.036804676055908,
      "epoch": 0.10399264951553625,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999953098942417,
      "loss": 6.1241,
      "mean_token_accuracy": 0.13158197253942489,
      "num_tokens": 2310790.0,
      "step": 1245
    },
    {
      "entropy": 6.110608148574829,
      "epoch": 0.10441029067824925,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999951157080541,
      "loss": 6.0361,
      "mean_token_accuracy": 0.13192710280418396,
      "num_tokens": 2319831.0,
      "step": 1250
    },
    {
      "entropy": 5.955552101135254,
      "epoch": 0.10482793184096224,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999949175830561,
      "loss": 5.8805,
      "mean_token_accuracy": 0.1360246866941452,
      "num_tokens": 2329474.0,
      "step": 1255
    },
    {
      "entropy": 6.155193710327149,
      "epoch": 0.10524557300367524,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999947155192514,
      "loss": 6.1836,
      "mean_token_accuracy": 0.12781498804688454,
      "num_tokens": 2338179.0,
      "step": 1260
    },
    {
      "entropy": 6.051835298538208,
      "epoch": 0.10566321416638824,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999945095166433,
      "loss": 5.9775,
      "mean_token_accuracy": 0.13405870422720909,
      "num_tokens": 2346899.0,
      "step": 1265
    },
    {
      "entropy": 6.139889907836914,
      "epoch": 0.10608085532910123,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999942995752354,
      "loss": 6.1316,
      "mean_token_accuracy": 0.12390393391251564,
      "num_tokens": 2356077.0,
      "step": 1270
    },
    {
      "entropy": 6.071658563613892,
      "epoch": 0.10649849649181423,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999940856950315,
      "loss": 6.0648,
      "mean_token_accuracy": 0.13771045356988906,
      "num_tokens": 2365224.0,
      "step": 1275
    },
    {
      "entropy": 6.026415109634399,
      "epoch": 0.10691613765452723,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999938678760354,
      "loss": 5.9646,
      "mean_token_accuracy": 0.1458335816860199,
      "num_tokens": 2375187.0,
      "step": 1280
    },
    {
      "entropy": 6.1658729076385494,
      "epoch": 0.10733377881724022,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999936461182507,
      "loss": 6.176,
      "mean_token_accuracy": 0.13546312600374222,
      "num_tokens": 2383460.0,
      "step": 1285
    },
    {
      "entropy": 6.111374711990356,
      "epoch": 0.10775141997995323,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999934204216814,
      "loss": 6.0838,
      "mean_token_accuracy": 0.13621672317385675,
      "num_tokens": 2392701.0,
      "step": 1290
    },
    {
      "entropy": 5.987979984283447,
      "epoch": 0.10816906114266622,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999931907863315,
      "loss": 6.0057,
      "mean_token_accuracy": 0.13854891061782837,
      "num_tokens": 2401516.0,
      "step": 1295
    },
    {
      "entropy": 6.11806287765503,
      "epoch": 0.10858670230537922,
      "grad_norm": 1.125,
      "learning_rate": 0.000499992957212205,
      "loss": 5.9854,
      "mean_token_accuracy": 0.13647850081324578,
      "num_tokens": 2410799.0,
      "step": 1300
    },
    {
      "entropy": 6.110712957382202,
      "epoch": 0.10900434346809222,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999927196993059,
      "loss": 5.9572,
      "mean_token_accuracy": 0.14616322070360183,
      "num_tokens": 2419431.0,
      "step": 1305
    },
    {
      "entropy": 5.93685736656189,
      "epoch": 0.10942198463080521,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999924782476384,
      "loss": 5.9997,
      "mean_token_accuracy": 0.14728242307901382,
      "num_tokens": 2427848.0,
      "step": 1310
    },
    {
      "entropy": 5.979397869110107,
      "epoch": 0.1098396257935182,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999922328572068,
      "loss": 5.959,
      "mean_token_accuracy": 0.1351412981748581,
      "num_tokens": 2437960.0,
      "step": 1315
    },
    {
      "entropy": 6.161928987503051,
      "epoch": 0.11025726695623121,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999919835280154,
      "loss": 6.0822,
      "mean_token_accuracy": 0.14213989973068236,
      "num_tokens": 2445835.0,
      "step": 1320
    },
    {
      "entropy": 5.935327577590942,
      "epoch": 0.1106749081189442,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999917302600684,
      "loss": 5.964,
      "mean_token_accuracy": 0.13782545179128647,
      "num_tokens": 2454556.0,
      "step": 1325
    },
    {
      "entropy": 6.0836340427398685,
      "epoch": 0.1110925492816572,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999914730533704,
      "loss": 6.0365,
      "mean_token_accuracy": 0.13404568061232566,
      "num_tokens": 2465301.0,
      "step": 1330
    },
    {
      "entropy": 6.053523588180542,
      "epoch": 0.1115101904443702,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999912119079258,
      "loss": 6.0297,
      "mean_token_accuracy": 0.13495606705546379,
      "num_tokens": 2474085.0,
      "step": 1335
    },
    {
      "entropy": 6.07110185623169,
      "epoch": 0.1119278316070832,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999909468237393,
      "loss": 5.9504,
      "mean_token_accuracy": 0.1391334593296051,
      "num_tokens": 2482439.0,
      "step": 1340
    },
    {
      "entropy": 5.964626693725586,
      "epoch": 0.11234547276979619,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999906778008153,
      "loss": 6.0029,
      "mean_token_accuracy": 0.14017563462257385,
      "num_tokens": 2491877.0,
      "step": 1345
    },
    {
      "entropy": 6.162881469726562,
      "epoch": 0.11276311393250919,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999904048391588,
      "loss": 6.1053,
      "mean_token_accuracy": 0.13479262962937355,
      "num_tokens": 2501669.0,
      "step": 1350
    },
    {
      "entropy": 5.949451684951782,
      "epoch": 0.11318075509522219,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999901279387743,
      "loss": 5.9907,
      "mean_token_accuracy": 0.1403600715100765,
      "num_tokens": 2511612.0,
      "step": 1355
    },
    {
      "entropy": 6.037490463256836,
      "epoch": 0.11359839625793518,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999898470996669,
      "loss": 5.9322,
      "mean_token_accuracy": 0.13493305146694184,
      "num_tokens": 2520791.0,
      "step": 1360
    },
    {
      "entropy": 6.008893966674805,
      "epoch": 0.11401603742064818,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999895623218415,
      "loss": 5.9933,
      "mean_token_accuracy": 0.14006343558430673,
      "num_tokens": 2529834.0,
      "step": 1365
    },
    {
      "entropy": 6.06941442489624,
      "epoch": 0.11443367858336118,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999892736053027,
      "loss": 6.0006,
      "mean_token_accuracy": 0.14169224575161934,
      "num_tokens": 2539053.0,
      "step": 1370
    },
    {
      "entropy": 5.955056619644165,
      "epoch": 0.11485131974607417,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999889809500561,
      "loss": 6.0424,
      "mean_token_accuracy": 0.1379003144800663,
      "num_tokens": 2549475.0,
      "step": 1375
    },
    {
      "entropy": 6.158879232406616,
      "epoch": 0.11526896090878717,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999886843561065,
      "loss": 6.0769,
      "mean_token_accuracy": 0.14125851467251777,
      "num_tokens": 2558718.0,
      "step": 1380
    },
    {
      "entropy": 5.925163269042969,
      "epoch": 0.11568660207150017,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999883838234592,
      "loss": 5.913,
      "mean_token_accuracy": 0.14257999509572983,
      "num_tokens": 2567454.0,
      "step": 1385
    },
    {
      "entropy": 5.924022912979126,
      "epoch": 0.11610424323421316,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999880793521194,
      "loss": 5.8263,
      "mean_token_accuracy": 0.1416394293308258,
      "num_tokens": 2575989.0,
      "step": 1390
    },
    {
      "entropy": 5.990731525421142,
      "epoch": 0.11652188439692616,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999877709420924,
      "loss": 6.0301,
      "mean_token_accuracy": 0.13861346766352653,
      "num_tokens": 2585240.0,
      "step": 1395
    },
    {
      "entropy": 6.079441165924072,
      "epoch": 0.11693952555963916,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999874585933837,
      "loss": 6.0616,
      "mean_token_accuracy": 0.1348680131137371,
      "num_tokens": 2593286.0,
      "step": 1400
    },
    {
      "entropy": 6.117265033721924,
      "epoch": 0.11735716672235215,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999871423059987,
      "loss": 6.0306,
      "mean_token_accuracy": 0.13561518117785454,
      "num_tokens": 2602992.0,
      "step": 1405
    },
    {
      "entropy": 5.919422197341919,
      "epoch": 0.11777480788506516,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499986822079943,
      "loss": 6.0325,
      "mean_token_accuracy": 0.13701212778687477,
      "num_tokens": 2611469.0,
      "step": 1410
    },
    {
      "entropy": 6.091077470779419,
      "epoch": 0.11819244904777815,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999864979152221,
      "loss": 5.8966,
      "mean_token_accuracy": 0.14297242164611818,
      "num_tokens": 2621456.0,
      "step": 1415
    },
    {
      "entropy": 5.908285808563233,
      "epoch": 0.11861009021049114,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999861698118417,
      "loss": 5.8972,
      "mean_token_accuracy": 0.13964127004146576,
      "num_tokens": 2630355.0,
      "step": 1420
    },
    {
      "entropy": 5.949282217025757,
      "epoch": 0.11902773137320415,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999858377698076,
      "loss": 5.9258,
      "mean_token_accuracy": 0.14554690718650817,
      "num_tokens": 2639508.0,
      "step": 1425
    },
    {
      "entropy": 5.863205432891846,
      "epoch": 0.11944537253591714,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999855017891256,
      "loss": 5.8353,
      "mean_token_accuracy": 0.15139148458838464,
      "num_tokens": 2648056.0,
      "step": 1430
    },
    {
      "entropy": 6.016834783554077,
      "epoch": 0.11986301369863013,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999851618698017,
      "loss": 6.0407,
      "mean_token_accuracy": 0.13517623394727707,
      "num_tokens": 2657731.0,
      "step": 1435
    },
    {
      "entropy": 6.055109977722168,
      "epoch": 0.12028065486134314,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999848180118416,
      "loss": 5.8825,
      "mean_token_accuracy": 0.14319568127393723,
      "num_tokens": 2666424.0,
      "step": 1440
    },
    {
      "entropy": 5.784654045104981,
      "epoch": 0.12069829602405613,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999844702152516,
      "loss": 5.8094,
      "mean_token_accuracy": 0.14697549343109131,
      "num_tokens": 2674646.0,
      "step": 1445
    },
    {
      "entropy": 6.079193449020385,
      "epoch": 0.12111593718676913,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999841184800374,
      "loss": 6.0074,
      "mean_token_accuracy": 0.139281989634037,
      "num_tokens": 2683295.0,
      "step": 1450
    },
    {
      "entropy": 5.972242116928101,
      "epoch": 0.12153357834948213,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999837628062055,
      "loss": 5.9096,
      "mean_token_accuracy": 0.14578085839748384,
      "num_tokens": 2692220.0,
      "step": 1455
    },
    {
      "entropy": 6.005794477462769,
      "epoch": 0.12195121951219512,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999834031937621,
      "loss": 6.1122,
      "mean_token_accuracy": 0.1357429787516594,
      "num_tokens": 2702095.0,
      "step": 1460
    },
    {
      "entropy": 6.048544406890869,
      "epoch": 0.12236886067490813,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999830396427133,
      "loss": 6.0392,
      "mean_token_accuracy": 0.13224004432559014,
      "num_tokens": 2711440.0,
      "step": 1465
    },
    {
      "entropy": 6.103414249420166,
      "epoch": 0.12278650183762112,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999826721530657,
      "loss": 6.0179,
      "mean_token_accuracy": 0.14071377664804458,
      "num_tokens": 2720564.0,
      "step": 1470
    },
    {
      "entropy": 5.971002054214478,
      "epoch": 0.12320414300033411,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999823007248255,
      "loss": 5.8644,
      "mean_token_accuracy": 0.13812775090336799,
      "num_tokens": 2729442.0,
      "step": 1475
    },
    {
      "entropy": 5.980379676818847,
      "epoch": 0.12362178416304712,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999819253579992,
      "loss": 5.9752,
      "mean_token_accuracy": 0.1382712244987488,
      "num_tokens": 2738683.0,
      "step": 1480
    },
    {
      "entropy": 5.91906156539917,
      "epoch": 0.12403942532576011,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999815460525938,
      "loss": 5.9438,
      "mean_token_accuracy": 0.13993872255086898,
      "num_tokens": 2747888.0,
      "step": 1485
    },
    {
      "entropy": 5.954296398162842,
      "epoch": 0.1244570664884731,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999811628086155,
      "loss": 5.8147,
      "mean_token_accuracy": 0.14942429587244987,
      "num_tokens": 2756667.0,
      "step": 1490
    },
    {
      "entropy": 5.962799215316773,
      "epoch": 0.1248747076511861,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499980775626071,
      "loss": 5.9992,
      "mean_token_accuracy": 0.1386643148958683,
      "num_tokens": 2765116.0,
      "step": 1495
    },
    {
      "entropy": 5.863136100769043,
      "epoch": 0.12529234881389908,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999803845049674,
      "loss": 5.8534,
      "mean_token_accuracy": 0.144050994515419,
      "num_tokens": 2774259.0,
      "step": 1500
    },
    {
      "entropy": 5.9968834400177,
      "epoch": 0.1257099899766121,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999799894453112,
      "loss": 5.8827,
      "mean_token_accuracy": 0.14217742010951043,
      "num_tokens": 2782988.0,
      "step": 1505
    },
    {
      "entropy": 5.882682275772095,
      "epoch": 0.1261276311393251,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999795904471095,
      "loss": 5.8867,
      "mean_token_accuracy": 0.14646163433790207,
      "num_tokens": 2791409.0,
      "step": 1510
    },
    {
      "entropy": 6.024690246582031,
      "epoch": 0.12654527230203808,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999791875103694,
      "loss": 5.9582,
      "mean_token_accuracy": 0.1408964402973652,
      "num_tokens": 2800644.0,
      "step": 1515
    },
    {
      "entropy": 5.89728102684021,
      "epoch": 0.12696291346475108,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999787806350977,
      "loss": 5.8566,
      "mean_token_accuracy": 0.1505585491657257,
      "num_tokens": 2809838.0,
      "step": 1520
    },
    {
      "entropy": 5.996928453445435,
      "epoch": 0.1273805546274641,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999783698213016,
      "loss": 5.979,
      "mean_token_accuracy": 0.14650033861398698,
      "num_tokens": 2818972.0,
      "step": 1525
    },
    {
      "entropy": 5.950346803665161,
      "epoch": 0.12779819579017707,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999779550689885,
      "loss": 5.9019,
      "mean_token_accuracy": 0.14394040852785112,
      "num_tokens": 2827733.0,
      "step": 1530
    },
    {
      "entropy": 6.0365362644195555,
      "epoch": 0.12821583695289007,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999775363781654,
      "loss": 5.9959,
      "mean_token_accuracy": 0.14076455160975457,
      "num_tokens": 2837300.0,
      "step": 1535
    },
    {
      "entropy": 5.97183837890625,
      "epoch": 0.12863347811560308,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999771137488395,
      "loss": 5.9156,
      "mean_token_accuracy": 0.14617259949445724,
      "num_tokens": 2846076.0,
      "step": 1540
    },
    {
      "entropy": 5.862637710571289,
      "epoch": 0.12905111927831608,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999766871810188,
      "loss": 5.8334,
      "mean_token_accuracy": 0.14296823143959045,
      "num_tokens": 2855523.0,
      "step": 1545
    },
    {
      "entropy": 5.930409288406372,
      "epoch": 0.12946876044102906,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999762566747101,
      "loss": 5.9713,
      "mean_token_accuracy": 0.13584407344460486,
      "num_tokens": 2864277.0,
      "step": 1550
    },
    {
      "entropy": 6.0211245059967045,
      "epoch": 0.12988640160374207,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999758222299214,
      "loss": 5.9505,
      "mean_token_accuracy": 0.14090201109647751,
      "num_tokens": 2873561.0,
      "step": 1555
    },
    {
      "entropy": 5.93624324798584,
      "epoch": 0.13030404276645507,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999753838466601,
      "loss": 5.8459,
      "mean_token_accuracy": 0.14749322459101677,
      "num_tokens": 2882449.0,
      "step": 1560
    },
    {
      "entropy": 5.935066652297974,
      "epoch": 0.13072168392916805,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999749415249338,
      "loss": 5.789,
      "mean_token_accuracy": 0.1558266080915928,
      "num_tokens": 2890862.0,
      "step": 1565
    },
    {
      "entropy": 5.831592226028443,
      "epoch": 0.13113932509188106,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999744952647505,
      "loss": 5.8754,
      "mean_token_accuracy": 0.14984015226364136,
      "num_tokens": 2899611.0,
      "step": 1570
    },
    {
      "entropy": 6.008182621002197,
      "epoch": 0.13155696625459407,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999740450661176,
      "loss": 5.9418,
      "mean_token_accuracy": 0.1445327214896679,
      "num_tokens": 2908111.0,
      "step": 1575
    },
    {
      "entropy": 6.0264496326446535,
      "epoch": 0.13197460741730704,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999735909290436,
      "loss": 5.9131,
      "mean_token_accuracy": 0.1450109988451004,
      "num_tokens": 2917194.0,
      "step": 1580
    },
    {
      "entropy": 5.816270923614502,
      "epoch": 0.13239224858002005,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999731328535358,
      "loss": 5.8718,
      "mean_token_accuracy": 0.14895178824663163,
      "num_tokens": 2926684.0,
      "step": 1585
    },
    {
      "entropy": 5.822729587554932,
      "epoch": 0.13280988974273306,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999726708396026,
      "loss": 5.8401,
      "mean_token_accuracy": 0.14757028147578238,
      "num_tokens": 2936094.0,
      "step": 1590
    },
    {
      "entropy": 5.919861888885498,
      "epoch": 0.13322753090544603,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499972204887252,
      "loss": 5.859,
      "mean_token_accuracy": 0.14859411120414734,
      "num_tokens": 2944389.0,
      "step": 1595
    },
    {
      "entropy": 5.91385817527771,
      "epoch": 0.13364517206815904,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999717349964922,
      "loss": 5.8669,
      "mean_token_accuracy": 0.14679256677627564,
      "num_tokens": 2954466.0,
      "step": 1600
    },
    {
      "entropy": 5.983486795425415,
      "epoch": 0.13406281323087205,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999712611673313,
      "loss": 6.0562,
      "mean_token_accuracy": 0.13626678064465522,
      "num_tokens": 2965403.0,
      "step": 1605
    },
    {
      "entropy": 6.138475179672241,
      "epoch": 0.13448045439358502,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999707833997777,
      "loss": 6.067,
      "mean_token_accuracy": 0.14294126704335214,
      "num_tokens": 2975896.0,
      "step": 1610
    },
    {
      "entropy": 5.808540487289429,
      "epoch": 0.13489809555629803,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999703016938398,
      "loss": 5.9174,
      "mean_token_accuracy": 0.1457381658256054,
      "num_tokens": 2984937.0,
      "step": 1615
    },
    {
      "entropy": 5.846891403198242,
      "epoch": 0.13531573671901104,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499969816049526,
      "loss": 5.8761,
      "mean_token_accuracy": 0.14553934037685395,
      "num_tokens": 2994052.0,
      "step": 1620
    },
    {
      "entropy": 5.9891563892364506,
      "epoch": 0.13573337788172402,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999693264668446,
      "loss": 5.7489,
      "mean_token_accuracy": 0.15324456840753556,
      "num_tokens": 3003274.0,
      "step": 1625
    },
    {
      "entropy": 5.889054298400879,
      "epoch": 0.13615101904443702,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999688329458045,
      "loss": 5.8496,
      "mean_token_accuracy": 0.1483180269598961,
      "num_tokens": 3011440.0,
      "step": 1630
    },
    {
      "entropy": 5.936949825286865,
      "epoch": 0.13656866020715003,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999683354864141,
      "loss": 5.9529,
      "mean_token_accuracy": 0.14169826358556747,
      "num_tokens": 3021032.0,
      "step": 1635
    },
    {
      "entropy": 5.959151649475098,
      "epoch": 0.136986301369863,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999678340886823,
      "loss": 5.9144,
      "mean_token_accuracy": 0.14800066351890565,
      "num_tokens": 3030160.0,
      "step": 1640
    },
    {
      "entropy": 5.9255006313323975,
      "epoch": 0.137403942532576,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999673287526176,
      "loss": 5.9463,
      "mean_token_accuracy": 0.14852431640028954,
      "num_tokens": 3039098.0,
      "step": 1645
    },
    {
      "entropy": 5.966451215744018,
      "epoch": 0.13782158369528902,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999668194782291,
      "loss": 5.8765,
      "mean_token_accuracy": 0.1453654244542122,
      "num_tokens": 3048679.0,
      "step": 1650
    },
    {
      "entropy": 5.845665216445923,
      "epoch": 0.138239224858002,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999663062655257,
      "loss": 5.8141,
      "mean_token_accuracy": 0.14851621985435487,
      "num_tokens": 3057613.0,
      "step": 1655
    },
    {
      "entropy": 5.967752313613891,
      "epoch": 0.138656866020715,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999657891145162,
      "loss": 6.0021,
      "mean_token_accuracy": 0.144612093269825,
      "num_tokens": 3066291.0,
      "step": 1660
    },
    {
      "entropy": 5.970147180557251,
      "epoch": 0.139074507183428,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999652680252099,
      "loss": 5.8986,
      "mean_token_accuracy": 0.14115611016750335,
      "num_tokens": 3075897.0,
      "step": 1665
    },
    {
      "entropy": 5.9527112483978275,
      "epoch": 0.139492148346141,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999647429976157,
      "loss": 5.8823,
      "mean_token_accuracy": 0.1441874161362648,
      "num_tokens": 3085193.0,
      "step": 1670
    },
    {
      "entropy": 5.956134605407715,
      "epoch": 0.139909789508854,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999642140317429,
      "loss": 5.88,
      "mean_token_accuracy": 0.14315434247255326,
      "num_tokens": 3094335.0,
      "step": 1675
    },
    {
      "entropy": 5.917848062515259,
      "epoch": 0.140327430671567,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999636811276007,
      "loss": 5.9943,
      "mean_token_accuracy": 0.13993434682488443,
      "num_tokens": 3103964.0,
      "step": 1680
    },
    {
      "entropy": 5.977165699005127,
      "epoch": 0.14074507183427998,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999631442851986,
      "loss": 5.8787,
      "mean_token_accuracy": 0.14386510774493216,
      "num_tokens": 3113321.0,
      "step": 1685
    },
    {
      "entropy": 5.884727621078492,
      "epoch": 0.14116271299699298,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999626035045458,
      "loss": 5.8754,
      "mean_token_accuracy": 0.14615907445549964,
      "num_tokens": 3122467.0,
      "step": 1690
    },
    {
      "entropy": 6.038534307479859,
      "epoch": 0.141580354159706,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999620587856519,
      "loss": 5.8367,
      "mean_token_accuracy": 0.14586948603391647,
      "num_tokens": 3131239.0,
      "step": 1695
    },
    {
      "entropy": 5.816018295288086,
      "epoch": 0.14199799532241897,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999615101285262,
      "loss": 5.9683,
      "mean_token_accuracy": 0.14061915054917334,
      "num_tokens": 3140568.0,
      "step": 1700
    },
    {
      "entropy": 5.981604433059692,
      "epoch": 0.14241563648513197,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999609575331786,
      "loss": 5.9145,
      "mean_token_accuracy": 0.14344506785273553,
      "num_tokens": 3151227.0,
      "step": 1705
    },
    {
      "entropy": 5.8707318782806395,
      "epoch": 0.14283327764784498,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999604009996186,
      "loss": 5.8108,
      "mean_token_accuracy": 0.15050104558467864,
      "num_tokens": 3159865.0,
      "step": 1710
    },
    {
      "entropy": 5.816216993331909,
      "epoch": 0.14325091881055796,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499959840527856,
      "loss": 5.7766,
      "mean_token_accuracy": 0.15594457387924193,
      "num_tokens": 3169832.0,
      "step": 1715
    },
    {
      "entropy": 5.904452180862426,
      "epoch": 0.14366855997327097,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999592761179007,
      "loss": 6.0157,
      "mean_token_accuracy": 0.14405087232589722,
      "num_tokens": 3180357.0,
      "step": 1720
    },
    {
      "entropy": 5.952069711685181,
      "epoch": 0.14408620113598397,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999587077697624,
      "loss": 5.8289,
      "mean_token_accuracy": 0.14626632928848265,
      "num_tokens": 3189663.0,
      "step": 1725
    },
    {
      "entropy": 5.891729259490967,
      "epoch": 0.14450384229869695,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999581354834512,
      "loss": 5.9203,
      "mean_token_accuracy": 0.14412349686026574,
      "num_tokens": 3199578.0,
      "step": 1730
    },
    {
      "entropy": 5.98569164276123,
      "epoch": 0.14492148346140996,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499957559258977,
      "loss": 5.879,
      "mean_token_accuracy": 0.1546083763241768,
      "num_tokens": 3208839.0,
      "step": 1735
    },
    {
      "entropy": 5.880684232711792,
      "epoch": 0.14533912462412296,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049995697909635,
      "loss": 5.8842,
      "mean_token_accuracy": 0.1447443701326847,
      "num_tokens": 3217449.0,
      "step": 1740
    },
    {
      "entropy": 5.84385576248169,
      "epoch": 0.14575676578683594,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999563949955803,
      "loss": 5.7728,
      "mean_token_accuracy": 0.1476930148899555,
      "num_tokens": 3226561.0,
      "step": 1745
    },
    {
      "entropy": 5.956462335586548,
      "epoch": 0.14617440694954895,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499955806956678,
      "loss": 5.8734,
      "mean_token_accuracy": 0.1562822014093399,
      "num_tokens": 3235400.0,
      "step": 1750
    },
    {
      "entropy": 5.929642915725708,
      "epoch": 0.14659204811226195,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999552149796536,
      "loss": 5.9135,
      "mean_token_accuracy": 0.14424390718340874,
      "num_tokens": 3244862.0,
      "step": 1755
    },
    {
      "entropy": 5.870706796646118,
      "epoch": 0.14700968927497493,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999546190645175,
      "loss": 5.8317,
      "mean_token_accuracy": 0.14855084866285323,
      "num_tokens": 3254370.0,
      "step": 1760
    },
    {
      "entropy": 5.925116539001465,
      "epoch": 0.14742733043768794,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999540192112799,
      "loss": 5.8673,
      "mean_token_accuracy": 0.14976725578308106,
      "num_tokens": 3264079.0,
      "step": 1765
    },
    {
      "entropy": 5.893246698379516,
      "epoch": 0.14784497160040094,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999534154199514,
      "loss": 5.9182,
      "mean_token_accuracy": 0.14768336340785027,
      "num_tokens": 3273315.0,
      "step": 1770
    },
    {
      "entropy": 5.878500699996948,
      "epoch": 0.14826261276311392,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999528076905426,
      "loss": 5.8005,
      "mean_token_accuracy": 0.1478862389922142,
      "num_tokens": 3282045.0,
      "step": 1775
    },
    {
      "entropy": 5.899237442016601,
      "epoch": 0.14868025392582693,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999521960230641,
      "loss": 5.9011,
      "mean_token_accuracy": 0.14373476877808572,
      "num_tokens": 3291596.0,
      "step": 1780
    },
    {
      "entropy": 5.820509672164917,
      "epoch": 0.14909789508853993,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999515804175268,
      "loss": 5.7256,
      "mean_token_accuracy": 0.1517377719283104,
      "num_tokens": 3300407.0,
      "step": 1785
    },
    {
      "entropy": 5.867762184143066,
      "epoch": 0.1495155362512529,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499950960873941,
      "loss": 5.7657,
      "mean_token_accuracy": 0.15600597709417344,
      "num_tokens": 3309848.0,
      "step": 1790
    },
    {
      "entropy": 5.801882362365722,
      "epoch": 0.14993317741396592,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999503373923182,
      "loss": 5.875,
      "mean_token_accuracy": 0.1472492039203644,
      "num_tokens": 3319031.0,
      "step": 1795
    },
    {
      "entropy": 5.922190523147583,
      "epoch": 0.15035081857667892,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999497099726687,
      "loss": 5.8236,
      "mean_token_accuracy": 0.14326767176389693,
      "num_tokens": 3327918.0,
      "step": 1800
    },
    {
      "entropy": 5.751756238937378,
      "epoch": 0.1507684597393919,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999490786150039,
      "loss": 5.8449,
      "mean_token_accuracy": 0.14434421062469482,
      "num_tokens": 3337071.0,
      "step": 1805
    },
    {
      "entropy": 6.079517030715943,
      "epoch": 0.1511861009021049,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999484433193348,
      "loss": 5.9039,
      "mean_token_accuracy": 0.14523821473121643,
      "num_tokens": 3346334.0,
      "step": 1810
    },
    {
      "entropy": 5.8847856521606445,
      "epoch": 0.15160374206481791,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999478040856722,
      "loss": 5.8307,
      "mean_token_accuracy": 0.1462329663336277,
      "num_tokens": 3354932.0,
      "step": 1815
    },
    {
      "entropy": 5.716736555099487,
      "epoch": 0.1520213832275309,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999471609140276,
      "loss": 5.6693,
      "mean_token_accuracy": 0.15522432029247285,
      "num_tokens": 3364345.0,
      "step": 1820
    },
    {
      "entropy": 5.929656648635865,
      "epoch": 0.1524390243902439,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999465138044122,
      "loss": 5.8707,
      "mean_token_accuracy": 0.1472846657037735,
      "num_tokens": 3372918.0,
      "step": 1825
    },
    {
      "entropy": 5.718468713760376,
      "epoch": 0.1528566655529569,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999458627568373,
      "loss": 5.7288,
      "mean_token_accuracy": 0.1567766509950161,
      "num_tokens": 3381815.0,
      "step": 1830
    },
    {
      "entropy": 5.843803644180298,
      "epoch": 0.15327430671566988,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999452077713144,
      "loss": 5.7803,
      "mean_token_accuracy": 0.14712713211774825,
      "num_tokens": 3391057.0,
      "step": 1835
    },
    {
      "entropy": 6.03563289642334,
      "epoch": 0.1536919478783829,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999445488478547,
      "loss": 5.961,
      "mean_token_accuracy": 0.1439568117260933,
      "num_tokens": 3401254.0,
      "step": 1840
    },
    {
      "entropy": 5.808206081390381,
      "epoch": 0.1541095890410959,
      "grad_norm": 0.953125,
      "learning_rate": 0.00049994388598647,
      "loss": 5.8897,
      "mean_token_accuracy": 0.14927712976932525,
      "num_tokens": 3411031.0,
      "step": 1845
    },
    {
      "entropy": 5.898315477371216,
      "epoch": 0.15452723020380887,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999432191871719,
      "loss": 5.8194,
      "mean_token_accuracy": 0.1547455981373787,
      "num_tokens": 3420154.0,
      "step": 1850
    },
    {
      "entropy": 5.990719413757324,
      "epoch": 0.15494487136652188,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999425484499718,
      "loss": 5.912,
      "mean_token_accuracy": 0.1456889122724533,
      "num_tokens": 3429285.0,
      "step": 1855
    },
    {
      "entropy": 5.868764209747314,
      "epoch": 0.1553625125292349,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999418737748817,
      "loss": 5.8969,
      "mean_token_accuracy": 0.14278291612863542,
      "num_tokens": 3438065.0,
      "step": 1860
    },
    {
      "entropy": 6.021544933319092,
      "epoch": 0.15578015369194786,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999411951619134,
      "loss": 5.877,
      "mean_token_accuracy": 0.145746049284935,
      "num_tokens": 3447534.0,
      "step": 1865
    },
    {
      "entropy": 5.836647796630859,
      "epoch": 0.15619779485466087,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999405126110786,
      "loss": 5.8529,
      "mean_token_accuracy": 0.15536943227052688,
      "num_tokens": 3456435.0,
      "step": 1870
    },
    {
      "entropy": 5.903014087677002,
      "epoch": 0.15661543601737388,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999398261223895,
      "loss": 5.8192,
      "mean_token_accuracy": 0.15386642515659332,
      "num_tokens": 3464982.0,
      "step": 1875
    },
    {
      "entropy": 5.938276910781861,
      "epoch": 0.15703307718008688,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999391356958579,
      "loss": 5.9591,
      "mean_token_accuracy": 0.14135469049215316,
      "num_tokens": 3474585.0,
      "step": 1880
    },
    {
      "entropy": 5.797132873535157,
      "epoch": 0.15745071834279986,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499938441331496,
      "loss": 5.7931,
      "mean_token_accuracy": 0.1504616156220436,
      "num_tokens": 3483164.0,
      "step": 1885
    },
    {
      "entropy": 5.866785955429077,
      "epoch": 0.15786835950551287,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999377430293159,
      "loss": 5.7921,
      "mean_token_accuracy": 0.15038491189479827,
      "num_tokens": 3492984.0,
      "step": 1890
    },
    {
      "entropy": 5.715900993347168,
      "epoch": 0.15828600066822587,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999370407893299,
      "loss": 5.6347,
      "mean_token_accuracy": 0.15855061560869216,
      "num_tokens": 3500356.0,
      "step": 1895
    },
    {
      "entropy": 5.965412521362305,
      "epoch": 0.15870364183093885,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999363346115501,
      "loss": 5.8402,
      "mean_token_accuracy": 0.1494261384010315,
      "num_tokens": 3509792.0,
      "step": 1900
    },
    {
      "entropy": 5.812705039978027,
      "epoch": 0.15912128299365186,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999356244959893,
      "loss": 5.7588,
      "mean_token_accuracy": 0.1485971011221409,
      "num_tokens": 3519340.0,
      "step": 1905
    },
    {
      "entropy": 5.796240758895874,
      "epoch": 0.15953892415636486,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999349104426594,
      "loss": 5.7562,
      "mean_token_accuracy": 0.155721578001976,
      "num_tokens": 3528846.0,
      "step": 1910
    },
    {
      "entropy": 5.790179777145386,
      "epoch": 0.15995656531907784,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999341924515732,
      "loss": 5.7495,
      "mean_token_accuracy": 0.15263200104236602,
      "num_tokens": 3537766.0,
      "step": 1915
    },
    {
      "entropy": 5.85261492729187,
      "epoch": 0.16037420648179085,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999334705227431,
      "loss": 5.8082,
      "mean_token_accuracy": 0.1515938565135002,
      "num_tokens": 3546724.0,
      "step": 1920
    },
    {
      "entropy": 5.695097827911377,
      "epoch": 0.16079184764450385,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499932744656182,
      "loss": 5.721,
      "mean_token_accuracy": 0.15915379524230958,
      "num_tokens": 3555599.0,
      "step": 1925
    },
    {
      "entropy": 5.895123863220215,
      "epoch": 0.16120948880721683,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999320148519024,
      "loss": 5.7653,
      "mean_token_accuracy": 0.15158458054065704,
      "num_tokens": 3564754.0,
      "step": 1930
    },
    {
      "entropy": 5.59046483039856,
      "epoch": 0.16162712996992984,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999312811099171,
      "loss": 5.6248,
      "mean_token_accuracy": 0.15768528878688812,
      "num_tokens": 3573166.0,
      "step": 1935
    },
    {
      "entropy": 5.958360719680786,
      "epoch": 0.16204477113264285,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999305434302391,
      "loss": 5.8977,
      "mean_token_accuracy": 0.14737681299448013,
      "num_tokens": 3581602.0,
      "step": 1940
    },
    {
      "entropy": 5.824247026443482,
      "epoch": 0.16246241229535582,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499929801812881,
      "loss": 5.6999,
      "mean_token_accuracy": 0.16191418766975402,
      "num_tokens": 3590946.0,
      "step": 1945
    },
    {
      "entropy": 5.746830797195434,
      "epoch": 0.16288005345806883,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999290562578561,
      "loss": 5.778,
      "mean_token_accuracy": 0.1522221401333809,
      "num_tokens": 3599888.0,
      "step": 1950
    },
    {
      "entropy": 5.830930376052857,
      "epoch": 0.16329769462078184,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999283067651773,
      "loss": 5.7984,
      "mean_token_accuracy": 0.15850079655647278,
      "num_tokens": 3609773.0,
      "step": 1955
    },
    {
      "entropy": 5.804241085052491,
      "epoch": 0.16371533578349481,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999275533348577,
      "loss": 5.749,
      "mean_token_accuracy": 0.15674369931221008,
      "num_tokens": 3618820.0,
      "step": 1960
    },
    {
      "entropy": 5.864436960220337,
      "epoch": 0.16413297694620782,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999267959669106,
      "loss": 5.8384,
      "mean_token_accuracy": 0.1484947569668293,
      "num_tokens": 3629010.0,
      "step": 1965
    },
    {
      "entropy": 5.776125764846801,
      "epoch": 0.16455061810892083,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999260346613491,
      "loss": 5.771,
      "mean_token_accuracy": 0.15559061765670776,
      "num_tokens": 3636915.0,
      "step": 1970
    },
    {
      "entropy": 5.798781251907348,
      "epoch": 0.1649682592716338,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999252694181867,
      "loss": 5.7535,
      "mean_token_accuracy": 0.1556951105594635,
      "num_tokens": 3646011.0,
      "step": 1975
    },
    {
      "entropy": 5.862015438079834,
      "epoch": 0.1653859004343468,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999245002374367,
      "loss": 5.7275,
      "mean_token_accuracy": 0.15725564807653428,
      "num_tokens": 3655059.0,
      "step": 1980
    },
    {
      "entropy": 5.859493160247803,
      "epoch": 0.16580354159705982,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999237271191125,
      "loss": 5.8152,
      "mean_token_accuracy": 0.15456283688545228,
      "num_tokens": 3664553.0,
      "step": 1985
    },
    {
      "entropy": 5.841559171676636,
      "epoch": 0.1662211827597728,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999229500632279,
      "loss": 5.7509,
      "mean_token_accuracy": 0.1524602711200714,
      "num_tokens": 3673604.0,
      "step": 1990
    },
    {
      "entropy": 5.810849189758301,
      "epoch": 0.1666388239224858,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999221690697961,
      "loss": 5.726,
      "mean_token_accuracy": 0.15616438090801238,
      "num_tokens": 3682634.0,
      "step": 1995
    },
    {
      "entropy": 5.778779172897339,
      "epoch": 0.1670564650851988,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999213841388312,
      "loss": 5.8043,
      "mean_token_accuracy": 0.1506836846470833,
      "num_tokens": 3691961.0,
      "step": 2000
    },
    {
      "entropy": 5.758075857162476,
      "epoch": 0.16747410624791179,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999205952703466,
      "loss": 5.7759,
      "mean_token_accuracy": 0.14631137028336524,
      "num_tokens": 3701412.0,
      "step": 2005
    },
    {
      "entropy": 5.8458953380584715,
      "epoch": 0.1678917474106248,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999198024643563,
      "loss": 5.8254,
      "mean_token_accuracy": 0.1556076392531395,
      "num_tokens": 3710449.0,
      "step": 2010
    },
    {
      "entropy": 5.8543816089630125,
      "epoch": 0.1683093885733378,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999190057208741,
      "loss": 5.7846,
      "mean_token_accuracy": 0.14759426265954972,
      "num_tokens": 3718654.0,
      "step": 2015
    },
    {
      "entropy": 5.727865552902221,
      "epoch": 0.16872702973605078,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499918205039914,
      "loss": 5.7796,
      "mean_token_accuracy": 0.1492263525724411,
      "num_tokens": 3728660.0,
      "step": 2020
    },
    {
      "entropy": 5.855515861511231,
      "epoch": 0.16914467089876378,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999174004214899,
      "loss": 5.8027,
      "mean_token_accuracy": 0.15391339361667633,
      "num_tokens": 3738085.0,
      "step": 2025
    },
    {
      "entropy": 5.804375648498535,
      "epoch": 0.1695623120614768,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999165918656161,
      "loss": 5.728,
      "mean_token_accuracy": 0.15261952951550484,
      "num_tokens": 3747727.0,
      "step": 2030
    },
    {
      "entropy": 5.843009185791016,
      "epoch": 0.16997995322418977,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999157793723065,
      "loss": 5.8405,
      "mean_token_accuracy": 0.14722344428300857,
      "num_tokens": 3757864.0,
      "step": 2035
    },
    {
      "entropy": 5.736613512039185,
      "epoch": 0.17039759438690277,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999149629415755,
      "loss": 5.7209,
      "mean_token_accuracy": 0.15391422659158707,
      "num_tokens": 3766843.0,
      "step": 2040
    },
    {
      "entropy": 5.846935081481933,
      "epoch": 0.17081523554961578,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999141425734374,
      "loss": 5.7794,
      "mean_token_accuracy": 0.15506419092416762,
      "num_tokens": 3776171.0,
      "step": 2045
    },
    {
      "entropy": 5.807502889633179,
      "epoch": 0.17123287671232876,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999133182679064,
      "loss": 5.7536,
      "mean_token_accuracy": 0.15907568335533143,
      "num_tokens": 3785545.0,
      "step": 2050
    },
    {
      "entropy": 5.7233068466186525,
      "epoch": 0.17165051787504176,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999124900249971,
      "loss": 5.7009,
      "mean_token_accuracy": 0.15531886667013167,
      "num_tokens": 3794110.0,
      "step": 2055
    },
    {
      "entropy": 5.827488946914673,
      "epoch": 0.17206815903775477,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999116578447239,
      "loss": 5.8275,
      "mean_token_accuracy": 0.15153776109218597,
      "num_tokens": 3803727.0,
      "step": 2060
    },
    {
      "entropy": 5.832465267181396,
      "epoch": 0.17248580020046775,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999108217271015,
      "loss": 5.8038,
      "mean_token_accuracy": 0.14992659762501717,
      "num_tokens": 3812904.0,
      "step": 2065
    },
    {
      "entropy": 6.044683599472046,
      "epoch": 0.17290344136318075,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999099816721444,
      "loss": 5.8619,
      "mean_token_accuracy": 0.14960845410823823,
      "num_tokens": 3822009.0,
      "step": 2070
    },
    {
      "entropy": 5.7061169147491455,
      "epoch": 0.17332108252589376,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999091376798673,
      "loss": 5.7898,
      "mean_token_accuracy": 0.16072025448083876,
      "num_tokens": 3830877.0,
      "step": 2075
    },
    {
      "entropy": 5.975783109664917,
      "epoch": 0.17373872368860674,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999082897502849,
      "loss": 5.8733,
      "mean_token_accuracy": 0.1454279713332653,
      "num_tokens": 3839876.0,
      "step": 2080
    },
    {
      "entropy": 5.8507367134094235,
      "epoch": 0.17415636485131974,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999074378834123,
      "loss": 5.8361,
      "mean_token_accuracy": 0.15541226267814637,
      "num_tokens": 3848769.0,
      "step": 2085
    },
    {
      "entropy": 5.8066075325012205,
      "epoch": 0.17457400601403275,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999065820792644,
      "loss": 5.6796,
      "mean_token_accuracy": 0.15266919434070586,
      "num_tokens": 3857676.0,
      "step": 2090
    },
    {
      "entropy": 5.805453491210938,
      "epoch": 0.17499164717674573,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999057223378559,
      "loss": 5.7524,
      "mean_token_accuracy": 0.15749767869710923,
      "num_tokens": 3867098.0,
      "step": 2095
    },
    {
      "entropy": 5.636935901641846,
      "epoch": 0.17540928833945874,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499904858659202,
      "loss": 5.71,
      "mean_token_accuracy": 0.15111172050237656,
      "num_tokens": 3876247.0,
      "step": 2100
    },
    {
      "entropy": 5.827657556533813,
      "epoch": 0.17582692950217174,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999039910433179,
      "loss": 5.7641,
      "mean_token_accuracy": 0.15805181115865707,
      "num_tokens": 3885747.0,
      "step": 2105
    },
    {
      "entropy": 5.9260594844818115,
      "epoch": 0.17624457066488472,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999031194902187,
      "loss": 5.879,
      "mean_token_accuracy": 0.15000726729631425,
      "num_tokens": 3896077.0,
      "step": 2110
    },
    {
      "entropy": 5.862950849533081,
      "epoch": 0.17666221182759773,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999022439999198,
      "loss": 5.8825,
      "mean_token_accuracy": 0.1381701298058033,
      "num_tokens": 3906323.0,
      "step": 2115
    },
    {
      "entropy": 5.887103509902954,
      "epoch": 0.17707985299031073,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999013645724362,
      "loss": 5.7701,
      "mean_token_accuracy": 0.15319084525108337,
      "num_tokens": 3915510.0,
      "step": 2120
    },
    {
      "entropy": 5.845855331420898,
      "epoch": 0.1774974941530237,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999004812077836,
      "loss": 5.8381,
      "mean_token_accuracy": 0.15369762405753135,
      "num_tokens": 3924799.0,
      "step": 2125
    },
    {
      "entropy": 5.783837032318115,
      "epoch": 0.17791513531573672,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998995939059773,
      "loss": 5.7876,
      "mean_token_accuracy": 0.1510419577360153,
      "num_tokens": 3933675.0,
      "step": 2130
    },
    {
      "entropy": 5.8255650997161865,
      "epoch": 0.17833277647844972,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499898702667033,
      "loss": 5.7387,
      "mean_token_accuracy": 0.15646950602531434,
      "num_tokens": 3942086.0,
      "step": 2135
    },
    {
      "entropy": 5.797743606567383,
      "epoch": 0.1787504176411627,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998978074909661,
      "loss": 5.7548,
      "mean_token_accuracy": 0.15239814668893814,
      "num_tokens": 3950626.0,
      "step": 2140
    },
    {
      "entropy": 5.686432552337647,
      "epoch": 0.1791680588038757,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998969083777925,
      "loss": 5.6455,
      "mean_token_accuracy": 0.16219130754470826,
      "num_tokens": 3959006.0,
      "step": 2145
    },
    {
      "entropy": 5.741449356079102,
      "epoch": 0.1795856999665887,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998960053275277,
      "loss": 5.7499,
      "mean_token_accuracy": 0.15800593346357344,
      "num_tokens": 3968177.0,
      "step": 2150
    },
    {
      "entropy": 5.87012619972229,
      "epoch": 0.1800033411293017,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998950983401875,
      "loss": 5.7029,
      "mean_token_accuracy": 0.15574318766593934,
      "num_tokens": 3977052.0,
      "step": 2155
    },
    {
      "entropy": 5.694927978515625,
      "epoch": 0.1804209822920147,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998941874157882,
      "loss": 5.712,
      "mean_token_accuracy": 0.15181038826704024,
      "num_tokens": 3985101.0,
      "step": 2160
    },
    {
      "entropy": 5.757119941711426,
      "epoch": 0.1808386234547277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998932725543453,
      "loss": 5.6385,
      "mean_token_accuracy": 0.16106333136558532,
      "num_tokens": 3992936.0,
      "step": 2165
    },
    {
      "entropy": 5.815750551223755,
      "epoch": 0.18125626461744068,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998923537558749,
      "loss": 5.8182,
      "mean_token_accuracy": 0.15310252755880355,
      "num_tokens": 4001702.0,
      "step": 2170
    },
    {
      "entropy": 5.8231817245483395,
      "epoch": 0.1816739057801537,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998914310203932,
      "loss": 5.8016,
      "mean_token_accuracy": 0.15513441413640977,
      "num_tokens": 4011639.0,
      "step": 2175
    },
    {
      "entropy": 5.750963401794434,
      "epoch": 0.1820915469428667,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998905043479163,
      "loss": 5.6073,
      "mean_token_accuracy": 0.16204652339220046,
      "num_tokens": 4021420.0,
      "step": 2180
    },
    {
      "entropy": 5.661520290374756,
      "epoch": 0.18250918810557967,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998895737384605,
      "loss": 5.6915,
      "mean_token_accuracy": 0.14866186380386354,
      "num_tokens": 4030869.0,
      "step": 2185
    },
    {
      "entropy": 5.72034273147583,
      "epoch": 0.18292682926829268,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998886391920418,
      "loss": 5.5454,
      "mean_token_accuracy": 0.16509579569101335,
      "num_tokens": 4039573.0,
      "step": 2190
    },
    {
      "entropy": 5.606214094161987,
      "epoch": 0.18334447043100568,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998877007086769,
      "loss": 5.6672,
      "mean_token_accuracy": 0.15629229545593262,
      "num_tokens": 4048684.0,
      "step": 2195
    },
    {
      "entropy": 5.723562002182007,
      "epoch": 0.18376211159371866,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004998867582883821,
      "loss": 5.8047,
      "mean_token_accuracy": 0.15196271240711212,
      "num_tokens": 4057968.0,
      "step": 2200
    },
    {
      "entropy": 5.904633522033691,
      "epoch": 0.18417975275643167,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998858119311739,
      "loss": 5.7148,
      "mean_token_accuracy": 0.15125639140605926,
      "num_tokens": 4068057.0,
      "step": 2205
    },
    {
      "entropy": 5.75338134765625,
      "epoch": 0.18459739391914468,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998848616370689,
      "loss": 5.6403,
      "mean_token_accuracy": 0.15991646945476531,
      "num_tokens": 4077141.0,
      "step": 2210
    },
    {
      "entropy": 5.744376277923584,
      "epoch": 0.18501503508185768,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998839074060837,
      "loss": 5.7003,
      "mean_token_accuracy": 0.15504838824272155,
      "num_tokens": 4085227.0,
      "step": 2215
    },
    {
      "entropy": 5.732967853546143,
      "epoch": 0.18543267624457066,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499882949238235,
      "loss": 5.6491,
      "mean_token_accuracy": 0.16520010381937028,
      "num_tokens": 4093674.0,
      "step": 2220
    },
    {
      "entropy": 5.81893310546875,
      "epoch": 0.18585031740728367,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998819871335396,
      "loss": 5.8264,
      "mean_token_accuracy": 0.1498129650950432,
      "num_tokens": 4103466.0,
      "step": 2225
    },
    {
      "entropy": 5.882892417907715,
      "epoch": 0.18626795856999667,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998810210920142,
      "loss": 5.6579,
      "mean_token_accuracy": 0.1567348837852478,
      "num_tokens": 4111833.0,
      "step": 2230
    },
    {
      "entropy": 5.696685934066773,
      "epoch": 0.18668559973270965,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998800511136759,
      "loss": 5.7082,
      "mean_token_accuracy": 0.1504954993724823,
      "num_tokens": 4121546.0,
      "step": 2235
    },
    {
      "entropy": 5.723720836639404,
      "epoch": 0.18710324089542266,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998790771985417,
      "loss": 5.6727,
      "mean_token_accuracy": 0.16130448281764984,
      "num_tokens": 4130408.0,
      "step": 2240
    },
    {
      "entropy": 5.756785297393799,
      "epoch": 0.18752088205813566,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998780993466284,
      "loss": 5.8075,
      "mean_token_accuracy": 0.15824081003665924,
      "num_tokens": 4139793.0,
      "step": 2245
    },
    {
      "entropy": 5.864732646942139,
      "epoch": 0.18793852322084864,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998771175579534,
      "loss": 5.6672,
      "mean_token_accuracy": 0.16205901950597762,
      "num_tokens": 4148192.0,
      "step": 2250
    },
    {
      "entropy": 5.560827684402466,
      "epoch": 0.18835616438356165,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998761318325337,
      "loss": 5.6092,
      "mean_token_accuracy": 0.15886650383472442,
      "num_tokens": 4157960.0,
      "step": 2255
    },
    {
      "entropy": 5.59301118850708,
      "epoch": 0.18877380554627465,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998751421703868,
      "loss": 5.4706,
      "mean_token_accuracy": 0.16533980667591094,
      "num_tokens": 4166356.0,
      "step": 2260
    },
    {
      "entropy": 5.80047197341919,
      "epoch": 0.18919144670898763,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998741485715297,
      "loss": 5.8187,
      "mean_token_accuracy": 0.14628092870116233,
      "num_tokens": 4176139.0,
      "step": 2265
    },
    {
      "entropy": 5.802434635162354,
      "epoch": 0.18960908787170064,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049987315103598,
      "loss": 5.67,
      "mean_token_accuracy": 0.15827721282839774,
      "num_tokens": 4185583.0,
      "step": 2270
    },
    {
      "entropy": 5.715236043930053,
      "epoch": 0.19002672903441364,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998721495637551,
      "loss": 5.78,
      "mean_token_accuracy": 0.1517994225025177,
      "num_tokens": 4194275.0,
      "step": 2275
    },
    {
      "entropy": 5.754878044128418,
      "epoch": 0.19044437019712662,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998711441548726,
      "loss": 5.7196,
      "mean_token_accuracy": 0.15217152535915374,
      "num_tokens": 4203424.0,
      "step": 2280
    },
    {
      "entropy": 5.72383394241333,
      "epoch": 0.19086201135983963,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998701348093499,
      "loss": 5.6788,
      "mean_token_accuracy": 0.15740089416503905,
      "num_tokens": 4212452.0,
      "step": 2285
    },
    {
      "entropy": 5.798124599456787,
      "epoch": 0.19127965252255263,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499869121527205,
      "loss": 5.7826,
      "mean_token_accuracy": 0.15464788377285005,
      "num_tokens": 4221639.0,
      "step": 2290
    },
    {
      "entropy": 5.684769153594971,
      "epoch": 0.1916972936852656,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998681043084553,
      "loss": 5.6652,
      "mean_token_accuracy": 0.15623509585857392,
      "num_tokens": 4231362.0,
      "step": 2295
    },
    {
      "entropy": 5.801016712188721,
      "epoch": 0.19211493484797862,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998670831531188,
      "loss": 5.7564,
      "mean_token_accuracy": 0.15280152559280397,
      "num_tokens": 4241035.0,
      "step": 2300
    },
    {
      "entropy": 5.819127893447876,
      "epoch": 0.19253257601069163,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998660580612135,
      "loss": 5.6748,
      "mean_token_accuracy": 0.15776295959949493,
      "num_tokens": 4250275.0,
      "step": 2305
    },
    {
      "entropy": 5.641567945480347,
      "epoch": 0.1929502171734046,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998650290327569,
      "loss": 5.7706,
      "mean_token_accuracy": 0.14925756752490998,
      "num_tokens": 4259952.0,
      "step": 2310
    },
    {
      "entropy": 5.823996067047119,
      "epoch": 0.1933678583361176,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998639960677675,
      "loss": 5.6086,
      "mean_token_accuracy": 0.1649850994348526,
      "num_tokens": 4269985.0,
      "step": 2315
    },
    {
      "entropy": 5.620763301849365,
      "epoch": 0.19378549949883062,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998629591662631,
      "loss": 5.6092,
      "mean_token_accuracy": 0.1591414988040924,
      "num_tokens": 4278317.0,
      "step": 2320
    },
    {
      "entropy": 5.700749444961548,
      "epoch": 0.1942031406615436,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499861918328262,
      "loss": 5.6617,
      "mean_token_accuracy": 0.15998956114053725,
      "num_tokens": 4286633.0,
      "step": 2325
    },
    {
      "entropy": 5.732091760635376,
      "epoch": 0.1946207818242566,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998608735537822,
      "loss": 5.6416,
      "mean_token_accuracy": 0.15784741938114166,
      "num_tokens": 4295696.0,
      "step": 2330
    },
    {
      "entropy": 5.74935622215271,
      "epoch": 0.1950384229869696,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998598248428423,
      "loss": 5.7175,
      "mean_token_accuracy": 0.15652929842472077,
      "num_tokens": 4305139.0,
      "step": 2335
    },
    {
      "entropy": 5.68619704246521,
      "epoch": 0.19545606414968258,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998587721954604,
      "loss": 5.5987,
      "mean_token_accuracy": 0.1622218132019043,
      "num_tokens": 4314029.0,
      "step": 2340
    },
    {
      "entropy": 5.705938053131104,
      "epoch": 0.1958737053123956,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998577156116551,
      "loss": 5.7354,
      "mean_token_accuracy": 0.15496601909399033,
      "num_tokens": 4323667.0,
      "step": 2345
    },
    {
      "entropy": 5.8132116317749025,
      "epoch": 0.1962913464751086,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998566550914448,
      "loss": 5.7757,
      "mean_token_accuracy": 0.1555030509829521,
      "num_tokens": 4333456.0,
      "step": 2350
    },
    {
      "entropy": 5.796841716766357,
      "epoch": 0.19670898763782158,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998555906348481,
      "loss": 5.7374,
      "mean_token_accuracy": 0.15186412781476974,
      "num_tokens": 4342527.0,
      "step": 2355
    },
    {
      "entropy": 5.733500623703003,
      "epoch": 0.19712662880053458,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998545222418836,
      "loss": 5.6578,
      "mean_token_accuracy": 0.15209388211369515,
      "num_tokens": 4351294.0,
      "step": 2360
    },
    {
      "entropy": 5.716574573516846,
      "epoch": 0.1975442699632476,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998534499125701,
      "loss": 5.664,
      "mean_token_accuracy": 0.15844270288944245,
      "num_tokens": 4361596.0,
      "step": 2365
    },
    {
      "entropy": 5.748371124267578,
      "epoch": 0.19796191112596057,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998523736469263,
      "loss": 5.6712,
      "mean_token_accuracy": 0.15820184350013733,
      "num_tokens": 4370917.0,
      "step": 2370
    },
    {
      "entropy": 5.66724534034729,
      "epoch": 0.19837955228867357,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499851293444971,
      "loss": 5.7026,
      "mean_token_accuracy": 0.15223393589258194,
      "num_tokens": 4380244.0,
      "step": 2375
    },
    {
      "entropy": 5.791155099868774,
      "epoch": 0.19879719345138658,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998502093067231,
      "loss": 5.6957,
      "mean_token_accuracy": 0.1528972566127777,
      "num_tokens": 4389852.0,
      "step": 2380
    },
    {
      "entropy": 5.754044246673584,
      "epoch": 0.19921483461409956,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998491212322017,
      "loss": 5.7307,
      "mean_token_accuracy": 0.15377760976552962,
      "num_tokens": 4399743.0,
      "step": 2385
    },
    {
      "entropy": 5.829539442062378,
      "epoch": 0.19963247577681256,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998480292214258,
      "loss": 5.6773,
      "mean_token_accuracy": 0.1586298957467079,
      "num_tokens": 4408661.0,
      "step": 2390
    },
    {
      "entropy": 5.641575002670288,
      "epoch": 0.20005011693952557,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998469332744144,
      "loss": 5.6336,
      "mean_token_accuracy": 0.16201509684324264,
      "num_tokens": 4417671.0,
      "step": 2395
    },
    {
      "entropy": 5.734684228897095,
      "epoch": 0.20046775810223855,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998458333911869,
      "loss": 5.7229,
      "mean_token_accuracy": 0.15162376314401627,
      "num_tokens": 4426850.0,
      "step": 2400
    },
    {
      "entropy": 5.829426002502442,
      "epoch": 0.20088539926495155,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998447295717624,
      "loss": 5.8166,
      "mean_token_accuracy": 0.15216446965932845,
      "num_tokens": 4435769.0,
      "step": 2405
    },
    {
      "entropy": 5.81864972114563,
      "epoch": 0.20130304042766456,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998436218161602,
      "loss": 5.7407,
      "mean_token_accuracy": 0.15777725130319595,
      "num_tokens": 4445521.0,
      "step": 2410
    },
    {
      "entropy": 5.732977390289307,
      "epoch": 0.20172068159037754,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998425101243999,
      "loss": 5.7357,
      "mean_token_accuracy": 0.1560164675116539,
      "num_tokens": 4454674.0,
      "step": 2415
    },
    {
      "entropy": 5.858603000640869,
      "epoch": 0.20213832275309054,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998413944965007,
      "loss": 5.7668,
      "mean_token_accuracy": 0.15934987515211105,
      "num_tokens": 4464035.0,
      "step": 2420
    },
    {
      "entropy": 5.726704406738281,
      "epoch": 0.20255596391580355,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998402749324823,
      "loss": 5.6002,
      "mean_token_accuracy": 0.1607057973742485,
      "num_tokens": 4472055.0,
      "step": 2425
    },
    {
      "entropy": 5.669188404083252,
      "epoch": 0.20297360507851653,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998391514323642,
      "loss": 5.7112,
      "mean_token_accuracy": 0.16063073128461838,
      "num_tokens": 4480669.0,
      "step": 2430
    },
    {
      "entropy": 5.813586473464966,
      "epoch": 0.20339124624122953,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998380239961661,
      "loss": 5.7006,
      "mean_token_accuracy": 0.16175678223371506,
      "num_tokens": 4489481.0,
      "step": 2435
    },
    {
      "entropy": 5.664542865753174,
      "epoch": 0.20380888740394254,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998368926239078,
      "loss": 5.5768,
      "mean_token_accuracy": 0.16573526561260224,
      "num_tokens": 4497508.0,
      "step": 2440
    },
    {
      "entropy": 5.650903415679932,
      "epoch": 0.20422652856665552,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499835757315609,
      "loss": 5.6594,
      "mean_token_accuracy": 0.16277491748332978,
      "num_tokens": 4506373.0,
      "step": 2445
    },
    {
      "entropy": 5.782987022399903,
      "epoch": 0.20464416972936852,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998346180712898,
      "loss": 5.7177,
      "mean_token_accuracy": 0.15605076253414155,
      "num_tokens": 4516283.0,
      "step": 2450
    },
    {
      "entropy": 5.701898193359375,
      "epoch": 0.20506181089208153,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998334748909698,
      "loss": 5.5897,
      "mean_token_accuracy": 0.1695755124092102,
      "num_tokens": 4524276.0,
      "step": 2455
    },
    {
      "entropy": 5.653223943710327,
      "epoch": 0.2054794520547945,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998323277746694,
      "loss": 5.667,
      "mean_token_accuracy": 0.16021030247211457,
      "num_tokens": 4532951.0,
      "step": 2460
    },
    {
      "entropy": 5.747987127304077,
      "epoch": 0.20589709321750752,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998311767224081,
      "loss": 5.6684,
      "mean_token_accuracy": 0.1644655093550682,
      "num_tokens": 4542229.0,
      "step": 2465
    },
    {
      "entropy": 5.686886167526245,
      "epoch": 0.20631473438022052,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998300217342067,
      "loss": 5.5713,
      "mean_token_accuracy": 0.16310817152261733,
      "num_tokens": 4551126.0,
      "step": 2470
    },
    {
      "entropy": 5.611570644378662,
      "epoch": 0.2067323755429335,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998288628100851,
      "loss": 5.679,
      "mean_token_accuracy": 0.16197994947433472,
      "num_tokens": 4560628.0,
      "step": 2475
    },
    {
      "entropy": 5.785298919677734,
      "epoch": 0.2071500167056465,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998276999500636,
      "loss": 5.7268,
      "mean_token_accuracy": 0.15670448988676072,
      "num_tokens": 4570184.0,
      "step": 2480
    },
    {
      "entropy": 5.70731635093689,
      "epoch": 0.2075676578683595,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998265331541627,
      "loss": 5.5896,
      "mean_token_accuracy": 0.16156910210847855,
      "num_tokens": 4580357.0,
      "step": 2485
    },
    {
      "entropy": 5.839704275131226,
      "epoch": 0.2079852990310725,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998253624224026,
      "loss": 5.7768,
      "mean_token_accuracy": 0.1487571582198143,
      "num_tokens": 4589965.0,
      "step": 2490
    },
    {
      "entropy": 5.740112066268921,
      "epoch": 0.2084029401937855,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499824187754804,
      "loss": 5.5971,
      "mean_token_accuracy": 0.16340472996234895,
      "num_tokens": 4598753.0,
      "step": 2495
    },
    {
      "entropy": 5.633929061889648,
      "epoch": 0.2088205813564985,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998230091513872,
      "loss": 5.6597,
      "mean_token_accuracy": 0.16298450231552125,
      "num_tokens": 4608698.0,
      "step": 2500
    },
    {
      "entropy": 5.745545530319214,
      "epoch": 0.20923822251921148,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004998218266121731,
      "loss": 5.7676,
      "mean_token_accuracy": 0.1568822406232357,
      "num_tokens": 4618137.0,
      "step": 2505
    },
    {
      "entropy": 5.694258832931519,
      "epoch": 0.2096558636819245,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998206401371823,
      "loss": 5.5901,
      "mean_token_accuracy": 0.16162560731172562,
      "num_tokens": 4627468.0,
      "step": 2510
    },
    {
      "entropy": 5.601826763153076,
      "epoch": 0.2100735048446375,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998194497264357,
      "loss": 5.4935,
      "mean_token_accuracy": 0.167683382332325,
      "num_tokens": 4636659.0,
      "step": 2515
    },
    {
      "entropy": 5.648157739639283,
      "epoch": 0.21049114600735047,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998182553799538,
      "loss": 5.6024,
      "mean_token_accuracy": 0.1581842228770256,
      "num_tokens": 4644969.0,
      "step": 2520
    },
    {
      "entropy": 5.7688682079315186,
      "epoch": 0.21090878717006348,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998170570977579,
      "loss": 5.7041,
      "mean_token_accuracy": 0.15401836931705476,
      "num_tokens": 4653930.0,
      "step": 2525
    },
    {
      "entropy": 5.706395292282105,
      "epoch": 0.21132642833277648,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998158548798687,
      "loss": 5.5867,
      "mean_token_accuracy": 0.1635697916150093,
      "num_tokens": 4663146.0,
      "step": 2530
    },
    {
      "entropy": 5.644166040420532,
      "epoch": 0.21174406949548946,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998146487263074,
      "loss": 5.7526,
      "mean_token_accuracy": 0.15531763583421707,
      "num_tokens": 4671874.0,
      "step": 2535
    },
    {
      "entropy": 5.625101566314697,
      "epoch": 0.21216171065820247,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998134386370951,
      "loss": 5.578,
      "mean_token_accuracy": 0.1623372033238411,
      "num_tokens": 4680811.0,
      "step": 2540
    },
    {
      "entropy": 5.786676645278931,
      "epoch": 0.21257935182091547,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998122246122529,
      "loss": 5.6417,
      "mean_token_accuracy": 0.1630154401063919,
      "num_tokens": 4689977.0,
      "step": 2545
    },
    {
      "entropy": 5.6473369121551515,
      "epoch": 0.21299699298362845,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998110066518021,
      "loss": 5.5277,
      "mean_token_accuracy": 0.1666818454861641,
      "num_tokens": 4698479.0,
      "step": 2550
    },
    {
      "entropy": 5.558276653289795,
      "epoch": 0.21341463414634146,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499809784755764,
      "loss": 5.5995,
      "mean_token_accuracy": 0.1562129810452461,
      "num_tokens": 4707575.0,
      "step": 2555
    },
    {
      "entropy": 5.7299939632415775,
      "epoch": 0.21383227530905446,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049980855892416,
      "loss": 5.6385,
      "mean_token_accuracy": 0.1641583114862442,
      "num_tokens": 4717015.0,
      "step": 2560
    },
    {
      "entropy": 5.690065193176269,
      "epoch": 0.21424991647176747,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998073291570116,
      "loss": 5.6671,
      "mean_token_accuracy": 0.15410696864128112,
      "num_tokens": 4726466.0,
      "step": 2565
    },
    {
      "entropy": 5.794637441635132,
      "epoch": 0.21466755763448045,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998060954543404,
      "loss": 5.7155,
      "mean_token_accuracy": 0.16310942322015762,
      "num_tokens": 4736497.0,
      "step": 2570
    },
    {
      "entropy": 5.648622655868531,
      "epoch": 0.21508519879719346,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998048578161678,
      "loss": 5.6435,
      "mean_token_accuracy": 0.1570078417658806,
      "num_tokens": 4745693.0,
      "step": 2575
    },
    {
      "entropy": 5.650327682495117,
      "epoch": 0.21550283995990646,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998036162425155,
      "loss": 5.562,
      "mean_token_accuracy": 0.17139142453670503,
      "num_tokens": 4754274.0,
      "step": 2580
    },
    {
      "entropy": 5.747675991058349,
      "epoch": 0.21592048112261944,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998023707334055,
      "loss": 5.7886,
      "mean_token_accuracy": 0.15042773261666298,
      "num_tokens": 4765607.0,
      "step": 2585
    },
    {
      "entropy": 5.7171971797943115,
      "epoch": 0.21633812228533245,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998011212888592,
      "loss": 5.5613,
      "mean_token_accuracy": 0.16136786639690398,
      "num_tokens": 4775291.0,
      "step": 2590
    },
    {
      "entropy": 5.59135890007019,
      "epoch": 0.21675576344804545,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997998679088988,
      "loss": 5.5037,
      "mean_token_accuracy": 0.16423833072185517,
      "num_tokens": 4783859.0,
      "step": 2595
    },
    {
      "entropy": 5.5703614234924315,
      "epoch": 0.21717340461075843,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997986105935461,
      "loss": 5.5722,
      "mean_token_accuracy": 0.16074053645133973,
      "num_tokens": 4792891.0,
      "step": 2600
    },
    {
      "entropy": 5.6992326259613035,
      "epoch": 0.21759104577347144,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997973493428231,
      "loss": 5.6581,
      "mean_token_accuracy": 0.15850717574357986,
      "num_tokens": 4801551.0,
      "step": 2605
    },
    {
      "entropy": 5.765617036819458,
      "epoch": 0.21800868693618444,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997960841567519,
      "loss": 5.617,
      "mean_token_accuracy": 0.1606222003698349,
      "num_tokens": 4810430.0,
      "step": 2610
    },
    {
      "entropy": 5.609820175170898,
      "epoch": 0.21842632809889742,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997948150353548,
      "loss": 5.6148,
      "mean_token_accuracy": 0.1620870128273964,
      "num_tokens": 4819156.0,
      "step": 2615
    },
    {
      "entropy": 5.7782470703125,
      "epoch": 0.21884396926161043,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997935419786537,
      "loss": 5.6495,
      "mean_token_accuracy": 0.15833018720149994,
      "num_tokens": 4827724.0,
      "step": 2620
    },
    {
      "entropy": 5.689379119873047,
      "epoch": 0.21926161042432343,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997922649866712,
      "loss": 5.5754,
      "mean_token_accuracy": 0.16985357105731963,
      "num_tokens": 4836954.0,
      "step": 2625
    },
    {
      "entropy": 5.694972991943359,
      "epoch": 0.2196792515870364,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997909840594294,
      "loss": 5.6895,
      "mean_token_accuracy": 0.15483302474021912,
      "num_tokens": 4847504.0,
      "step": 2630
    },
    {
      "entropy": 5.720165061950683,
      "epoch": 0.22009689274974942,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997896991969509,
      "loss": 5.6209,
      "mean_token_accuracy": 0.16614843010902405,
      "num_tokens": 4856105.0,
      "step": 2635
    },
    {
      "entropy": 5.61157546043396,
      "epoch": 0.22051453391246242,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997884103992581,
      "loss": 5.6128,
      "mean_token_accuracy": 0.16402403861284257,
      "num_tokens": 4866118.0,
      "step": 2640
    },
    {
      "entropy": 5.756872844696045,
      "epoch": 0.2209321750751754,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997871176663736,
      "loss": 5.6501,
      "mean_token_accuracy": 0.16168341040611267,
      "num_tokens": 4875242.0,
      "step": 2645
    },
    {
      "entropy": 5.771356153488159,
      "epoch": 0.2213498162378884,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997858209983201,
      "loss": 5.7122,
      "mean_token_accuracy": 0.16138535365462303,
      "num_tokens": 4884726.0,
      "step": 2650
    },
    {
      "entropy": 5.623738479614258,
      "epoch": 0.22176745740060141,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997845203951202,
      "loss": 5.6596,
      "mean_token_accuracy": 0.15921541154384614,
      "num_tokens": 4893178.0,
      "step": 2655
    },
    {
      "entropy": 5.808769464492798,
      "epoch": 0.2221850985633144,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997832158567967,
      "loss": 5.7487,
      "mean_token_accuracy": 0.16165247559547424,
      "num_tokens": 4903126.0,
      "step": 2660
    },
    {
      "entropy": 5.619534492492676,
      "epoch": 0.2226027397260274,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997819073833724,
      "loss": 5.4728,
      "mean_token_accuracy": 0.17080911844968796,
      "num_tokens": 4912069.0,
      "step": 2665
    },
    {
      "entropy": 5.607888507843017,
      "epoch": 0.2230203808887404,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997805949748702,
      "loss": 5.5288,
      "mean_token_accuracy": 0.16749428808689118,
      "num_tokens": 4921014.0,
      "step": 2670
    },
    {
      "entropy": 5.537565088272094,
      "epoch": 0.22343802205145338,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997792786313132,
      "loss": 5.5862,
      "mean_token_accuracy": 0.16741064935922623,
      "num_tokens": 4930382.0,
      "step": 2675
    },
    {
      "entropy": 5.759979391098023,
      "epoch": 0.2238556632141664,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997779583527243,
      "loss": 5.652,
      "mean_token_accuracy": 0.15423294603824617,
      "num_tokens": 4939672.0,
      "step": 2680
    },
    {
      "entropy": 5.817228889465332,
      "epoch": 0.2242733043768794,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997766341391268,
      "loss": 5.6812,
      "mean_token_accuracy": 0.15111590325832366,
      "num_tokens": 4949186.0,
      "step": 2685
    },
    {
      "entropy": 5.6610682010650635,
      "epoch": 0.22469094553959237,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997753059905437,
      "loss": 5.6584,
      "mean_token_accuracy": 0.15599125772714614,
      "num_tokens": 4958155.0,
      "step": 2690
    },
    {
      "entropy": 5.574031972885132,
      "epoch": 0.22510858670230538,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997739739069981,
      "loss": 5.5189,
      "mean_token_accuracy": 0.16694373190402984,
      "num_tokens": 4966524.0,
      "step": 2695
    },
    {
      "entropy": 5.665495204925537,
      "epoch": 0.22552622786501839,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997726378885138,
      "loss": 5.5993,
      "mean_token_accuracy": 0.16144806295633315,
      "num_tokens": 4976066.0,
      "step": 2700
    },
    {
      "entropy": 5.661115884780884,
      "epoch": 0.22594386902773136,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997712979351137,
      "loss": 5.5056,
      "mean_token_accuracy": 0.17269045412540435,
      "num_tokens": 4984957.0,
      "step": 2705
    },
    {
      "entropy": 5.5876758098602295,
      "epoch": 0.22636151019044437,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997699540468216,
      "loss": 5.4869,
      "mean_token_accuracy": 0.1622341826558113,
      "num_tokens": 4994430.0,
      "step": 2710
    },
    {
      "entropy": 5.5405677318572994,
      "epoch": 0.22677915135315738,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997686062236609,
      "loss": 5.623,
      "mean_token_accuracy": 0.1629298061132431,
      "num_tokens": 5003893.0,
      "step": 2715
    },
    {
      "entropy": 5.660926103591919,
      "epoch": 0.22719679251587035,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997672544656553,
      "loss": 5.609,
      "mean_token_accuracy": 0.1635463997721672,
      "num_tokens": 5014083.0,
      "step": 2720
    },
    {
      "entropy": 5.621028232574463,
      "epoch": 0.22761443367858336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997658987728283,
      "loss": 5.4988,
      "mean_token_accuracy": 0.16685888916254044,
      "num_tokens": 5023025.0,
      "step": 2725
    },
    {
      "entropy": 5.596793222427368,
      "epoch": 0.22803207484129637,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997645391452035,
      "loss": 5.631,
      "mean_token_accuracy": 0.15907337963581086,
      "num_tokens": 5032620.0,
      "step": 2730
    },
    {
      "entropy": 5.791715097427368,
      "epoch": 0.22844971600400935,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499763175582805,
      "loss": 5.6414,
      "mean_token_accuracy": 0.15750645697116852,
      "num_tokens": 5041773.0,
      "step": 2735
    },
    {
      "entropy": 5.659903049468994,
      "epoch": 0.22886735716672235,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997618080856566,
      "loss": 5.6808,
      "mean_token_accuracy": 0.1559741899371147,
      "num_tokens": 5051450.0,
      "step": 2740
    },
    {
      "entropy": 5.641678237915039,
      "epoch": 0.22928499832943536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997604366537823,
      "loss": 5.6401,
      "mean_token_accuracy": 0.1551765412092209,
      "num_tokens": 5061172.0,
      "step": 2745
    },
    {
      "entropy": 5.652740812301635,
      "epoch": 0.22970263949214834,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997590612872058,
      "loss": 5.5455,
      "mean_token_accuracy": 0.1640415072441101,
      "num_tokens": 5071161.0,
      "step": 2750
    },
    {
      "entropy": 5.610337162017823,
      "epoch": 0.23012028065486134,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997576819859515,
      "loss": 5.631,
      "mean_token_accuracy": 0.1642090708017349,
      "num_tokens": 5080884.0,
      "step": 2755
    },
    {
      "entropy": 5.643430471420288,
      "epoch": 0.23053792181757435,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997562987500434,
      "loss": 5.5696,
      "mean_token_accuracy": 0.16299497783184053,
      "num_tokens": 5090164.0,
      "step": 2760
    },
    {
      "entropy": 5.54939866065979,
      "epoch": 0.23095556298028733,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997549115795058,
      "loss": 5.5187,
      "mean_token_accuracy": 0.1628969594836235,
      "num_tokens": 5098700.0,
      "step": 2765
    },
    {
      "entropy": 5.79123969078064,
      "epoch": 0.23137320414300033,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499753520474363,
      "loss": 5.769,
      "mean_token_accuracy": 0.14941132217645645,
      "num_tokens": 5107713.0,
      "step": 2770
    },
    {
      "entropy": 5.765942716598511,
      "epoch": 0.23179084530571334,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997521254346391,
      "loss": 5.7571,
      "mean_token_accuracy": 0.15038875490427017,
      "num_tokens": 5117532.0,
      "step": 2775
    },
    {
      "entropy": 5.685317707061768,
      "epoch": 0.23220848646842632,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997507264603589,
      "loss": 5.5187,
      "mean_token_accuracy": 0.16634505540132521,
      "num_tokens": 5126891.0,
      "step": 2780
    },
    {
      "entropy": 5.571315050125122,
      "epoch": 0.23262612763113932,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997493235515466,
      "loss": 5.4884,
      "mean_token_accuracy": 0.16462970823049544,
      "num_tokens": 5135608.0,
      "step": 2785
    },
    {
      "entropy": 5.643984651565551,
      "epoch": 0.23304376879385233,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997479167082269,
      "loss": 5.5878,
      "mean_token_accuracy": 0.15804445147514343,
      "num_tokens": 5145055.0,
      "step": 2790
    },
    {
      "entropy": 5.702338075637817,
      "epoch": 0.2334614099565653,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997465059304243,
      "loss": 5.6301,
      "mean_token_accuracy": 0.16131939738988876,
      "num_tokens": 5154742.0,
      "step": 2795
    },
    {
      "entropy": 5.640383243560791,
      "epoch": 0.23387905111927831,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997450912181637,
      "loss": 5.6043,
      "mean_token_accuracy": 0.15742480158805847,
      "num_tokens": 5163876.0,
      "step": 2800
    },
    {
      "entropy": 5.638509082794189,
      "epoch": 0.23429669228199132,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997436725714696,
      "loss": 5.6155,
      "mean_token_accuracy": 0.1609307512640953,
      "num_tokens": 5173614.0,
      "step": 2805
    },
    {
      "entropy": 5.70156888961792,
      "epoch": 0.2347143334447043,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997422499903671,
      "loss": 5.6353,
      "mean_token_accuracy": 0.1623079940676689,
      "num_tokens": 5182061.0,
      "step": 2810
    },
    {
      "entropy": 5.7618965148925785,
      "epoch": 0.2351319746074173,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499740823474881,
      "loss": 5.8081,
      "mean_token_accuracy": 0.14517047330737115,
      "num_tokens": 5191661.0,
      "step": 2815
    },
    {
      "entropy": 5.767496252059937,
      "epoch": 0.2355496157701303,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997393930250363,
      "loss": 5.7083,
      "mean_token_accuracy": 0.15727835297584533,
      "num_tokens": 5200624.0,
      "step": 2820
    },
    {
      "entropy": 5.6672691822052,
      "epoch": 0.2359672569328433,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499737958640858,
      "loss": 5.6324,
      "mean_token_accuracy": 0.15578907281160354,
      "num_tokens": 5209741.0,
      "step": 2825
    },
    {
      "entropy": 5.71718430519104,
      "epoch": 0.2363848980955563,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997365203223711,
      "loss": 5.5589,
      "mean_token_accuracy": 0.16819592118263244,
      "num_tokens": 5218877.0,
      "step": 2830
    },
    {
      "entropy": 5.673211860656738,
      "epoch": 0.2368025392582693,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499735078069601,
      "loss": 5.6996,
      "mean_token_accuracy": 0.15367967188358306,
      "num_tokens": 5229153.0,
      "step": 2835
    },
    {
      "entropy": 5.635143899917603,
      "epoch": 0.23722018042098228,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997336318825728,
      "loss": 5.5574,
      "mean_token_accuracy": 0.16222159564495087,
      "num_tokens": 5238561.0,
      "step": 2840
    },
    {
      "entropy": 5.709632921218872,
      "epoch": 0.23763782158369529,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997321817613119,
      "loss": 5.6481,
      "mean_token_accuracy": 0.16289447247982025,
      "num_tokens": 5248135.0,
      "step": 2845
    },
    {
      "entropy": 5.6386292457580565,
      "epoch": 0.2380554627464083,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997307277058436,
      "loss": 5.5654,
      "mean_token_accuracy": 0.16689260303974152,
      "num_tokens": 5257286.0,
      "step": 2850
    },
    {
      "entropy": 5.671214628219604,
      "epoch": 0.23847310390912127,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997292697161935,
      "loss": 5.7306,
      "mean_token_accuracy": 0.15510661229491235,
      "num_tokens": 5266595.0,
      "step": 2855
    },
    {
      "entropy": 5.700104284286499,
      "epoch": 0.23889074507183428,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499727807792387,
      "loss": 5.5257,
      "mean_token_accuracy": 0.1626771479845047,
      "num_tokens": 5275651.0,
      "step": 2860
    },
    {
      "entropy": 5.505528211593628,
      "epoch": 0.23930838623454728,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997263419344495,
      "loss": 5.473,
      "mean_token_accuracy": 0.16999822109937668,
      "num_tokens": 5284596.0,
      "step": 2865
    },
    {
      "entropy": 5.670961904525757,
      "epoch": 0.23972602739726026,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997248721424071,
      "loss": 5.6304,
      "mean_token_accuracy": 0.16004635393619537,
      "num_tokens": 5292410.0,
      "step": 2870
    },
    {
      "entropy": 5.590500545501709,
      "epoch": 0.24014366855997327,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997233984162853,
      "loss": 5.5005,
      "mean_token_accuracy": 0.16961989551782608,
      "num_tokens": 5301124.0,
      "step": 2875
    },
    {
      "entropy": 5.697359323501587,
      "epoch": 0.24056130972268627,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997219207561099,
      "loss": 5.6064,
      "mean_token_accuracy": 0.16648687422275543,
      "num_tokens": 5310571.0,
      "step": 2880
    },
    {
      "entropy": 5.662830972671509,
      "epoch": 0.24097895088539925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997204391619068,
      "loss": 5.5717,
      "mean_token_accuracy": 0.16593839973211288,
      "num_tokens": 5319746.0,
      "step": 2885
    },
    {
      "entropy": 5.622473859786988,
      "epoch": 0.24139659204811226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997189536337018,
      "loss": 5.6595,
      "mean_token_accuracy": 0.16147310808300971,
      "num_tokens": 5329734.0,
      "step": 2890
    },
    {
      "entropy": 5.6491858005523685,
      "epoch": 0.24181423321082526,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997174641715211,
      "loss": 5.4694,
      "mean_token_accuracy": 0.17032976001501082,
      "num_tokens": 5338405.0,
      "step": 2895
    },
    {
      "entropy": 5.620096254348755,
      "epoch": 0.24223187437353827,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997159707753906,
      "loss": 5.6705,
      "mean_token_accuracy": 0.1593254551291466,
      "num_tokens": 5348179.0,
      "step": 2900
    },
    {
      "entropy": 5.836646986007691,
      "epoch": 0.24264951553625125,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997144734453367,
      "loss": 5.7065,
      "mean_token_accuracy": 0.15110059678554535,
      "num_tokens": 5358035.0,
      "step": 2905
    },
    {
      "entropy": 5.599517440795898,
      "epoch": 0.24306715669896425,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997129721813853,
      "loss": 5.5921,
      "mean_token_accuracy": 0.17012663632631303,
      "num_tokens": 5366713.0,
      "step": 2910
    },
    {
      "entropy": 5.542704105377197,
      "epoch": 0.24348479786167726,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997114669835629,
      "loss": 5.4764,
      "mean_token_accuracy": 0.16873355507850646,
      "num_tokens": 5375245.0,
      "step": 2915
    },
    {
      "entropy": 5.587150430679321,
      "epoch": 0.24390243902439024,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997099578518959,
      "loss": 5.5601,
      "mean_token_accuracy": 0.16696805357933045,
      "num_tokens": 5384097.0,
      "step": 2920
    },
    {
      "entropy": 5.721682119369507,
      "epoch": 0.24432008018710324,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997084447864105,
      "loss": 5.6815,
      "mean_token_accuracy": 0.15409955456852914,
      "num_tokens": 5393611.0,
      "step": 2925
    },
    {
      "entropy": 5.687287282943726,
      "epoch": 0.24473772134981625,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997069277871333,
      "loss": 5.6073,
      "mean_token_accuracy": 0.16065794229507446,
      "num_tokens": 5402332.0,
      "step": 2930
    },
    {
      "entropy": 5.523683404922485,
      "epoch": 0.24515536251252923,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997054068540908,
      "loss": 5.5574,
      "mean_token_accuracy": 0.1698683112859726,
      "num_tokens": 5411415.0,
      "step": 2935
    },
    {
      "entropy": 5.7273612976074215,
      "epoch": 0.24557300367524224,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997038819873098,
      "loss": 5.4526,
      "mean_token_accuracy": 0.17287943363189698,
      "num_tokens": 5419742.0,
      "step": 2940
    },
    {
      "entropy": 5.498819160461426,
      "epoch": 0.24599064483795524,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997023531868167,
      "loss": 5.498,
      "mean_token_accuracy": 0.17199851274490358,
      "num_tokens": 5428689.0,
      "step": 2945
    },
    {
      "entropy": 5.7247068881988525,
      "epoch": 0.24640828600066822,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997008204526386,
      "loss": 5.6771,
      "mean_token_accuracy": 0.1559805117547512,
      "num_tokens": 5438096.0,
      "step": 2950
    },
    {
      "entropy": 5.657676792144775,
      "epoch": 0.24682592716338123,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996992837848021,
      "loss": 5.5497,
      "mean_token_accuracy": 0.16129023730754852,
      "num_tokens": 5446726.0,
      "step": 2955
    },
    {
      "entropy": 5.5872002124786375,
      "epoch": 0.24724356832609423,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004996977431833342,
      "loss": 5.6677,
      "mean_token_accuracy": 0.15905494391918182,
      "num_tokens": 5456275.0,
      "step": 2960
    },
    {
      "entropy": 5.700059175491333,
      "epoch": 0.2476612094888072,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996961986482618,
      "loss": 5.6299,
      "mean_token_accuracy": 0.16636704355478288,
      "num_tokens": 5466135.0,
      "step": 2965
    },
    {
      "entropy": 5.551216793060303,
      "epoch": 0.24807885065152022,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996946501796119,
      "loss": 5.5014,
      "mean_token_accuracy": 0.17257890850305557,
      "num_tokens": 5474882.0,
      "step": 2970
    },
    {
      "entropy": 5.448128414154053,
      "epoch": 0.24849649181423322,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996930977774118,
      "loss": 5.4279,
      "mean_token_accuracy": 0.17420113384723662,
      "num_tokens": 5484856.0,
      "step": 2975
    },
    {
      "entropy": 5.69349365234375,
      "epoch": 0.2489141329769462,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996915414416885,
      "loss": 5.6235,
      "mean_token_accuracy": 0.15520315021276473,
      "num_tokens": 5493840.0,
      "step": 2980
    },
    {
      "entropy": 5.653994464874268,
      "epoch": 0.2493317741396592,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996899811724693,
      "loss": 5.5977,
      "mean_token_accuracy": 0.16332905516028404,
      "num_tokens": 5502655.0,
      "step": 2985
    },
    {
      "entropy": 5.607406282424927,
      "epoch": 0.2497494153023722,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996884169697816,
      "loss": 5.5109,
      "mean_token_accuracy": 0.17501745223999024,
      "num_tokens": 5511640.0,
      "step": 2990
    },
    {
      "entropy": 5.5393799304962155,
      "epoch": 0.2501670564650852,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996868488336526,
      "loss": 5.4996,
      "mean_token_accuracy": 0.1669514074921608,
      "num_tokens": 5520704.0,
      "step": 2995
    },
    {
      "entropy": 5.605220365524292,
      "epoch": 0.25058469762779817,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996852767641099,
      "loss": 5.5816,
      "mean_token_accuracy": 0.16140030473470687,
      "num_tokens": 5531090.0,
      "step": 3000
    },
    {
      "epoch": 0.25058469762779817,
      "eval_entropy": 5.445592117976193,
      "eval_loss": 5.563920497894287,
      "eval_mean_token_accuracy": 0.17285948360267395,
      "eval_num_tokens": 5531090.0,
      "eval_runtime": 27.6695,
      "eval_samples_per_second": 1364.789,
      "eval_steps_per_second": 170.621,
      "step": 3000
    },
    {
      "entropy": 5.619354915618897,
      "epoch": 0.2510023387905112,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996837007611811,
      "loss": 5.6279,
      "mean_token_accuracy": 0.16189804971218108,
      "num_tokens": 5539905.0,
      "step": 3005
    },
    {
      "entropy": 5.790455722808838,
      "epoch": 0.2514199799532242,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004996821208248935,
      "loss": 5.66,
      "mean_token_accuracy": 0.15732589215040207,
      "num_tokens": 5549052.0,
      "step": 3010
    },
    {
      "entropy": 5.793611192703247,
      "epoch": 0.25183762111593716,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996805369552749,
      "loss": 5.6652,
      "mean_token_accuracy": 0.1621344655752182,
      "num_tokens": 5558156.0,
      "step": 3015
    },
    {
      "entropy": 5.588698101043701,
      "epoch": 0.2522552622786502,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996789491523532,
      "loss": 5.574,
      "mean_token_accuracy": 0.1657869517803192,
      "num_tokens": 5567367.0,
      "step": 3020
    },
    {
      "entropy": 5.6186281681060795,
      "epoch": 0.2526729034413632,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996773574161559,
      "loss": 5.6607,
      "mean_token_accuracy": 0.16135193556547164,
      "num_tokens": 5576883.0,
      "step": 3025
    },
    {
      "entropy": 5.7037670612335205,
      "epoch": 0.25309054460407615,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996757617467112,
      "loss": 5.5479,
      "mean_token_accuracy": 0.16849476248025894,
      "num_tokens": 5586050.0,
      "step": 3030
    },
    {
      "entropy": 5.73937029838562,
      "epoch": 0.2535081857667892,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996741621440466,
      "loss": 5.719,
      "mean_token_accuracy": 0.15612117648124696,
      "num_tokens": 5595933.0,
      "step": 3035
    },
    {
      "entropy": 5.689214468002319,
      "epoch": 0.25392582692950216,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996725586081906,
      "loss": 5.5941,
      "mean_token_accuracy": 0.16068582087755204,
      "num_tokens": 5604832.0,
      "step": 3040
    },
    {
      "entropy": 5.606196403503418,
      "epoch": 0.25434346809221514,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996709511391709,
      "loss": 5.5841,
      "mean_token_accuracy": 0.16500435769557953,
      "num_tokens": 5614077.0,
      "step": 3045
    },
    {
      "entropy": 5.654494285583496,
      "epoch": 0.2547611092549282,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996693397370158,
      "loss": 5.5317,
      "mean_token_accuracy": 0.16322938948869706,
      "num_tokens": 5622971.0,
      "step": 3050
    },
    {
      "entropy": 5.576167821884155,
      "epoch": 0.25517875041764115,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996677244017534,
      "loss": 5.4771,
      "mean_token_accuracy": 0.16635890752077104,
      "num_tokens": 5631356.0,
      "step": 3055
    },
    {
      "entropy": 5.55122127532959,
      "epoch": 0.25559639158035413,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996661051334121,
      "loss": 5.5086,
      "mean_token_accuracy": 0.16659390032291413,
      "num_tokens": 5641095.0,
      "step": 3060
    },
    {
      "entropy": 5.564494895935058,
      "epoch": 0.25601403274306717,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996644819320201,
      "loss": 5.4975,
      "mean_token_accuracy": 0.17219194620847703,
      "num_tokens": 5651311.0,
      "step": 3065
    },
    {
      "entropy": 5.630851888656617,
      "epoch": 0.25643167390578014,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996628547976061,
      "loss": 5.5506,
      "mean_token_accuracy": 0.16819642186164857,
      "num_tokens": 5659957.0,
      "step": 3070
    },
    {
      "entropy": 5.5410982131958,
      "epoch": 0.2568493150684932,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996612237301983,
      "loss": 5.5318,
      "mean_token_accuracy": 0.16320393234491348,
      "num_tokens": 5668876.0,
      "step": 3075
    },
    {
      "entropy": 5.545115947723389,
      "epoch": 0.25726695623120616,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996595887298254,
      "loss": 5.4886,
      "mean_token_accuracy": 0.16713176518678666,
      "num_tokens": 5677809.0,
      "step": 3080
    },
    {
      "entropy": 5.5903150081634525,
      "epoch": 0.25768459739391913,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499657949796516,
      "loss": 5.4582,
      "mean_token_accuracy": 0.17611153572797775,
      "num_tokens": 5687055.0,
      "step": 3085
    },
    {
      "entropy": 5.475444793701172,
      "epoch": 0.25810223855663217,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996563069302986,
      "loss": 5.5073,
      "mean_token_accuracy": 0.16294189989566804,
      "num_tokens": 5696394.0,
      "step": 3090
    },
    {
      "entropy": 5.687202072143554,
      "epoch": 0.25851987971934515,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996546601312023,
      "loss": 5.5786,
      "mean_token_accuracy": 0.16350055485963821,
      "num_tokens": 5704372.0,
      "step": 3095
    },
    {
      "entropy": 5.4938570022583,
      "epoch": 0.2589375208820581,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996530093992558,
      "loss": 5.4259,
      "mean_token_accuracy": 0.17429800778627397,
      "num_tokens": 5713535.0,
      "step": 3100
    },
    {
      "entropy": 5.456874084472656,
      "epoch": 0.25935516204477116,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996513547344877,
      "loss": 5.4753,
      "mean_token_accuracy": 0.17394353300333024,
      "num_tokens": 5722559.0,
      "step": 3105
    },
    {
      "entropy": 5.599130296707154,
      "epoch": 0.25977280320748414,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996496961369274,
      "loss": 5.547,
      "mean_token_accuracy": 0.16444521844387056,
      "num_tokens": 5731765.0,
      "step": 3110
    },
    {
      "entropy": 5.572427988052368,
      "epoch": 0.2601904443701971,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996480336066037,
      "loss": 5.4365,
      "mean_token_accuracy": 0.18244351595640182,
      "num_tokens": 5739989.0,
      "step": 3115
    },
    {
      "entropy": 5.59587516784668,
      "epoch": 0.26060808553291015,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996463671435458,
      "loss": 5.4658,
      "mean_token_accuracy": 0.17085543423891067,
      "num_tokens": 5748461.0,
      "step": 3120
    },
    {
      "entropy": 5.626722288131714,
      "epoch": 0.26102572669562313,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996446967477829,
      "loss": 5.5934,
      "mean_token_accuracy": 0.16824812889099122,
      "num_tokens": 5757495.0,
      "step": 3125
    },
    {
      "entropy": 5.670326280593872,
      "epoch": 0.2614433678583361,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996430224193441,
      "loss": 5.6468,
      "mean_token_accuracy": 0.16008122116327286,
      "num_tokens": 5767951.0,
      "step": 3130
    },
    {
      "entropy": 5.680183029174804,
      "epoch": 0.26186100902104914,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996413441582589,
      "loss": 5.5937,
      "mean_token_accuracy": 0.1659947469830513,
      "num_tokens": 5777021.0,
      "step": 3135
    },
    {
      "entropy": 5.572238922119141,
      "epoch": 0.2622786501837621,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996396619645564,
      "loss": 5.5446,
      "mean_token_accuracy": 0.16491442322731018,
      "num_tokens": 5785794.0,
      "step": 3140
    },
    {
      "entropy": 5.611316061019897,
      "epoch": 0.2626962913464751,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996379758382662,
      "loss": 5.4394,
      "mean_token_accuracy": 0.17236423194408418,
      "num_tokens": 5794480.0,
      "step": 3145
    },
    {
      "entropy": 5.437555170059204,
      "epoch": 0.26311393250918813,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499636285779418,
      "loss": 5.3734,
      "mean_token_accuracy": 0.18060459941625595,
      "num_tokens": 5802341.0,
      "step": 3150
    },
    {
      "entropy": 5.53844838142395,
      "epoch": 0.2635315736719011,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996345917880412,
      "loss": 5.5307,
      "mean_token_accuracy": 0.16787216514348985,
      "num_tokens": 5811365.0,
      "step": 3155
    },
    {
      "entropy": 5.558381080627441,
      "epoch": 0.2639492148346141,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996328938641654,
      "loss": 5.4675,
      "mean_token_accuracy": 0.17300861477851867,
      "num_tokens": 5819957.0,
      "step": 3160
    },
    {
      "entropy": 5.53677339553833,
      "epoch": 0.2643668559973271,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996311920078204,
      "loss": 5.4983,
      "mean_token_accuracy": 0.1693764254450798,
      "num_tokens": 5828378.0,
      "step": 3165
    },
    {
      "entropy": 5.64333815574646,
      "epoch": 0.2647844971600401,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996294862190361,
      "loss": 5.5793,
      "mean_token_accuracy": 0.16261188238859176,
      "num_tokens": 5837310.0,
      "step": 3170
    },
    {
      "entropy": 5.55024790763855,
      "epoch": 0.2652021383227531,
      "grad_norm": 1.0,
      "learning_rate": 0.000499627776497842,
      "loss": 5.518,
      "mean_token_accuracy": 0.17094084322452546,
      "num_tokens": 5845945.0,
      "step": 3175
    },
    {
      "entropy": 5.5680609226226805,
      "epoch": 0.2656197794854661,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996260628442686,
      "loss": 5.5254,
      "mean_token_accuracy": 0.16151418387889863,
      "num_tokens": 5855954.0,
      "step": 3180
    },
    {
      "entropy": 5.720606994628906,
      "epoch": 0.2660374206481791,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004996243452583454,
      "loss": 5.6397,
      "mean_token_accuracy": 0.15828690603375434,
      "num_tokens": 5865327.0,
      "step": 3185
    },
    {
      "entropy": 5.719430017471313,
      "epoch": 0.26645506181089207,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996226237401028,
      "loss": 5.6038,
      "mean_token_accuracy": 0.16542584747076033,
      "num_tokens": 5873959.0,
      "step": 3190
    },
    {
      "entropy": 5.533983564376831,
      "epoch": 0.2668727029736051,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996208982895706,
      "loss": 5.4574,
      "mean_token_accuracy": 0.17117766588926314,
      "num_tokens": 5882566.0,
      "step": 3195
    },
    {
      "entropy": 5.5592443466186525,
      "epoch": 0.2672903441363181,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996191689067794,
      "loss": 5.5055,
      "mean_token_accuracy": 0.1669450059533119,
      "num_tokens": 5891407.0,
      "step": 3200
    },
    {
      "entropy": 5.60461163520813,
      "epoch": 0.26770798529903106,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996174355917591,
      "loss": 5.4663,
      "mean_token_accuracy": 0.17221603840589522,
      "num_tokens": 5900123.0,
      "step": 3205
    },
    {
      "entropy": 5.5325319290161135,
      "epoch": 0.2681256264617441,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996156983445404,
      "loss": 5.5292,
      "mean_token_accuracy": 0.1720262810587883,
      "num_tokens": 5909043.0,
      "step": 3210
    },
    {
      "entropy": 5.637628793716431,
      "epoch": 0.26854326762445707,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996139571651534,
      "loss": 5.5511,
      "mean_token_accuracy": 0.16693660020828247,
      "num_tokens": 5918728.0,
      "step": 3215
    },
    {
      "entropy": 5.648838138580322,
      "epoch": 0.26896090878717005,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996122120536288,
      "loss": 5.5609,
      "mean_token_accuracy": 0.16625168174505234,
      "num_tokens": 5927281.0,
      "step": 3220
    },
    {
      "entropy": 5.529385042190552,
      "epoch": 0.2693785499498831,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996104630099969,
      "loss": 5.4818,
      "mean_token_accuracy": 0.17687633484601975,
      "num_tokens": 5935696.0,
      "step": 3225
    },
    {
      "entropy": 5.519879579544067,
      "epoch": 0.26979619111259606,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996087100342885,
      "loss": 5.5259,
      "mean_token_accuracy": 0.16781273633241653,
      "num_tokens": 5944858.0,
      "step": 3230
    },
    {
      "entropy": 5.741100788116455,
      "epoch": 0.27021383227530904,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996069531265345,
      "loss": 5.6246,
      "mean_token_accuracy": 0.1606270983815193,
      "num_tokens": 5954243.0,
      "step": 3235
    },
    {
      "entropy": 5.536730432510376,
      "epoch": 0.2706314734380221,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996051922867653,
      "loss": 5.4993,
      "mean_token_accuracy": 0.16611584275960922,
      "num_tokens": 5963318.0,
      "step": 3240
    },
    {
      "entropy": 5.564297723770141,
      "epoch": 0.27104911460073505,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996034275150118,
      "loss": 5.4059,
      "mean_token_accuracy": 0.17894322723150252,
      "num_tokens": 5971487.0,
      "step": 3245
    },
    {
      "entropy": 5.509364891052246,
      "epoch": 0.27146675576344803,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499601658811305,
      "loss": 5.4881,
      "mean_token_accuracy": 0.17391130924224854,
      "num_tokens": 5979981.0,
      "step": 3250
    },
    {
      "entropy": 5.6284205436706545,
      "epoch": 0.27188439692616106,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995998861756758,
      "loss": 5.5299,
      "mean_token_accuracy": 0.1685034915804863,
      "num_tokens": 5989427.0,
      "step": 3255
    },
    {
      "entropy": 5.515089416503907,
      "epoch": 0.27230203808887404,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995981096081552,
      "loss": 5.4914,
      "mean_token_accuracy": 0.16262538433074952,
      "num_tokens": 5999434.0,
      "step": 3260
    },
    {
      "entropy": 5.607756948471069,
      "epoch": 0.272719679251587,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995963291087744,
      "loss": 5.5435,
      "mean_token_accuracy": 0.16143085807561874,
      "num_tokens": 6008963.0,
      "step": 3265
    },
    {
      "entropy": 5.676263093948364,
      "epoch": 0.27313732041430006,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995945446775645,
      "loss": 5.5692,
      "mean_token_accuracy": 0.15726133584976196,
      "num_tokens": 6017693.0,
      "step": 3270
    },
    {
      "entropy": 5.591120767593384,
      "epoch": 0.27355496157701303,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995927563145566,
      "loss": 5.6032,
      "mean_token_accuracy": 0.16171587631106377,
      "num_tokens": 6027943.0,
      "step": 3275
    },
    {
      "entropy": 5.644890785217285,
      "epoch": 0.273972602739726,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995909640197825,
      "loss": 5.5546,
      "mean_token_accuracy": 0.16676364094018936,
      "num_tokens": 6036750.0,
      "step": 3280
    },
    {
      "entropy": 5.55715970993042,
      "epoch": 0.27439024390243905,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995891677932729,
      "loss": 5.4183,
      "mean_token_accuracy": 0.17173196673393248,
      "num_tokens": 6045722.0,
      "step": 3285
    },
    {
      "entropy": 5.5623070240020756,
      "epoch": 0.274807885065152,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995873676350597,
      "loss": 5.5829,
      "mean_token_accuracy": 0.1611317977309227,
      "num_tokens": 6055001.0,
      "step": 3290
    },
    {
      "entropy": 5.6479048252105715,
      "epoch": 0.275225526227865,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995855635451743,
      "loss": 5.5473,
      "mean_token_accuracy": 0.1628773733973503,
      "num_tokens": 6063420.0,
      "step": 3295
    },
    {
      "entropy": 5.502745962142944,
      "epoch": 0.27564316739057804,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995837555236484,
      "loss": 5.526,
      "mean_token_accuracy": 0.16666587591171264,
      "num_tokens": 6073075.0,
      "step": 3300
    },
    {
      "entropy": 5.6657538414001465,
      "epoch": 0.276060808553291,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995819435705133,
      "loss": 5.5826,
      "mean_token_accuracy": 0.16148816794157028,
      "num_tokens": 6083035.0,
      "step": 3305
    },
    {
      "entropy": 5.520883941650391,
      "epoch": 0.276478449716004,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995801276858011,
      "loss": 5.4804,
      "mean_token_accuracy": 0.16716577261686325,
      "num_tokens": 6090980.0,
      "step": 3310
    },
    {
      "entropy": 5.501849842071533,
      "epoch": 0.276896090878717,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995783078695434,
      "loss": 5.4708,
      "mean_token_accuracy": 0.170765620470047,
      "num_tokens": 6099402.0,
      "step": 3315
    },
    {
      "entropy": 5.643212795257568,
      "epoch": 0.27731373204143,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499576484121772,
      "loss": 5.5253,
      "mean_token_accuracy": 0.17255675792694092,
      "num_tokens": 6109251.0,
      "step": 3320
    },
    {
      "entropy": 5.637518215179443,
      "epoch": 0.277731373204143,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995746564425191,
      "loss": 5.6402,
      "mean_token_accuracy": 0.15654925331473352,
      "num_tokens": 6119005.0,
      "step": 3325
    },
    {
      "entropy": 5.625385427474976,
      "epoch": 0.278149014366856,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004995728248318163,
      "loss": 5.4926,
      "mean_token_accuracy": 0.168638151884079,
      "num_tokens": 6127566.0,
      "step": 3330
    },
    {
      "entropy": 5.5461078643798825,
      "epoch": 0.278566655529569,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995709892896961,
      "loss": 5.5165,
      "mean_token_accuracy": 0.1676493614912033,
      "num_tokens": 6137189.0,
      "step": 3335
    },
    {
      "entropy": 5.64869122505188,
      "epoch": 0.278984296692282,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995691498161903,
      "loss": 5.5612,
      "mean_token_accuracy": 0.1613749474287033,
      "num_tokens": 6147038.0,
      "step": 3340
    },
    {
      "entropy": 5.587482118606568,
      "epoch": 0.279401937854995,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995673064113313,
      "loss": 5.5687,
      "mean_token_accuracy": 0.16784363389015197,
      "num_tokens": 6155770.0,
      "step": 3345
    },
    {
      "entropy": 5.560370349884034,
      "epoch": 0.279819579017708,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995654590751513,
      "loss": 5.524,
      "mean_token_accuracy": 0.16733328104019166,
      "num_tokens": 6164030.0,
      "step": 3350
    },
    {
      "entropy": 5.487758684158325,
      "epoch": 0.28023722018042097,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995636078076825,
      "loss": 5.4619,
      "mean_token_accuracy": 0.16856031417846679,
      "num_tokens": 6173046.0,
      "step": 3355
    },
    {
      "entropy": 5.645107221603394,
      "epoch": 0.280654861343134,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995617526089575,
      "loss": 5.5023,
      "mean_token_accuracy": 0.1745447725057602,
      "num_tokens": 6182193.0,
      "step": 3360
    },
    {
      "entropy": 5.584012460708618,
      "epoch": 0.281072502505847,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995598934790087,
      "loss": 5.4964,
      "mean_token_accuracy": 0.1712668612599373,
      "num_tokens": 6190791.0,
      "step": 3365
    },
    {
      "entropy": 5.609999179840088,
      "epoch": 0.28149014366855996,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995580304178687,
      "loss": 5.5344,
      "mean_token_accuracy": 0.1656983897089958,
      "num_tokens": 6199245.0,
      "step": 3370
    },
    {
      "entropy": 5.507687950134278,
      "epoch": 0.281907784831273,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995561634255701,
      "loss": 5.537,
      "mean_token_accuracy": 0.1721063256263733,
      "num_tokens": 6208420.0,
      "step": 3375
    },
    {
      "entropy": 5.61139760017395,
      "epoch": 0.28232542599398597,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995542925021455,
      "loss": 5.5553,
      "mean_token_accuracy": 0.16631663143634795,
      "num_tokens": 6217059.0,
      "step": 3380
    },
    {
      "entropy": 5.538406038284302,
      "epoch": 0.28274306715669895,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995524176476277,
      "loss": 5.4275,
      "mean_token_accuracy": 0.17101717442274095,
      "num_tokens": 6225955.0,
      "step": 3385
    },
    {
      "entropy": 5.52253565788269,
      "epoch": 0.283160708319412,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995505388620497,
      "loss": 5.5014,
      "mean_token_accuracy": 0.1696779116988182,
      "num_tokens": 6235989.0,
      "step": 3390
    },
    {
      "entropy": 5.645891809463501,
      "epoch": 0.28357834948212496,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499548656145444,
      "loss": 5.527,
      "mean_token_accuracy": 0.1621383935213089,
      "num_tokens": 6244821.0,
      "step": 3395
    },
    {
      "entropy": 5.566744995117188,
      "epoch": 0.28399599064483794,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995467694978439,
      "loss": 5.6931,
      "mean_token_accuracy": 0.16034929305315018,
      "num_tokens": 6254157.0,
      "step": 3400
    },
    {
      "entropy": 5.738871717453003,
      "epoch": 0.28441363180755097,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995448789192823,
      "loss": 5.6091,
      "mean_token_accuracy": 0.1633436217904091,
      "num_tokens": 6263807.0,
      "step": 3405
    },
    {
      "entropy": 5.5871929168701175,
      "epoch": 0.28483127297026395,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995429844097922,
      "loss": 5.5241,
      "mean_token_accuracy": 0.17041266113519668,
      "num_tokens": 6273075.0,
      "step": 3410
    },
    {
      "entropy": 5.594843816757202,
      "epoch": 0.2852489141329769,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995410859694071,
      "loss": 5.4851,
      "mean_token_accuracy": 0.16699558347463608,
      "num_tokens": 6282079.0,
      "step": 3415
    },
    {
      "entropy": 5.517852783203125,
      "epoch": 0.28566655529568996,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995391835981599,
      "loss": 5.514,
      "mean_token_accuracy": 0.16514699906110764,
      "num_tokens": 6291957.0,
      "step": 3420
    },
    {
      "entropy": 5.594090270996094,
      "epoch": 0.28608419645840294,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499537277296084,
      "loss": 5.5319,
      "mean_token_accuracy": 0.1670241139829159,
      "num_tokens": 6301144.0,
      "step": 3425
    },
    {
      "entropy": 5.57917799949646,
      "epoch": 0.2865018376211159,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995353670632127,
      "loss": 5.5072,
      "mean_token_accuracy": 0.17103310972452163,
      "num_tokens": 6310654.0,
      "step": 3430
    },
    {
      "entropy": 5.5719961643219,
      "epoch": 0.28691947878382895,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995334528995797,
      "loss": 5.4683,
      "mean_token_accuracy": 0.170123153924942,
      "num_tokens": 6319562.0,
      "step": 3435
    },
    {
      "entropy": 5.5296674251556395,
      "epoch": 0.28733711994654193,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995315348052182,
      "loss": 5.511,
      "mean_token_accuracy": 0.1664817675948143,
      "num_tokens": 6329056.0,
      "step": 3440
    },
    {
      "entropy": 5.548776292800904,
      "epoch": 0.2877547611092549,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499529612780162,
      "loss": 5.4789,
      "mean_token_accuracy": 0.17050102949142457,
      "num_tokens": 6337857.0,
      "step": 3445
    },
    {
      "entropy": 5.552810192108154,
      "epoch": 0.28817240227196794,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995276868244446,
      "loss": 5.4435,
      "mean_token_accuracy": 0.16510596126317978,
      "num_tokens": 6346876.0,
      "step": 3450
    },
    {
      "entropy": 5.532234477996826,
      "epoch": 0.2885900434346809,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995257569380998,
      "loss": 5.485,
      "mean_token_accuracy": 0.169155353307724,
      "num_tokens": 6356200.0,
      "step": 3455
    },
    {
      "entropy": 5.622051954269409,
      "epoch": 0.2890076845973939,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995238231211614,
      "loss": 5.5402,
      "mean_token_accuracy": 0.1686449870467186,
      "num_tokens": 6365761.0,
      "step": 3460
    },
    {
      "entropy": 5.538770389556885,
      "epoch": 0.28942532576010693,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995218853736631,
      "loss": 5.5776,
      "mean_token_accuracy": 0.16255938410758972,
      "num_tokens": 6375326.0,
      "step": 3465
    },
    {
      "entropy": 5.648449420928955,
      "epoch": 0.2898429669228199,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995199436956391,
      "loss": 5.5974,
      "mean_token_accuracy": 0.16999968513846397,
      "num_tokens": 6384662.0,
      "step": 3470
    },
    {
      "entropy": 5.5590479373931885,
      "epoch": 0.2902606080855329,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995179980871231,
      "loss": 5.494,
      "mean_token_accuracy": 0.16821259111166,
      "num_tokens": 6393576.0,
      "step": 3475
    },
    {
      "entropy": 5.578520345687866,
      "epoch": 0.2906782492482459,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995160485481493,
      "loss": 5.4442,
      "mean_token_accuracy": 0.17245304733514785,
      "num_tokens": 6402436.0,
      "step": 3480
    },
    {
      "entropy": 5.55627121925354,
      "epoch": 0.2910958904109589,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499514095078752,
      "loss": 5.5024,
      "mean_token_accuracy": 0.16973026543855668,
      "num_tokens": 6411471.0,
      "step": 3485
    },
    {
      "entropy": 5.6449315547943115,
      "epoch": 0.2915135315736719,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499512137678965,
      "loss": 5.6409,
      "mean_token_accuracy": 0.16155346035957335,
      "num_tokens": 6421355.0,
      "step": 3490
    },
    {
      "entropy": 5.545068979263306,
      "epoch": 0.2919311727363849,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995101763488229,
      "loss": 5.5681,
      "mean_token_accuracy": 0.1627734988927841,
      "num_tokens": 6430579.0,
      "step": 3495
    },
    {
      "entropy": 5.560308027267456,
      "epoch": 0.2923488138990979,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995082110883599,
      "loss": 5.4993,
      "mean_token_accuracy": 0.1718488723039627,
      "num_tokens": 6439939.0,
      "step": 3500
    },
    {
      "entropy": 5.518435144424439,
      "epoch": 0.29276645506181087,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995062418976104,
      "loss": 5.3474,
      "mean_token_accuracy": 0.1709216669201851,
      "num_tokens": 6448598.0,
      "step": 3505
    },
    {
      "entropy": 5.572875738143921,
      "epoch": 0.2931840962245239,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995042687766089,
      "loss": 5.5441,
      "mean_token_accuracy": 0.16180262267589568,
      "num_tokens": 6457647.0,
      "step": 3510
    },
    {
      "entropy": 5.482194232940674,
      "epoch": 0.2936017373872369,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995022917253899,
      "loss": 5.3749,
      "mean_token_accuracy": 0.17720404118299485,
      "num_tokens": 6466391.0,
      "step": 3515
    },
    {
      "entropy": 5.519830751419067,
      "epoch": 0.29401937854994986,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995003107439882,
      "loss": 5.4836,
      "mean_token_accuracy": 0.17875840663909912,
      "num_tokens": 6475252.0,
      "step": 3520
    },
    {
      "entropy": 5.581446790695191,
      "epoch": 0.2944370197126629,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994983258324383,
      "loss": 5.5141,
      "mean_token_accuracy": 0.16880967915058137,
      "num_tokens": 6484959.0,
      "step": 3525
    },
    {
      "entropy": 5.558564567565918,
      "epoch": 0.2948546608753759,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994963369907748,
      "loss": 5.5838,
      "mean_token_accuracy": 0.1725819930434227,
      "num_tokens": 6495221.0,
      "step": 3530
    },
    {
      "entropy": 5.506700468063355,
      "epoch": 0.29527230203808885,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994943442190328,
      "loss": 5.4285,
      "mean_token_accuracy": 0.1691438838839531,
      "num_tokens": 6504390.0,
      "step": 3535
    },
    {
      "entropy": 5.481767988204956,
      "epoch": 0.2956899432008019,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994923475172471,
      "loss": 5.4351,
      "mean_token_accuracy": 0.17185622900724412,
      "num_tokens": 6514105.0,
      "step": 3540
    },
    {
      "entropy": 5.554193353652954,
      "epoch": 0.29610758436351486,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994903468854527,
      "loss": 5.4727,
      "mean_token_accuracy": 0.1658230811357498,
      "num_tokens": 6523673.0,
      "step": 3545
    },
    {
      "entropy": 5.5826500415802,
      "epoch": 0.29652522552622784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994883423236843,
      "loss": 5.4356,
      "mean_token_accuracy": 0.17110901027917863,
      "num_tokens": 6532158.0,
      "step": 3550
    },
    {
      "entropy": 5.557160758972168,
      "epoch": 0.2969428666889409,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994863338319775,
      "loss": 5.5548,
      "mean_token_accuracy": 0.16146010607481004,
      "num_tokens": 6541580.0,
      "step": 3555
    },
    {
      "entropy": 5.58383092880249,
      "epoch": 0.29736050785165385,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994843214103672,
      "loss": 5.4685,
      "mean_token_accuracy": 0.1715245321393013,
      "num_tokens": 6550298.0,
      "step": 3560
    },
    {
      "entropy": 5.671181774139404,
      "epoch": 0.29777814901436683,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994823050588886,
      "loss": 5.6423,
      "mean_token_accuracy": 0.1643599845468998,
      "num_tokens": 6560774.0,
      "step": 3565
    },
    {
      "entropy": 5.6066991806030275,
      "epoch": 0.29819579017707987,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499480284777577,
      "loss": 5.4538,
      "mean_token_accuracy": 0.1764713555574417,
      "num_tokens": 6568946.0,
      "step": 3570
    },
    {
      "entropy": 5.569188070297241,
      "epoch": 0.29861343133979285,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994782605664677,
      "loss": 5.5869,
      "mean_token_accuracy": 0.16506663262844085,
      "num_tokens": 6579798.0,
      "step": 3575
    },
    {
      "entropy": 5.546876001358032,
      "epoch": 0.2990310725025058,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994762324255965,
      "loss": 5.4352,
      "mean_token_accuracy": 0.16500298976898192,
      "num_tokens": 6589309.0,
      "step": 3580
    },
    {
      "entropy": 5.606908178329467,
      "epoch": 0.29944871366521886,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994742003549986,
      "loss": 5.5996,
      "mean_token_accuracy": 0.16275236755609512,
      "num_tokens": 6598835.0,
      "step": 3585
    },
    {
      "entropy": 5.657209491729736,
      "epoch": 0.29986635482793184,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994721643547095,
      "loss": 5.4766,
      "mean_token_accuracy": 0.1674236461520195,
      "num_tokens": 6608167.0,
      "step": 3590
    },
    {
      "entropy": 5.4105946063995365,
      "epoch": 0.3002839959906448,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499470124424765,
      "loss": 5.4033,
      "mean_token_accuracy": 0.17718791365623474,
      "num_tokens": 6616904.0,
      "step": 3595
    },
    {
      "entropy": 5.476817989349366,
      "epoch": 0.30070163715335785,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004994680805652008,
      "loss": 5.5156,
      "mean_token_accuracy": 0.17451889514923097,
      "num_tokens": 6625841.0,
      "step": 3600
    },
    {
      "entropy": 5.500431299209595,
      "epoch": 0.3011192783160708,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004994660327760527,
      "loss": 5.3398,
      "mean_token_accuracy": 0.1821831837296486,
      "num_tokens": 6633755.0,
      "step": 3605
    },
    {
      "entropy": 5.499605131149292,
      "epoch": 0.3015369194787838,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994639810573565,
      "loss": 5.481,
      "mean_token_accuracy": 0.17698049694299697,
      "num_tokens": 6642372.0,
      "step": 3610
    },
    {
      "entropy": 5.665440702438355,
      "epoch": 0.30195456064149684,
      "grad_norm": 1.0,
      "learning_rate": 0.000499461925409148,
      "loss": 5.458,
      "mean_token_accuracy": 0.17390639185905457,
      "num_tokens": 6652271.0,
      "step": 3615
    },
    {
      "entropy": 5.537015962600708,
      "epoch": 0.3023722018042098,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994598658314634,
      "loss": 5.5743,
      "mean_token_accuracy": 0.16500245183706283,
      "num_tokens": 6661463.0,
      "step": 3620
    },
    {
      "entropy": 5.605558681488037,
      "epoch": 0.3027898429669228,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994578023243386,
      "loss": 5.453,
      "mean_token_accuracy": 0.1709505945444107,
      "num_tokens": 6671423.0,
      "step": 3625
    },
    {
      "entropy": 5.5377373695373535,
      "epoch": 0.30320748412963583,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004994557348878098,
      "loss": 5.4589,
      "mean_token_accuracy": 0.16418076828122138,
      "num_tokens": 6680218.0,
      "step": 3630
    },
    {
      "entropy": 5.4533233642578125,
      "epoch": 0.3036251252923488,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994536635219133,
      "loss": 5.4554,
      "mean_token_accuracy": 0.16532421708106995,
      "num_tokens": 6689117.0,
      "step": 3635
    },
    {
      "entropy": 5.557939243316651,
      "epoch": 0.3040427664550618,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994515882266852,
      "loss": 5.5161,
      "mean_token_accuracy": 0.1704023465514183,
      "num_tokens": 6698024.0,
      "step": 3640
    },
    {
      "entropy": 5.60575761795044,
      "epoch": 0.3044604076177748,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994495090021618,
      "loss": 5.6485,
      "mean_token_accuracy": 0.16607705354690552,
      "num_tokens": 6706253.0,
      "step": 3645
    },
    {
      "entropy": 5.681532382965088,
      "epoch": 0.3048780487804878,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994474258483796,
      "loss": 5.5898,
      "mean_token_accuracy": 0.16927950829267502,
      "num_tokens": 6716440.0,
      "step": 3650
    },
    {
      "entropy": 5.574979066848755,
      "epoch": 0.3052956899432008,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994453387653751,
      "loss": 5.4285,
      "mean_token_accuracy": 0.17709684371948242,
      "num_tokens": 6725263.0,
      "step": 3655
    },
    {
      "entropy": 5.4725415229797365,
      "epoch": 0.3057133311059138,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994432477531847,
      "loss": 5.5857,
      "mean_token_accuracy": 0.16625414788722992,
      "num_tokens": 6735023.0,
      "step": 3660
    },
    {
      "entropy": 5.616141366958618,
      "epoch": 0.3061309722686268,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994411528118452,
      "loss": 5.5083,
      "mean_token_accuracy": 0.1685444086790085,
      "num_tokens": 6744659.0,
      "step": 3665
    },
    {
      "entropy": 5.57385892868042,
      "epoch": 0.30654861343133977,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994390539413931,
      "loss": 5.5147,
      "mean_token_accuracy": 0.1631895735859871,
      "num_tokens": 6753844.0,
      "step": 3670
    },
    {
      "entropy": 5.567027425765991,
      "epoch": 0.3069662545940528,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994369511418652,
      "loss": 5.5014,
      "mean_token_accuracy": 0.17073792219161987,
      "num_tokens": 6762574.0,
      "step": 3675
    },
    {
      "entropy": 5.649561262130737,
      "epoch": 0.3073838957567658,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994348444132983,
      "loss": 5.6345,
      "mean_token_accuracy": 0.1632675498723984,
      "num_tokens": 6772991.0,
      "step": 3680
    },
    {
      "entropy": 5.517704200744629,
      "epoch": 0.30780153691947876,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994327337557293,
      "loss": 5.4742,
      "mean_token_accuracy": 0.16668988019227982,
      "num_tokens": 6781380.0,
      "step": 3685
    },
    {
      "entropy": 5.524003982543945,
      "epoch": 0.3082191780821918,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994306191691952,
      "loss": 5.4849,
      "mean_token_accuracy": 0.1661317154765129,
      "num_tokens": 6791125.0,
      "step": 3690
    },
    {
      "entropy": 5.672194337844848,
      "epoch": 0.30863681924490477,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994285006537329,
      "loss": 5.5211,
      "mean_token_accuracy": 0.1653213694691658,
      "num_tokens": 6801038.0,
      "step": 3695
    },
    {
      "entropy": 5.461750936508179,
      "epoch": 0.30905446040761775,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994263782093796,
      "loss": 5.4273,
      "mean_token_accuracy": 0.1715899422764778,
      "num_tokens": 6810049.0,
      "step": 3700
    },
    {
      "entropy": 5.51236925125122,
      "epoch": 0.3094721015703308,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994242518361723,
      "loss": 5.431,
      "mean_token_accuracy": 0.16919803619384766,
      "num_tokens": 6819375.0,
      "step": 3705
    },
    {
      "entropy": 5.621548652648926,
      "epoch": 0.30988974273304376,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994221215341484,
      "loss": 5.5203,
      "mean_token_accuracy": 0.1661919519305229,
      "num_tokens": 6828872.0,
      "step": 3710
    },
    {
      "entropy": 5.496156454086304,
      "epoch": 0.31030738389575674,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994199873033452,
      "loss": 5.4009,
      "mean_token_accuracy": 0.17521113753318787,
      "num_tokens": 6838698.0,
      "step": 3715
    },
    {
      "entropy": 5.474640941619873,
      "epoch": 0.3107250250584698,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994178491437999,
      "loss": 5.525,
      "mean_token_accuracy": 0.17458565533161163,
      "num_tokens": 6848147.0,
      "step": 3720
    },
    {
      "entropy": 5.586211156845093,
      "epoch": 0.31114266622118275,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994157070555501,
      "loss": 5.4856,
      "mean_token_accuracy": 0.1678055003285408,
      "num_tokens": 6857777.0,
      "step": 3725
    },
    {
      "entropy": 5.558766460418701,
      "epoch": 0.31156030738389573,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994135610386331,
      "loss": 5.4467,
      "mean_token_accuracy": 0.17062433809041977,
      "num_tokens": 6866340.0,
      "step": 3730
    },
    {
      "entropy": 5.511882162094116,
      "epoch": 0.31197794854660876,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994114110930867,
      "loss": 5.4634,
      "mean_token_accuracy": 0.1747218683362007,
      "num_tokens": 6875670.0,
      "step": 3735
    },
    {
      "entropy": 5.596552181243896,
      "epoch": 0.31239558970932174,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994092572189483,
      "loss": 5.4967,
      "mean_token_accuracy": 0.16636160612106324,
      "num_tokens": 6885722.0,
      "step": 3740
    },
    {
      "entropy": 5.552028846740723,
      "epoch": 0.3128132308720348,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994070994162558,
      "loss": 5.5494,
      "mean_token_accuracy": 0.16789107471704484,
      "num_tokens": 6894267.0,
      "step": 3745
    },
    {
      "entropy": 5.624865961074829,
      "epoch": 0.31323087203474775,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004994049376850471,
      "loss": 5.5182,
      "mean_token_accuracy": 0.158856088668108,
      "num_tokens": 6904488.0,
      "step": 3750
    },
    {
      "entropy": 5.531492900848389,
      "epoch": 0.31364851319746073,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994027720253596,
      "loss": 5.4755,
      "mean_token_accuracy": 0.1683632478117943,
      "num_tokens": 6913329.0,
      "step": 3755
    },
    {
      "entropy": 5.519823312759399,
      "epoch": 0.31406615436017377,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994006024372315,
      "loss": 5.4304,
      "mean_token_accuracy": 0.17840514183044434,
      "num_tokens": 6921722.0,
      "step": 3760
    },
    {
      "entropy": 5.622691059112549,
      "epoch": 0.31448379552288674,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993984289207007,
      "loss": 5.5234,
      "mean_token_accuracy": 0.17229442149400712,
      "num_tokens": 6931183.0,
      "step": 3765
    },
    {
      "entropy": 5.582337999343872,
      "epoch": 0.3149014366855997,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993962514758054,
      "loss": 5.6009,
      "mean_token_accuracy": 0.1696538582444191,
      "num_tokens": 6939813.0,
      "step": 3770
    },
    {
      "entropy": 5.6612132549285885,
      "epoch": 0.31531907784831276,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004993940701025836,
      "loss": 5.6299,
      "mean_token_accuracy": 0.16207285374403,
      "num_tokens": 6950227.0,
      "step": 3775
    },
    {
      "entropy": 5.546162462234497,
      "epoch": 0.31573671901102573,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993918848010734,
      "loss": 5.4395,
      "mean_token_accuracy": 0.1773684024810791,
      "num_tokens": 6959971.0,
      "step": 3780
    },
    {
      "entropy": 5.462558126449585,
      "epoch": 0.3161543601737387,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993896955713133,
      "loss": 5.4037,
      "mean_token_accuracy": 0.17802790701389312,
      "num_tokens": 6968557.0,
      "step": 3785
    },
    {
      "entropy": 5.589730453491211,
      "epoch": 0.31657200133645175,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993875024133414,
      "loss": 5.4472,
      "mean_token_accuracy": 0.17268655300140381,
      "num_tokens": 6978058.0,
      "step": 3790
    },
    {
      "entropy": 5.557629013061524,
      "epoch": 0.3169896424991647,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993853053271962,
      "loss": 5.5142,
      "mean_token_accuracy": 0.17178971618413924,
      "num_tokens": 6986549.0,
      "step": 3795
    },
    {
      "entropy": 5.5617804527282715,
      "epoch": 0.3174072836618777,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993831043129161,
      "loss": 5.4836,
      "mean_token_accuracy": 0.17168736606836318,
      "num_tokens": 6995907.0,
      "step": 3800
    },
    {
      "entropy": 5.513281345367432,
      "epoch": 0.31782492482459074,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993808993705397,
      "loss": 5.3945,
      "mean_token_accuracy": 0.1667625993490219,
      "num_tokens": 7006195.0,
      "step": 3805
    },
    {
      "entropy": 5.4175151824951175,
      "epoch": 0.3182425659873037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993786905001055,
      "loss": 5.3814,
      "mean_token_accuracy": 0.17235345691442489,
      "num_tokens": 7015223.0,
      "step": 3810
    },
    {
      "entropy": 5.5245832920074465,
      "epoch": 0.3186602071500167,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993764777016524,
      "loss": 5.4344,
      "mean_token_accuracy": 0.17326589077711105,
      "num_tokens": 7024191.0,
      "step": 3815
    },
    {
      "entropy": 5.429194736480713,
      "epoch": 0.31907784831272973,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993742609752188,
      "loss": 5.4682,
      "mean_token_accuracy": 0.17181134819984437,
      "num_tokens": 7032947.0,
      "step": 3820
    },
    {
      "entropy": 5.534834289550782,
      "epoch": 0.3194954894754427,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993720403208438,
      "loss": 5.333,
      "mean_token_accuracy": 0.1813531905412674,
      "num_tokens": 7041422.0,
      "step": 3825
    },
    {
      "entropy": 5.5460773468017575,
      "epoch": 0.3199131306381557,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499369815738566,
      "loss": 5.5954,
      "mean_token_accuracy": 0.16053829938173295,
      "num_tokens": 7051051.0,
      "step": 3830
    },
    {
      "entropy": 5.490616798400879,
      "epoch": 0.3203307718008687,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993675872284248,
      "loss": 5.4714,
      "mean_token_accuracy": 0.175313900411129,
      "num_tokens": 7060116.0,
      "step": 3835
    },
    {
      "entropy": 5.624402809143066,
      "epoch": 0.3207484129635817,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993653547904586,
      "loss": 5.4635,
      "mean_token_accuracy": 0.16506005823612213,
      "num_tokens": 7068759.0,
      "step": 3840
    },
    {
      "entropy": 5.41002345085144,
      "epoch": 0.3211660541262947,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499363118424707,
      "loss": 5.3636,
      "mean_token_accuracy": 0.17519752979278563,
      "num_tokens": 7078321.0,
      "step": 3845
    },
    {
      "entropy": 5.59229097366333,
      "epoch": 0.3215836952890077,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993608781312088,
      "loss": 5.4448,
      "mean_token_accuracy": 0.17415895462036132,
      "num_tokens": 7088126.0,
      "step": 3850
    },
    {
      "entropy": 5.495071697235107,
      "epoch": 0.3220013364517207,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993586339100035,
      "loss": 5.4175,
      "mean_token_accuracy": 0.16729627251625062,
      "num_tokens": 7096948.0,
      "step": 3855
    },
    {
      "entropy": 5.383945178985596,
      "epoch": 0.32241897761443367,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993563857611301,
      "loss": 5.3991,
      "mean_token_accuracy": 0.17286474704742433,
      "num_tokens": 7106867.0,
      "step": 3860
    },
    {
      "entropy": 5.511750364303589,
      "epoch": 0.3228366187771467,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993541336846283,
      "loss": 5.5111,
      "mean_token_accuracy": 0.1696945309638977,
      "num_tokens": 7115971.0,
      "step": 3865
    },
    {
      "entropy": 5.57295298576355,
      "epoch": 0.3232542599398597,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993518776805372,
      "loss": 5.3619,
      "mean_token_accuracy": 0.1746579512953758,
      "num_tokens": 7124659.0,
      "step": 3870
    },
    {
      "entropy": 5.586831712722779,
      "epoch": 0.32367190110257266,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993496177488965,
      "loss": 5.6524,
      "mean_token_accuracy": 0.16620706468820573,
      "num_tokens": 7133456.0,
      "step": 3875
    },
    {
      "entropy": 5.535101461410522,
      "epoch": 0.3240895422652857,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993473538897457,
      "loss": 5.4391,
      "mean_token_accuracy": 0.17057373225688935,
      "num_tokens": 7142649.0,
      "step": 3880
    },
    {
      "entropy": 5.619561052322387,
      "epoch": 0.32450718342799867,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993450861031243,
      "loss": 5.5659,
      "mean_token_accuracy": 0.16672575026750563,
      "num_tokens": 7153151.0,
      "step": 3885
    },
    {
      "entropy": 5.502790641784668,
      "epoch": 0.32492482459071165,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993428143890722,
      "loss": 5.4504,
      "mean_token_accuracy": 0.1734729528427124,
      "num_tokens": 7161638.0,
      "step": 3890
    },
    {
      "entropy": 5.493521595001221,
      "epoch": 0.3253424657534247,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993405387476291,
      "loss": 5.4303,
      "mean_token_accuracy": 0.1750929281115532,
      "num_tokens": 7170832.0,
      "step": 3895
    },
    {
      "entropy": 5.491485261917115,
      "epoch": 0.32576010691613766,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993382591788349,
      "loss": 5.4248,
      "mean_token_accuracy": 0.1764598712325096,
      "num_tokens": 7180025.0,
      "step": 3900
    },
    {
      "entropy": 5.5754554271698,
      "epoch": 0.32617774807885064,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993359756827293,
      "loss": 5.4719,
      "mean_token_accuracy": 0.17411546036601067,
      "num_tokens": 7189287.0,
      "step": 3905
    },
    {
      "entropy": 5.603695440292358,
      "epoch": 0.32659538924156367,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993336882593524,
      "loss": 5.4604,
      "mean_token_accuracy": 0.1673969268798828,
      "num_tokens": 7198438.0,
      "step": 3910
    },
    {
      "entropy": 5.409713411331177,
      "epoch": 0.32701303040427665,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993313969087444,
      "loss": 5.4936,
      "mean_token_accuracy": 0.17143895775079726,
      "num_tokens": 7208006.0,
      "step": 3915
    },
    {
      "entropy": 5.584083938598633,
      "epoch": 0.32743067156698963,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993291016309452,
      "loss": 5.4972,
      "mean_token_accuracy": 0.16895750910043716,
      "num_tokens": 7216975.0,
      "step": 3920
    },
    {
      "entropy": 5.5208659172058105,
      "epoch": 0.32784831272970266,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499326802425995,
      "loss": 5.489,
      "mean_token_accuracy": 0.16583355739712716,
      "num_tokens": 7225664.0,
      "step": 3925
    },
    {
      "entropy": 5.565141344070435,
      "epoch": 0.32826595389241564,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993244992939342,
      "loss": 5.5415,
      "mean_token_accuracy": 0.1687723606824875,
      "num_tokens": 7234780.0,
      "step": 3930
    },
    {
      "entropy": 5.550216579437256,
      "epoch": 0.3286835950551286,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499322192234803,
      "loss": 5.3683,
      "mean_token_accuracy": 0.1774860978126526,
      "num_tokens": 7242648.0,
      "step": 3935
    },
    {
      "entropy": 5.367749166488648,
      "epoch": 0.32910123621784165,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993198812486418,
      "loss": 5.4683,
      "mean_token_accuracy": 0.1734187424182892,
      "num_tokens": 7251856.0,
      "step": 3940
    },
    {
      "entropy": 5.658477258682251,
      "epoch": 0.32951887738055463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499317566335491,
      "loss": 5.4898,
      "mean_token_accuracy": 0.17640756517648698,
      "num_tokens": 7260552.0,
      "step": 3945
    },
    {
      "entropy": 5.52726149559021,
      "epoch": 0.3299365185432676,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993152474953913,
      "loss": 5.4654,
      "mean_token_accuracy": 0.1719620943069458,
      "num_tokens": 7270328.0,
      "step": 3950
    },
    {
      "entropy": 5.438219451904297,
      "epoch": 0.33035415970598064,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993129247283831,
      "loss": 5.4562,
      "mean_token_accuracy": 0.17517218589782715,
      "num_tokens": 7279797.0,
      "step": 3955
    },
    {
      "entropy": 5.579410600662231,
      "epoch": 0.3307718008686936,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993105980345071,
      "loss": 5.5082,
      "mean_token_accuracy": 0.16999913454055787,
      "num_tokens": 7290205.0,
      "step": 3960
    },
    {
      "entropy": 5.531829833984375,
      "epoch": 0.3311894420314066,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993082674138042,
      "loss": 5.4724,
      "mean_token_accuracy": 0.17171881347894669,
      "num_tokens": 7298544.0,
      "step": 3965
    },
    {
      "entropy": 5.451128578186035,
      "epoch": 0.33160708319411963,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499305932866315,
      "loss": 5.4439,
      "mean_token_accuracy": 0.17368571907281877,
      "num_tokens": 7307662.0,
      "step": 3970
    },
    {
      "entropy": 5.569617891311646,
      "epoch": 0.3320247243568326,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993035943920805,
      "loss": 5.4979,
      "mean_token_accuracy": 0.1646683320403099,
      "num_tokens": 7317085.0,
      "step": 3975
    },
    {
      "entropy": 5.573373508453369,
      "epoch": 0.3324423655195456,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993012519911416,
      "loss": 5.4695,
      "mean_token_accuracy": 0.17162806242704393,
      "num_tokens": 7326190.0,
      "step": 3980
    },
    {
      "entropy": 5.546729230880738,
      "epoch": 0.3328600066822586,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992989056635392,
      "loss": 5.4943,
      "mean_token_accuracy": 0.1697179466485977,
      "num_tokens": 7336143.0,
      "step": 3985
    },
    {
      "entropy": 5.567217826843262,
      "epoch": 0.3332776478449716,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004992965554093146,
      "loss": 5.4788,
      "mean_token_accuracy": 0.16804200112819673,
      "num_tokens": 7346024.0,
      "step": 3990
    },
    {
      "entropy": 5.536852121353149,
      "epoch": 0.3336952890076846,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992942012285086,
      "loss": 5.5328,
      "mean_token_accuracy": 0.16853790283203124,
      "num_tokens": 7355477.0,
      "step": 3995
    },
    {
      "entropy": 5.608371210098267,
      "epoch": 0.3341129301703976,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004992918431211629,
      "loss": 5.363,
      "mean_token_accuracy": 0.17802691608667373,
      "num_tokens": 7364720.0,
      "step": 4000
    },
    {
      "entropy": 5.42276668548584,
      "epoch": 0.3345305713331106,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992894810873183,
      "loss": 5.3953,
      "mean_token_accuracy": 0.1722881615161896,
      "num_tokens": 7373793.0,
      "step": 4005
    },
    {
      "entropy": 5.514430809020996,
      "epoch": 0.33494821249582357,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992871151270165,
      "loss": 5.5193,
      "mean_token_accuracy": 0.16925701051950454,
      "num_tokens": 7383886.0,
      "step": 4010
    },
    {
      "entropy": 5.647746133804321,
      "epoch": 0.3353658536585366,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992847452402986,
      "loss": 5.573,
      "mean_token_accuracy": 0.16909781023859977,
      "num_tokens": 7393617.0,
      "step": 4015
    },
    {
      "entropy": 5.511100721359253,
      "epoch": 0.3357834948212496,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992823714272064,
      "loss": 5.3591,
      "mean_token_accuracy": 0.18106982558965684,
      "num_tokens": 7401949.0,
      "step": 4020
    },
    {
      "entropy": 5.340199756622314,
      "epoch": 0.33620113598396256,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992799936877813,
      "loss": 5.3761,
      "mean_token_accuracy": 0.17852676808834075,
      "num_tokens": 7410180.0,
      "step": 4025
    },
    {
      "entropy": 5.602356719970703,
      "epoch": 0.3366187771466756,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992776120220649,
      "loss": 5.577,
      "mean_token_accuracy": 0.1651108369231224,
      "num_tokens": 7420506.0,
      "step": 4030
    },
    {
      "entropy": 5.607547473907471,
      "epoch": 0.3370364183093886,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992752264300989,
      "loss": 5.4913,
      "mean_token_accuracy": 0.1773996263742447,
      "num_tokens": 7429644.0,
      "step": 4035
    },
    {
      "entropy": 5.612040662765503,
      "epoch": 0.33745405947210155,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992728369119252,
      "loss": 5.5258,
      "mean_token_accuracy": 0.17017731964588165,
      "num_tokens": 7439129.0,
      "step": 4040
    },
    {
      "entropy": 5.4839338779449465,
      "epoch": 0.3378717006348146,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004992704434675855,
      "loss": 5.5053,
      "mean_token_accuracy": 0.17029713541269303,
      "num_tokens": 7449355.0,
      "step": 4045
    },
    {
      "entropy": 5.56695556640625,
      "epoch": 0.33828934179752757,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992680460971218,
      "loss": 5.483,
      "mean_token_accuracy": 0.1702693894505501,
      "num_tokens": 7458763.0,
      "step": 4050
    },
    {
      "entropy": 5.539506673812866,
      "epoch": 0.33870698296024054,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992656448005758,
      "loss": 5.3955,
      "mean_token_accuracy": 0.17542738765478133,
      "num_tokens": 7467697.0,
      "step": 4055
    },
    {
      "entropy": 5.481296968460083,
      "epoch": 0.3391246241229536,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004992632395779899,
      "loss": 5.5317,
      "mean_token_accuracy": 0.16871478855609895,
      "num_tokens": 7476728.0,
      "step": 4060
    },
    {
      "entropy": 5.515885925292968,
      "epoch": 0.33954226528566656,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992608304294061,
      "loss": 5.4981,
      "mean_token_accuracy": 0.17235045433044432,
      "num_tokens": 7486794.0,
      "step": 4065
    },
    {
      "entropy": 5.618357086181641,
      "epoch": 0.33995990644837953,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992584173548664,
      "loss": 5.4498,
      "mean_token_accuracy": 0.17357994019985198,
      "num_tokens": 7496286.0,
      "step": 4070
    },
    {
      "entropy": 5.474039125442505,
      "epoch": 0.34037754761109257,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992560003544131,
      "loss": 5.4542,
      "mean_token_accuracy": 0.17892095372080802,
      "num_tokens": 7505670.0,
      "step": 4075
    },
    {
      "entropy": 5.562310600280762,
      "epoch": 0.34079518877380555,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992535794280887,
      "loss": 5.5205,
      "mean_token_accuracy": 0.17938163354992867,
      "num_tokens": 7514702.0,
      "step": 4080
    },
    {
      "entropy": 5.54784574508667,
      "epoch": 0.3412128299365185,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992511545759354,
      "loss": 5.4674,
      "mean_token_accuracy": 0.17029086202383042,
      "num_tokens": 7523705.0,
      "step": 4085
    },
    {
      "entropy": 5.482650947570801,
      "epoch": 0.34163047109923156,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992487257979958,
      "loss": 5.3594,
      "mean_token_accuracy": 0.1778412491083145,
      "num_tokens": 7532491.0,
      "step": 4090
    },
    {
      "entropy": 5.489047718048096,
      "epoch": 0.34204811226194454,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992462930943123,
      "loss": 5.409,
      "mean_token_accuracy": 0.16939541548490525,
      "num_tokens": 7540761.0,
      "step": 4095
    },
    {
      "entropy": 5.589512634277344,
      "epoch": 0.3424657534246575,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004992438564649275,
      "loss": 5.4827,
      "mean_token_accuracy": 0.1699928969144821,
      "num_tokens": 7550897.0,
      "step": 4100
    },
    {
      "entropy": 5.49046139717102,
      "epoch": 0.34288339458737055,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004992414159098841,
      "loss": 5.3683,
      "mean_token_accuracy": 0.1745118945837021,
      "num_tokens": 7559982.0,
      "step": 4105
    },
    {
      "entropy": 5.518819141387939,
      "epoch": 0.3433010357500835,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992389714292247,
      "loss": 5.4551,
      "mean_token_accuracy": 0.16522257626056672,
      "num_tokens": 7569358.0,
      "step": 4110
    },
    {
      "entropy": 5.5108489990234375,
      "epoch": 0.3437186769127965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992365230229922,
      "loss": 5.4217,
      "mean_token_accuracy": 0.17053637355566026,
      "num_tokens": 7578437.0,
      "step": 4115
    },
    {
      "entropy": 5.44957389831543,
      "epoch": 0.34413631807550954,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992340706912295,
      "loss": 5.3934,
      "mean_token_accuracy": 0.1758326917886734,
      "num_tokens": 7587528.0,
      "step": 4120
    },
    {
      "entropy": 5.625531435012817,
      "epoch": 0.3445539592382225,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992316144339794,
      "loss": 5.4328,
      "mean_token_accuracy": 0.18261457085609437,
      "num_tokens": 7595991.0,
      "step": 4125
    },
    {
      "entropy": 5.461813974380493,
      "epoch": 0.3449716004009355,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992291542512852,
      "loss": 5.3865,
      "mean_token_accuracy": 0.17654262781143187,
      "num_tokens": 7604538.0,
      "step": 4130
    },
    {
      "entropy": 5.406788349151611,
      "epoch": 0.34538924156364853,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992266901431895,
      "loss": 5.4169,
      "mean_token_accuracy": 0.17029373198747635,
      "num_tokens": 7613853.0,
      "step": 4135
    },
    {
      "entropy": 5.491995239257813,
      "epoch": 0.3458068827263615,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992242221097358,
      "loss": 5.3651,
      "mean_token_accuracy": 0.17852593064308167,
      "num_tokens": 7623538.0,
      "step": 4140
    },
    {
      "entropy": 5.483587026596069,
      "epoch": 0.3462245238890745,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992217501509672,
      "loss": 5.3846,
      "mean_token_accuracy": 0.17738809436559677,
      "num_tokens": 7632121.0,
      "step": 4145
    },
    {
      "entropy": 5.5288893699646,
      "epoch": 0.3466421650517875,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499219274266927,
      "loss": 5.4581,
      "mean_token_accuracy": 0.1706435889005661,
      "num_tokens": 7640779.0,
      "step": 4150
    },
    {
      "entropy": 5.489064121246338,
      "epoch": 0.3470598062145005,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992167944576584,
      "loss": 5.3844,
      "mean_token_accuracy": 0.16829655468463897,
      "num_tokens": 7649138.0,
      "step": 4155
    },
    {
      "entropy": 5.43375883102417,
      "epoch": 0.3474774473772135,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499214310723205,
      "loss": 5.3733,
      "mean_token_accuracy": 0.18074534982442855,
      "num_tokens": 7658510.0,
      "step": 4160
    },
    {
      "entropy": 5.4592814445495605,
      "epoch": 0.3478950885399265,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992118230636102,
      "loss": 5.3127,
      "mean_token_accuracy": 0.18564201146364212,
      "num_tokens": 7666946.0,
      "step": 4165
    },
    {
      "entropy": 5.445179462432861,
      "epoch": 0.3483127297026395,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992093314789176,
      "loss": 5.3549,
      "mean_token_accuracy": 0.17630248218774797,
      "num_tokens": 7675572.0,
      "step": 4170
    },
    {
      "entropy": 5.374283218383789,
      "epoch": 0.34873037086535247,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992068359691707,
      "loss": 5.3844,
      "mean_token_accuracy": 0.17325544804334642,
      "num_tokens": 7684719.0,
      "step": 4175
    },
    {
      "entropy": 5.452366304397583,
      "epoch": 0.3491480120280655,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992043365344132,
      "loss": 5.4128,
      "mean_token_accuracy": 0.16606821864843369,
      "num_tokens": 7694370.0,
      "step": 4180
    },
    {
      "entropy": 5.538264179229737,
      "epoch": 0.3495656531907785,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992018331746891,
      "loss": 5.4927,
      "mean_token_accuracy": 0.16647274494171144,
      "num_tokens": 7703572.0,
      "step": 4185
    },
    {
      "entropy": 5.54348931312561,
      "epoch": 0.34998329435349146,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991993258900418,
      "loss": 5.4592,
      "mean_token_accuracy": 0.17307280451059343,
      "num_tokens": 7712732.0,
      "step": 4190
    },
    {
      "entropy": 5.545743179321289,
      "epoch": 0.3504009355162045,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991968146805156,
      "loss": 5.5309,
      "mean_token_accuracy": 0.16450876593589783,
      "num_tokens": 7722213.0,
      "step": 4195
    },
    {
      "entropy": 5.567177391052246,
      "epoch": 0.35081857667891747,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991942995461542,
      "loss": 5.4382,
      "mean_token_accuracy": 0.16758489459753037,
      "num_tokens": 7732228.0,
      "step": 4200
    },
    {
      "entropy": 5.551746988296509,
      "epoch": 0.35123621784163045,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991917804870018,
      "loss": 5.4993,
      "mean_token_accuracy": 0.16597847491502762,
      "num_tokens": 7741662.0,
      "step": 4205
    },
    {
      "entropy": 5.602847003936768,
      "epoch": 0.3516538590043435,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991892575031024,
      "loss": 5.4378,
      "mean_token_accuracy": 0.17829289734363557,
      "num_tokens": 7749608.0,
      "step": 4210
    },
    {
      "entropy": 5.340467071533203,
      "epoch": 0.35207150016705646,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991867305945001,
      "loss": 5.4167,
      "mean_token_accuracy": 0.17135734856128693,
      "num_tokens": 7758387.0,
      "step": 4215
    },
    {
      "entropy": 5.479982471466064,
      "epoch": 0.35248914132976944,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991841997612394,
      "loss": 5.45,
      "mean_token_accuracy": 0.16724161356687545,
      "num_tokens": 7768103.0,
      "step": 4220
    },
    {
      "entropy": 5.4644652843475345,
      "epoch": 0.3529067824924825,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991816650033643,
      "loss": 5.2988,
      "mean_token_accuracy": 0.17720498591661454,
      "num_tokens": 7776826.0,
      "step": 4225
    },
    {
      "entropy": 5.408760690689087,
      "epoch": 0.35332442365519545,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991791263209194,
      "loss": 5.419,
      "mean_token_accuracy": 0.1694740518927574,
      "num_tokens": 7785866.0,
      "step": 4230
    },
    {
      "entropy": 5.6010815620422365,
      "epoch": 0.35374206481790843,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991765837139491,
      "loss": 5.4461,
      "mean_token_accuracy": 0.16579159647226333,
      "num_tokens": 7795140.0,
      "step": 4235
    },
    {
      "entropy": 5.424951791763306,
      "epoch": 0.35415970598062146,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991740371824977,
      "loss": 5.4321,
      "mean_token_accuracy": 0.17311078310012817,
      "num_tokens": 7804256.0,
      "step": 4240
    },
    {
      "entropy": 5.505328559875489,
      "epoch": 0.35457734714333444,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049917148672661,
      "loss": 5.3538,
      "mean_token_accuracy": 0.18129236549139022,
      "num_tokens": 7812295.0,
      "step": 4245
    },
    {
      "entropy": 5.6016477108001705,
      "epoch": 0.3549949883060474,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991689323463306,
      "loss": 5.4417,
      "mean_token_accuracy": 0.17255054265260697,
      "num_tokens": 7821208.0,
      "step": 4250
    },
    {
      "entropy": 5.403783226013184,
      "epoch": 0.35541262946876045,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991663740417042,
      "loss": 5.4886,
      "mean_token_accuracy": 0.16946886703372002,
      "num_tokens": 7830542.0,
      "step": 4255
    },
    {
      "entropy": 5.468491840362549,
      "epoch": 0.35583027063147343,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991638118127757,
      "loss": 5.2673,
      "mean_token_accuracy": 0.18574574887752532,
      "num_tokens": 7839469.0,
      "step": 4260
    },
    {
      "entropy": 5.495815372467041,
      "epoch": 0.3562479117941864,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991612456595896,
      "loss": 5.4631,
      "mean_token_accuracy": 0.17240444272756578,
      "num_tokens": 7849350.0,
      "step": 4265
    },
    {
      "entropy": 5.512340116500854,
      "epoch": 0.35666555295689945,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991586755821913,
      "loss": 5.3893,
      "mean_token_accuracy": 0.1797077938914299,
      "num_tokens": 7858882.0,
      "step": 4270
    },
    {
      "entropy": 5.492542552947998,
      "epoch": 0.3570831941196124,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991561015806254,
      "loss": 5.4564,
      "mean_token_accuracy": 0.17558517456054687,
      "num_tokens": 7868704.0,
      "step": 4275
    },
    {
      "entropy": 5.568697166442871,
      "epoch": 0.3575008352823254,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991535236549372,
      "loss": 5.4523,
      "mean_token_accuracy": 0.17207656651735306,
      "num_tokens": 7878103.0,
      "step": 4280
    },
    {
      "entropy": 5.341040420532226,
      "epoch": 0.35791847644503844,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991509418051718,
      "loss": 5.2134,
      "mean_token_accuracy": 0.18514634370803834,
      "num_tokens": 7887228.0,
      "step": 4285
    },
    {
      "entropy": 5.495812606811524,
      "epoch": 0.3583361176077514,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991483560313743,
      "loss": 5.4461,
      "mean_token_accuracy": 0.17493630200624466,
      "num_tokens": 7897205.0,
      "step": 4290
    },
    {
      "entropy": 5.463666105270386,
      "epoch": 0.3587537587704644,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049914576633359,
      "loss": 5.3476,
      "mean_token_accuracy": 0.17855667918920518,
      "num_tokens": 7906336.0,
      "step": 4295
    },
    {
      "entropy": 5.425225639343262,
      "epoch": 0.3591713999331774,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991431727118643,
      "loss": 5.3859,
      "mean_token_accuracy": 0.17622471898794173,
      "num_tokens": 7914773.0,
      "step": 4300
    },
    {
      "entropy": 5.416539430618286,
      "epoch": 0.3595890410958904,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991405751662426,
      "loss": 5.3227,
      "mean_token_accuracy": 0.181936876475811,
      "num_tokens": 7922759.0,
      "step": 4305
    },
    {
      "entropy": 5.421218681335449,
      "epoch": 0.3600066822586034,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991379736967703,
      "loss": 5.33,
      "mean_token_accuracy": 0.17022594213485717,
      "num_tokens": 7932199.0,
      "step": 4310
    },
    {
      "entropy": 5.403014087677002,
      "epoch": 0.3604243234213164,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991353683034929,
      "loss": 5.3416,
      "mean_token_accuracy": 0.17926559150218963,
      "num_tokens": 7941045.0,
      "step": 4315
    },
    {
      "entropy": 5.430142784118653,
      "epoch": 0.3608419645840294,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991327589864563,
      "loss": 5.3243,
      "mean_token_accuracy": 0.18629411309957505,
      "num_tokens": 7949339.0,
      "step": 4320
    },
    {
      "entropy": 5.44664740562439,
      "epoch": 0.3612596057467424,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991301457457057,
      "loss": 5.3631,
      "mean_token_accuracy": 0.17302195131778716,
      "num_tokens": 7958182.0,
      "step": 4325
    },
    {
      "entropy": 5.492476940155029,
      "epoch": 0.3616772469094554,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004991275285812873,
      "loss": 5.4332,
      "mean_token_accuracy": 0.17579357475042343,
      "num_tokens": 7966711.0,
      "step": 4330
    },
    {
      "entropy": 5.4342657089233395,
      "epoch": 0.3620948880721684,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991249074932466,
      "loss": 5.3002,
      "mean_token_accuracy": 0.1751624494791031,
      "num_tokens": 7974864.0,
      "step": 4335
    },
    {
      "entropy": 5.4517271518707275,
      "epoch": 0.36251252923488136,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991222824816296,
      "loss": 5.463,
      "mean_token_accuracy": 0.16784718185663222,
      "num_tokens": 7984812.0,
      "step": 4340
    },
    {
      "entropy": 5.4975443363189695,
      "epoch": 0.3629301703975944,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991196535464824,
      "loss": 5.434,
      "mean_token_accuracy": 0.16960262209177018,
      "num_tokens": 7994862.0,
      "step": 4345
    },
    {
      "entropy": 5.454056358337402,
      "epoch": 0.3633478115603074,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991170206878508,
      "loss": 5.3731,
      "mean_token_accuracy": 0.17012314200401307,
      "num_tokens": 8003943.0,
      "step": 4350
    },
    {
      "entropy": 5.456601524353028,
      "epoch": 0.36376545272302035,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499114383905781,
      "loss": 5.3758,
      "mean_token_accuracy": 0.17089642137289046,
      "num_tokens": 8012841.0,
      "step": 4355
    },
    {
      "entropy": 5.4201741218566895,
      "epoch": 0.3641830938857334,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991117432003192,
      "loss": 5.3394,
      "mean_token_accuracy": 0.1764964520931244,
      "num_tokens": 8021766.0,
      "step": 4360
    },
    {
      "entropy": 5.509561347961426,
      "epoch": 0.36460073504844637,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991090985715116,
      "loss": 5.3568,
      "mean_token_accuracy": 0.17222451120615007,
      "num_tokens": 8030491.0,
      "step": 4365
    },
    {
      "entropy": 5.377471208572388,
      "epoch": 0.36501837621115935,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991064500194043,
      "loss": 5.3375,
      "mean_token_accuracy": 0.17822521328926086,
      "num_tokens": 8039325.0,
      "step": 4370
    },
    {
      "entropy": 5.492970132827759,
      "epoch": 0.3654360173738724,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991037975440439,
      "loss": 5.4225,
      "mean_token_accuracy": 0.17352834790945054,
      "num_tokens": 8048673.0,
      "step": 4375
    },
    {
      "entropy": 5.6428838729858395,
      "epoch": 0.36585365853658536,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991011411454769,
      "loss": 5.58,
      "mean_token_accuracy": 0.16775976866483688,
      "num_tokens": 8057934.0,
      "step": 4380
    },
    {
      "entropy": 5.385854768753052,
      "epoch": 0.36627129969929834,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004990984808237495,
      "loss": 5.3637,
      "mean_token_accuracy": 0.1815803900361061,
      "num_tokens": 8066923.0,
      "step": 4385
    },
    {
      "entropy": 5.499070501327514,
      "epoch": 0.36668894086201137,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990958165789086,
      "loss": 5.4665,
      "mean_token_accuracy": 0.16589635163545607,
      "num_tokens": 8076263.0,
      "step": 4390
    },
    {
      "entropy": 5.543005561828613,
      "epoch": 0.36710658202472435,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004990931484110005,
      "loss": 5.4713,
      "mean_token_accuracy": 0.1719420939683914,
      "num_tokens": 8085741.0,
      "step": 4395
    },
    {
      "entropy": 5.430317115783692,
      "epoch": 0.3675242231874373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990904763200722,
      "loss": 5.4401,
      "mean_token_accuracy": 0.1729919657111168,
      "num_tokens": 8094646.0,
      "step": 4400
    },
    {
      "entropy": 5.463187503814697,
      "epoch": 0.36794186435015036,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990878003061705,
      "loss": 5.4221,
      "mean_token_accuracy": 0.1726178929209709,
      "num_tokens": 8103215.0,
      "step": 4405
    },
    {
      "entropy": 5.4622879981994625,
      "epoch": 0.36835950551286334,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990851203693419,
      "loss": 5.2573,
      "mean_token_accuracy": 0.1726316198706627,
      "num_tokens": 8112381.0,
      "step": 4410
    },
    {
      "entropy": 5.422245883941651,
      "epoch": 0.3687771466755763,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990824365096336,
      "loss": 5.3613,
      "mean_token_accuracy": 0.17220450937747955,
      "num_tokens": 8121139.0,
      "step": 4415
    },
    {
      "entropy": 5.507014894485474,
      "epoch": 0.36919478783828935,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990797487270926,
      "loss": 5.4698,
      "mean_token_accuracy": 0.16886930018663407,
      "num_tokens": 8130011.0,
      "step": 4420
    },
    {
      "entropy": 5.4787599563598635,
      "epoch": 0.36961242900100233,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990770570217659,
      "loss": 5.4126,
      "mean_token_accuracy": 0.17649679034948348,
      "num_tokens": 8138757.0,
      "step": 4425
    },
    {
      "entropy": 5.397525072097778,
      "epoch": 0.37003007016371536,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004990743613937005,
      "loss": 5.3981,
      "mean_token_accuracy": 0.172449491918087,
      "num_tokens": 8147862.0,
      "step": 4430
    },
    {
      "entropy": 5.484035444259644,
      "epoch": 0.37044771132642834,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990716618429436,
      "loss": 5.3451,
      "mean_token_accuracy": 0.18334268629550934,
      "num_tokens": 8156754.0,
      "step": 4435
    },
    {
      "entropy": 5.413478517532349,
      "epoch": 0.3708653524891413,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990689583695428,
      "loss": 5.4094,
      "mean_token_accuracy": 0.18084121197462083,
      "num_tokens": 8165647.0,
      "step": 4440
    },
    {
      "entropy": 5.431940793991089,
      "epoch": 0.37128299365185435,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499066250973545,
      "loss": 5.347,
      "mean_token_accuracy": 0.174612420797348,
      "num_tokens": 8174811.0,
      "step": 4445
    },
    {
      "entropy": 5.43572039604187,
      "epoch": 0.37170063481456733,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004990635396549978,
      "loss": 5.3379,
      "mean_token_accuracy": 0.17819695621728898,
      "num_tokens": 8184270.0,
      "step": 4450
    },
    {
      "entropy": 5.539177322387696,
      "epoch": 0.3721182759772803,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990608244139486,
      "loss": 5.4612,
      "mean_token_accuracy": 0.1732149139046669,
      "num_tokens": 8193209.0,
      "step": 4455
    },
    {
      "entropy": 5.62060956954956,
      "epoch": 0.37253591713999334,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499058105250445,
      "loss": 5.5469,
      "mean_token_accuracy": 0.16869076937437058,
      "num_tokens": 8202502.0,
      "step": 4460
    },
    {
      "entropy": 5.546969890594482,
      "epoch": 0.3729535583027063,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990553821645346,
      "loss": 5.4956,
      "mean_token_accuracy": 0.17091006189584732,
      "num_tokens": 8212226.0,
      "step": 4465
    },
    {
      "entropy": 5.5280375480651855,
      "epoch": 0.3733711994654193,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499052655156265,
      "loss": 5.4079,
      "mean_token_accuracy": 0.17574627101421356,
      "num_tokens": 8220729.0,
      "step": 4470
    },
    {
      "entropy": 5.452630853652954,
      "epoch": 0.37378884062813234,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990499242256839,
      "loss": 5.4652,
      "mean_token_accuracy": 0.1738543137907982,
      "num_tokens": 8230385.0,
      "step": 4475
    },
    {
      "entropy": 5.495773553848267,
      "epoch": 0.3742064817908453,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990471893728394,
      "loss": 5.391,
      "mean_token_accuracy": 0.17303333729505538,
      "num_tokens": 8240057.0,
      "step": 4480
    },
    {
      "entropy": 5.340406799316407,
      "epoch": 0.3746241229535583,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499044450597779,
      "loss": 5.2553,
      "mean_token_accuracy": 0.18496009856462478,
      "num_tokens": 8249903.0,
      "step": 4485
    },
    {
      "entropy": 5.464671802520752,
      "epoch": 0.3750417641162713,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004990417079005508,
      "loss": 5.396,
      "mean_token_accuracy": 0.17637877762317658,
      "num_tokens": 8259334.0,
      "step": 4490
    },
    {
      "entropy": 5.3831353187561035,
      "epoch": 0.3754594052789843,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499038961281203,
      "loss": 5.2977,
      "mean_token_accuracy": 0.18332188874483107,
      "num_tokens": 8268543.0,
      "step": 4495
    },
    {
      "entropy": 5.516524982452393,
      "epoch": 0.3758770464416973,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990362107397834,
      "loss": 5.4426,
      "mean_token_accuracy": 0.17564595937728883,
      "num_tokens": 8278257.0,
      "step": 4500
    },
    {
      "entropy": 5.46676197052002,
      "epoch": 0.3762946876044103,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990334562763401,
      "loss": 5.3112,
      "mean_token_accuracy": 0.18334500193595887,
      "num_tokens": 8287506.0,
      "step": 4505
    },
    {
      "entropy": 5.370918893814087,
      "epoch": 0.3767123287671233,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990306978909216,
      "loss": 5.328,
      "mean_token_accuracy": 0.182416196167469,
      "num_tokens": 8295803.0,
      "step": 4510
    },
    {
      "entropy": 5.386123561859131,
      "epoch": 0.3771299699298363,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990279355835761,
      "loss": 5.3914,
      "mean_token_accuracy": 0.1722537860274315,
      "num_tokens": 8305344.0,
      "step": 4515
    },
    {
      "entropy": 5.535409212112427,
      "epoch": 0.3775476110925493,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990251693543519,
      "loss": 5.4944,
      "mean_token_accuracy": 0.16697886884212493,
      "num_tokens": 8315226.0,
      "step": 4520
    },
    {
      "entropy": 5.498746633529663,
      "epoch": 0.3779652522552623,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990223992032975,
      "loss": 5.4555,
      "mean_token_accuracy": 0.1674830958247185,
      "num_tokens": 8324614.0,
      "step": 4525
    },
    {
      "entropy": 5.3684216976165775,
      "epoch": 0.37838289341797526,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004990196251304612,
      "loss": 5.2618,
      "mean_token_accuracy": 0.17755249738693238,
      "num_tokens": 8334014.0,
      "step": 4530
    },
    {
      "entropy": 5.432075262069702,
      "epoch": 0.3788005345806883,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499016847135892,
      "loss": 5.3333,
      "mean_token_accuracy": 0.17807306349277496,
      "num_tokens": 8343038.0,
      "step": 4535
    },
    {
      "entropy": 5.534339857101441,
      "epoch": 0.3792181757434013,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499014065219638,
      "loss": 5.4958,
      "mean_token_accuracy": 0.16782399713993074,
      "num_tokens": 8353170.0,
      "step": 4540
    },
    {
      "entropy": 5.49351806640625,
      "epoch": 0.37963581690611425,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990112793817481,
      "loss": 5.3361,
      "mean_token_accuracy": 0.18302312940359117,
      "num_tokens": 8362240.0,
      "step": 4545
    },
    {
      "entropy": 5.361939573287964,
      "epoch": 0.3800534580688273,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004990084896222713,
      "loss": 5.3468,
      "mean_token_accuracy": 0.17659430652856828,
      "num_tokens": 8370723.0,
      "step": 4550
    },
    {
      "entropy": 5.353510475158691,
      "epoch": 0.38047109923154027,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004990056959412562,
      "loss": 5.2238,
      "mean_token_accuracy": 0.18136699199676515,
      "num_tokens": 8379282.0,
      "step": 4555
    },
    {
      "entropy": 5.375656175613403,
      "epoch": 0.38088874039425324,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990028983387517,
      "loss": 5.3093,
      "mean_token_accuracy": 0.19318119585514068,
      "num_tokens": 8387918.0,
      "step": 4560
    },
    {
      "entropy": 5.548627471923828,
      "epoch": 0.3813063815569663,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990000968148068,
      "loss": 5.4563,
      "mean_token_accuracy": 0.1697971135377884,
      "num_tokens": 8397755.0,
      "step": 4565
    },
    {
      "entropy": 5.497386264801025,
      "epoch": 0.38172402271967926,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989972913694707,
      "loss": 5.4646,
      "mean_token_accuracy": 0.17076701074838638,
      "num_tokens": 8408229.0,
      "step": 4570
    },
    {
      "entropy": 5.427838087081909,
      "epoch": 0.38214166388239224,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004989944820027923,
      "loss": 5.4039,
      "mean_token_accuracy": 0.17245035618543625,
      "num_tokens": 8417039.0,
      "step": 4575
    },
    {
      "entropy": 5.42997784614563,
      "epoch": 0.38255930504510527,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989916687148209,
      "loss": 5.2803,
      "mean_token_accuracy": 0.1772854745388031,
      "num_tokens": 8426656.0,
      "step": 4580
    },
    {
      "entropy": 5.5097325325012205,
      "epoch": 0.38297694620781825,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989888515056057,
      "loss": 5.3533,
      "mean_token_accuracy": 0.1768176957964897,
      "num_tokens": 8435254.0,
      "step": 4585
    },
    {
      "entropy": 5.303733777999878,
      "epoch": 0.3833945873705312,
      "grad_norm": 1.140625,
      "learning_rate": 0.000498986030375196,
      "loss": 5.2728,
      "mean_token_accuracy": 0.1777205228805542,
      "num_tokens": 8444594.0,
      "step": 4590
    },
    {
      "entropy": 5.355650806427002,
      "epoch": 0.38381222853324426,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989832053236413,
      "loss": 5.2771,
      "mean_token_accuracy": 0.18417982757091522,
      "num_tokens": 8453737.0,
      "step": 4595
    },
    {
      "entropy": 5.469804000854492,
      "epoch": 0.38422986969595724,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000498980376350991,
      "loss": 5.3665,
      "mean_token_accuracy": 0.17435781806707382,
      "num_tokens": 8463033.0,
      "step": 4600
    },
    {
      "entropy": 5.415114831924439,
      "epoch": 0.3846475108586702,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989775434572945,
      "loss": 5.3722,
      "mean_token_accuracy": 0.1721029609441757,
      "num_tokens": 8472875.0,
      "step": 4605
    },
    {
      "entropy": 5.531055068969726,
      "epoch": 0.38506515202138325,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989747066426015,
      "loss": 5.4918,
      "mean_token_accuracy": 0.16320496201515197,
      "num_tokens": 8482281.0,
      "step": 4610
    },
    {
      "entropy": 5.556277322769165,
      "epoch": 0.38548279318409623,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989718659069618,
      "loss": 5.4176,
      "mean_token_accuracy": 0.1776779368519783,
      "num_tokens": 8490871.0,
      "step": 4615
    },
    {
      "entropy": 5.384867572784424,
      "epoch": 0.3859004343468092,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000498969021250425,
      "loss": 5.3735,
      "mean_token_accuracy": 0.17717715203762055,
      "num_tokens": 8499765.0,
      "step": 4620
    },
    {
      "entropy": 5.410521936416626,
      "epoch": 0.38631807550952224,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004989661726730407,
      "loss": 5.3426,
      "mean_token_accuracy": 0.18398708254098892,
      "num_tokens": 8508163.0,
      "step": 4625
    },
    {
      "entropy": 5.484644365310669,
      "epoch": 0.3867357166722352,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989633201748591,
      "loss": 5.4098,
      "mean_token_accuracy": 0.17248052060604097,
      "num_tokens": 8517791.0,
      "step": 4630
    },
    {
      "entropy": 5.41528902053833,
      "epoch": 0.3871533578349482,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989604637559299,
      "loss": 5.3477,
      "mean_token_accuracy": 0.1795007422566414,
      "num_tokens": 8526051.0,
      "step": 4635
    },
    {
      "entropy": 5.43937463760376,
      "epoch": 0.38757099899766123,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004989576034163033,
      "loss": 5.3687,
      "mean_token_accuracy": 0.17532350718975068,
      "num_tokens": 8536102.0,
      "step": 4640
    },
    {
      "entropy": 5.432383728027344,
      "epoch": 0.3879886401603742,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989547391560293,
      "loss": 5.3781,
      "mean_token_accuracy": 0.16846679151058197,
      "num_tokens": 8544566.0,
      "step": 4645
    },
    {
      "entropy": 5.533143711090088,
      "epoch": 0.3884062813230872,
      "grad_norm": 1.09375,
      "learning_rate": 0.000498951870975158,
      "loss": 5.576,
      "mean_token_accuracy": 0.1642456069588661,
      "num_tokens": 8555397.0,
      "step": 4650
    },
    {
      "entropy": 5.570744943618775,
      "epoch": 0.3888239224858002,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989489988737395,
      "loss": 5.3884,
      "mean_token_accuracy": 0.1791091486811638,
      "num_tokens": 8564184.0,
      "step": 4655
    },
    {
      "entropy": 5.567000961303711,
      "epoch": 0.3892415636485132,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004989461228518243,
      "loss": 5.4094,
      "mean_token_accuracy": 0.1724994361400604,
      "num_tokens": 8572967.0,
      "step": 4660
    },
    {
      "entropy": 5.394959306716919,
      "epoch": 0.3896592048112262,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004989432429094627,
      "loss": 5.3859,
      "mean_token_accuracy": 0.17433421611785888,
      "num_tokens": 8581946.0,
      "step": 4665
    },
    {
      "entropy": 5.366177749633789,
      "epoch": 0.3900768459739392,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004989403590467049,
      "loss": 5.2841,
      "mean_token_accuracy": 0.18927239775657653,
      "num_tokens": 8590528.0,
      "step": 4670
    },
    {
      "entropy": 5.448218202590942,
      "epoch": 0.3904944871366522,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989374712636017,
      "loss": 5.2817,
      "mean_token_accuracy": 0.17978343069553376,
      "num_tokens": 8600399.0,
      "step": 4675
    },
    {
      "entropy": 5.4937997341156,
      "epoch": 0.39091212829936517,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004989345795602036,
      "loss": 5.3416,
      "mean_token_accuracy": 0.17784634977579117,
      "num_tokens": 8609620.0,
      "step": 4680
    },
    {
      "entropy": 5.337210035324096,
      "epoch": 0.3913297694620782,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004989316839365609,
      "loss": 5.3764,
      "mean_token_accuracy": 0.18280318081378938,
      "num_tokens": 8618077.0,
      "step": 4685
    },
    {
      "entropy": 5.354756498336792,
      "epoch": 0.3917474106247912,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989287843927246,
      "loss": 5.3177,
      "mean_token_accuracy": 0.18373266011476516,
      "num_tokens": 8626672.0,
      "step": 4690
    },
    {
      "entropy": 5.448664379119873,
      "epoch": 0.39216505178750416,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989258809287455,
      "loss": 5.3058,
      "mean_token_accuracy": 0.1821853294968605,
      "num_tokens": 8635997.0,
      "step": 4695
    },
    {
      "entropy": 5.444197797775269,
      "epoch": 0.3925826929502172,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989229735446743,
      "loss": 5.3413,
      "mean_token_accuracy": 0.16682087033987045,
      "num_tokens": 8645296.0,
      "step": 4700
    },
    {
      "entropy": 5.477824544906616,
      "epoch": 0.39300033411293017,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989200622405619,
      "loss": 5.4731,
      "mean_token_accuracy": 0.17398236244916915,
      "num_tokens": 8654772.0,
      "step": 4705
    },
    {
      "entropy": 5.435440587997436,
      "epoch": 0.39341797527564315,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989171470164592,
      "loss": 5.3319,
      "mean_token_accuracy": 0.17901189625263214,
      "num_tokens": 8664279.0,
      "step": 4710
    },
    {
      "entropy": 5.4493427753448485,
      "epoch": 0.3938356164383562,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989142278724173,
      "loss": 5.4138,
      "mean_token_accuracy": 0.1713800996541977,
      "num_tokens": 8673405.0,
      "step": 4715
    },
    {
      "entropy": 5.4495786190032955,
      "epoch": 0.39425325760106916,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989113048084874,
      "loss": 5.3752,
      "mean_token_accuracy": 0.17766857296228408,
      "num_tokens": 8682254.0,
      "step": 4720
    },
    {
      "entropy": 5.473398065567016,
      "epoch": 0.39467089876378214,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989083778247206,
      "loss": 5.2786,
      "mean_token_accuracy": 0.18520117402076722,
      "num_tokens": 8690751.0,
      "step": 4725
    },
    {
      "entropy": 5.514026927947998,
      "epoch": 0.3950885399264952,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498905446921168,
      "loss": 5.4732,
      "mean_token_accuracy": 0.16697628647089005,
      "num_tokens": 8700485.0,
      "step": 4730
    },
    {
      "entropy": 5.420544815063477,
      "epoch": 0.39550618108920815,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989025120978812,
      "loss": 5.3883,
      "mean_token_accuracy": 0.16980446875095367,
      "num_tokens": 8709650.0,
      "step": 4735
    },
    {
      "entropy": 5.374988794326782,
      "epoch": 0.39592382225192113,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004988995733549113,
      "loss": 5.3366,
      "mean_token_accuracy": 0.17936188727617264,
      "num_tokens": 8719090.0,
      "step": 4740
    },
    {
      "entropy": 5.456218910217285,
      "epoch": 0.39634146341463417,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988966306923099,
      "loss": 5.4868,
      "mean_token_accuracy": 0.17431129068136214,
      "num_tokens": 8728563.0,
      "step": 4745
    },
    {
      "entropy": 5.582469940185547,
      "epoch": 0.39675910457734714,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004988936841101284,
      "loss": 5.3821,
      "mean_token_accuracy": 0.17387064397335053,
      "num_tokens": 8737587.0,
      "step": 4750
    },
    {
      "entropy": 5.496310997009277,
      "epoch": 0.3971767457400601,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988907336084186,
      "loss": 5.4406,
      "mean_token_accuracy": 0.17855045199394226,
      "num_tokens": 8747377.0,
      "step": 4755
    },
    {
      "entropy": 5.516774272918701,
      "epoch": 0.39759438690277316,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988877791872319,
      "loss": 5.4181,
      "mean_token_accuracy": 0.17951357066631318,
      "num_tokens": 8756471.0,
      "step": 4760
    },
    {
      "entropy": 5.376832675933838,
      "epoch": 0.39801202806548613,
      "grad_norm": 1.125,
      "learning_rate": 0.0004988848208466202,
      "loss": 5.4289,
      "mean_token_accuracy": 0.1693958595395088,
      "num_tokens": 8765445.0,
      "step": 4765
    },
    {
      "entropy": 5.581160211563111,
      "epoch": 0.3984296692281991,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988818585866351,
      "loss": 5.4503,
      "mean_token_accuracy": 0.1728968620300293,
      "num_tokens": 8774946.0,
      "step": 4770
    },
    {
      "entropy": 5.478467082977295,
      "epoch": 0.39884731039091215,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004988788924073288,
      "loss": 5.3677,
      "mean_token_accuracy": 0.17615725547075273,
      "num_tokens": 8783652.0,
      "step": 4775
    },
    {
      "entropy": 5.364590787887574,
      "epoch": 0.3992649515536251,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004988759223087528,
      "loss": 5.2824,
      "mean_token_accuracy": 0.1788276180624962,
      "num_tokens": 8792633.0,
      "step": 4780
    },
    {
      "entropy": 5.34464955329895,
      "epoch": 0.3996825927163381,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988729482909595,
      "loss": 5.2876,
      "mean_token_accuracy": 0.1825534701347351,
      "num_tokens": 8802376.0,
      "step": 4785
    },
    {
      "entropy": 5.389295625686645,
      "epoch": 0.40010023387905114,
      "grad_norm": 1.125,
      "learning_rate": 0.0004988699703540006,
      "loss": 5.3487,
      "mean_token_accuracy": 0.18201276659965515,
      "num_tokens": 8812122.0,
      "step": 4790
    },
    {
      "entropy": 5.443384647369385,
      "epoch": 0.4005178750417641,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988669884979284,
      "loss": 5.2828,
      "mean_token_accuracy": 0.17886264026165008,
      "num_tokens": 8821111.0,
      "step": 4795
    },
    {
      "entropy": 5.4228722095489506,
      "epoch": 0.4009355162044771,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004988640027227951,
      "loss": 5.3983,
      "mean_token_accuracy": 0.17484676614403724,
      "num_tokens": 8830354.0,
      "step": 4800
    },
    {
      "entropy": 5.367304134368896,
      "epoch": 0.4013531573671901,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498861013028653,
      "loss": 5.2865,
      "mean_token_accuracy": 0.18212662786245346,
      "num_tokens": 8839500.0,
      "step": 4805
    },
    {
      "entropy": 5.448452091217041,
      "epoch": 0.4017707985299031,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988580194155544,
      "loss": 5.3719,
      "mean_token_accuracy": 0.174194598197937,
      "num_tokens": 8849632.0,
      "step": 4810
    },
    {
      "entropy": 5.622716903686523,
      "epoch": 0.4021884396926161,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004988550218835517,
      "loss": 5.4885,
      "mean_token_accuracy": 0.1704885631799698,
      "num_tokens": 8858383.0,
      "step": 4815
    },
    {
      "entropy": 5.349063301086426,
      "epoch": 0.4026060808553291,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004988520204326975,
      "loss": 5.2539,
      "mean_token_accuracy": 0.18768793791532518,
      "num_tokens": 8866755.0,
      "step": 4820
    },
    {
      "entropy": 5.398096561431885,
      "epoch": 0.4030237220180421,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988490150630441,
      "loss": 5.4592,
      "mean_token_accuracy": 0.1728242516517639,
      "num_tokens": 8876896.0,
      "step": 4825
    },
    {
      "entropy": 5.594376134872436,
      "epoch": 0.4034413631807551,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004988460057746443,
      "loss": 5.3984,
      "mean_token_accuracy": 0.17692140489816666,
      "num_tokens": 8886029.0,
      "step": 4830
    },
    {
      "entropy": 5.4225725650787355,
      "epoch": 0.4038590043434681,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988429925675507,
      "loss": 5.3057,
      "mean_token_accuracy": 0.17887427508831025,
      "num_tokens": 8895396.0,
      "step": 4835
    },
    {
      "entropy": 5.404714584350586,
      "epoch": 0.4042766455061811,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004988399754418161,
      "loss": 5.361,
      "mean_token_accuracy": 0.18620169758796692,
      "num_tokens": 8904125.0,
      "step": 4840
    },
    {
      "entropy": 5.404704236984253,
      "epoch": 0.40469428666889407,
      "grad_norm": 1.125,
      "learning_rate": 0.0004988369543974932,
      "loss": 5.4318,
      "mean_token_accuracy": 0.16959863007068635,
      "num_tokens": 8913011.0,
      "step": 4845
    },
    {
      "entropy": 5.402275943756104,
      "epoch": 0.4051119278316071,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004988339294346351,
      "loss": 5.3426,
      "mean_token_accuracy": 0.1808636337518692,
      "num_tokens": 8921596.0,
      "step": 4850
    },
    {
      "entropy": 5.448325061798096,
      "epoch": 0.4055295689943201,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004988309005532946,
      "loss": 5.3277,
      "mean_token_accuracy": 0.17477848529815673,
      "num_tokens": 8931267.0,
      "step": 4855
    },
    {
      "entropy": 5.298576974868775,
      "epoch": 0.40594721015703306,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988278677535247,
      "loss": 5.2007,
      "mean_token_accuracy": 0.18315067142248154,
      "num_tokens": 8940082.0,
      "step": 4860
    },
    {
      "entropy": 5.485443592071533,
      "epoch": 0.4063648513197461,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004988248310353788,
      "loss": 5.4679,
      "mean_token_accuracy": 0.17353947907686235,
      "num_tokens": 8949838.0,
      "step": 4865
    },
    {
      "entropy": 5.467548894882202,
      "epoch": 0.40678249248245907,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004988217903989095,
      "loss": 5.3867,
      "mean_token_accuracy": 0.17952940315008165,
      "num_tokens": 8958510.0,
      "step": 4870
    },
    {
      "entropy": 5.433662223815918,
      "epoch": 0.40720013364517205,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988187458441704,
      "loss": 5.382,
      "mean_token_accuracy": 0.17126011699438096,
      "num_tokens": 8968372.0,
      "step": 4875
    },
    {
      "entropy": 5.3494861125946045,
      "epoch": 0.4076177748078851,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004988156973712149,
      "loss": 5.2433,
      "mean_token_accuracy": 0.18010528683662413,
      "num_tokens": 8977121.0,
      "step": 4880
    },
    {
      "entropy": 5.4491739749908445,
      "epoch": 0.40803541597059806,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004988126449800961,
      "loss": 5.3756,
      "mean_token_accuracy": 0.1782987281680107,
      "num_tokens": 8986643.0,
      "step": 4885
    },
    {
      "entropy": 5.380730676651001,
      "epoch": 0.40845305713331104,
      "grad_norm": 1.125,
      "learning_rate": 0.0004988095886708676,
      "loss": 5.2101,
      "mean_token_accuracy": 0.18655198961496353,
      "num_tokens": 8994925.0,
      "step": 4890
    },
    {
      "entropy": 5.299397134780884,
      "epoch": 0.40887069829602407,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004988065284435828,
      "loss": 5.3264,
      "mean_token_accuracy": 0.1842572957277298,
      "num_tokens": 9003844.0,
      "step": 4895
    },
    {
      "entropy": 5.466626310348511,
      "epoch": 0.40928833945873705,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004988034642982954,
      "loss": 5.3381,
      "mean_token_accuracy": 0.17607323825359344,
      "num_tokens": 9013351.0,
      "step": 4900
    },
    {
      "entropy": 5.445525598526001,
      "epoch": 0.40970598062145,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988003962350589,
      "loss": 5.4327,
      "mean_token_accuracy": 0.17783276438713075,
      "num_tokens": 9023488.0,
      "step": 4905
    },
    {
      "entropy": 5.423844528198242,
      "epoch": 0.41012362178416306,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498797324253927,
      "loss": 5.3436,
      "mean_token_accuracy": 0.1794862851500511,
      "num_tokens": 9033018.0,
      "step": 4910
    },
    {
      "entropy": 5.3908610343933105,
      "epoch": 0.41054126294687604,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987942483549537,
      "loss": 5.3205,
      "mean_token_accuracy": 0.18037871420383453,
      "num_tokens": 9041572.0,
      "step": 4915
    },
    {
      "entropy": 5.5957917213439945,
      "epoch": 0.410958904109589,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987911685381926,
      "loss": 5.5562,
      "mean_token_accuracy": 0.16397008448839187,
      "num_tokens": 9051505.0,
      "step": 4920
    },
    {
      "entropy": 5.426666593551635,
      "epoch": 0.41137654527230205,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987880848036977,
      "loss": 5.3448,
      "mean_token_accuracy": 0.17821265608072281,
      "num_tokens": 9061168.0,
      "step": 4925
    },
    {
      "entropy": 5.2480797290802,
      "epoch": 0.41179418643501503,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987849971515232,
      "loss": 5.2327,
      "mean_token_accuracy": 0.17862533032894135,
      "num_tokens": 9069776.0,
      "step": 4930
    },
    {
      "entropy": 5.4414534091949465,
      "epoch": 0.412211827597728,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004987819055817227,
      "loss": 5.2979,
      "mean_token_accuracy": 0.1817397803068161,
      "num_tokens": 9078995.0,
      "step": 4935
    },
    {
      "entropy": 5.348909711837768,
      "epoch": 0.41262946876044104,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004987788100943507,
      "loss": 5.3274,
      "mean_token_accuracy": 0.18326059132814407,
      "num_tokens": 9088291.0,
      "step": 4940
    },
    {
      "entropy": 5.423943901062012,
      "epoch": 0.413047109923154,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004987757106894613,
      "loss": 5.257,
      "mean_token_accuracy": 0.17726942151784897,
      "num_tokens": 9097013.0,
      "step": 4945
    },
    {
      "entropy": 5.427419185638428,
      "epoch": 0.413464751085867,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004987726073671086,
      "loss": 5.3776,
      "mean_token_accuracy": 0.17360134273767472,
      "num_tokens": 9106909.0,
      "step": 4950
    },
    {
      "entropy": 5.4259748458862305,
      "epoch": 0.41388239224858003,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987695001273471,
      "loss": 5.3981,
      "mean_token_accuracy": 0.1686091035604477,
      "num_tokens": 9116056.0,
      "step": 4955
    },
    {
      "entropy": 5.408644199371338,
      "epoch": 0.414300033411293,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004987663889702313,
      "loss": 5.3426,
      "mean_token_accuracy": 0.1770440310239792,
      "num_tokens": 9124968.0,
      "step": 4960
    },
    {
      "entropy": 5.380485582351684,
      "epoch": 0.414717674574006,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987632738958155,
      "loss": 5.2845,
      "mean_token_accuracy": 0.17902038991451263,
      "num_tokens": 9134423.0,
      "step": 4965
    },
    {
      "entropy": 5.514896488189697,
      "epoch": 0.415135315736719,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987601549041541,
      "loss": 5.4178,
      "mean_token_accuracy": 0.17550164312124253,
      "num_tokens": 9144179.0,
      "step": 4970
    },
    {
      "entropy": 5.399398517608643,
      "epoch": 0.415552956899432,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987570319953018,
      "loss": 5.4532,
      "mean_token_accuracy": 0.16906047165393828,
      "num_tokens": 9154045.0,
      "step": 4975
    },
    {
      "entropy": 5.315214443206787,
      "epoch": 0.415970598062145,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987539051693134,
      "loss": 5.2424,
      "mean_token_accuracy": 0.1848849982023239,
      "num_tokens": 9162922.0,
      "step": 4980
    },
    {
      "entropy": 5.340002155303955,
      "epoch": 0.416388239224858,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004987507744262436,
      "loss": 5.2502,
      "mean_token_accuracy": 0.18313726782798767,
      "num_tokens": 9172327.0,
      "step": 4985
    },
    {
      "entropy": 5.3172860622406,
      "epoch": 0.416805880387571,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987476397661472,
      "loss": 5.2397,
      "mean_token_accuracy": 0.18388236463069915,
      "num_tokens": 9180830.0,
      "step": 4990
    },
    {
      "entropy": 5.461140155792236,
      "epoch": 0.41722352155028397,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004987445011890789,
      "loss": 5.4757,
      "mean_token_accuracy": 0.17659415304660797,
      "num_tokens": 9190787.0,
      "step": 4995
    },
    {
      "entropy": 5.486403846740723,
      "epoch": 0.417641162712997,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498741358695094,
      "loss": 5.3959,
      "mean_token_accuracy": 0.17391360849142073,
      "num_tokens": 9199554.0,
      "step": 5000
    },
    {
      "entropy": 5.36230354309082,
      "epoch": 0.41805880387571,
      "grad_norm": 1.03125,
      "learning_rate": 0.000498738212284247,
      "loss": 5.2918,
      "mean_token_accuracy": 0.17479039132595062,
      "num_tokens": 9209309.0,
      "step": 5005
    },
    {
      "entropy": 5.442480516433716,
      "epoch": 0.41847644503842296,
      "grad_norm": 1.0,
      "learning_rate": 0.0004987350619565934,
      "loss": 5.3546,
      "mean_token_accuracy": 0.1775646835565567,
      "num_tokens": 9218529.0,
      "step": 5010
    },
    {
      "entropy": 5.397161149978638,
      "epoch": 0.418894086201136,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987319077121884,
      "loss": 5.2956,
      "mean_token_accuracy": 0.17779555320739746,
      "num_tokens": 9226924.0,
      "step": 5015
    },
    {
      "entropy": 5.345029497146607,
      "epoch": 0.419311727363849,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004987287495510869,
      "loss": 5.3549,
      "mean_token_accuracy": 0.17008619159460067,
      "num_tokens": 9237602.0,
      "step": 5020
    },
    {
      "entropy": 5.487488460540772,
      "epoch": 0.41972936852656195,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004987255874733444,
      "loss": 5.3437,
      "mean_token_accuracy": 0.17842262536287307,
      "num_tokens": 9246209.0,
      "step": 5025
    },
    {
      "entropy": 5.351472520828247,
      "epoch": 0.420147009689275,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004987224214790161,
      "loss": 5.3118,
      "mean_token_accuracy": 0.17697940915822982,
      "num_tokens": 9255464.0,
      "step": 5030
    },
    {
      "entropy": 5.360794496536255,
      "epoch": 0.42056465085198796,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987192515681575,
      "loss": 5.3421,
      "mean_token_accuracy": 0.17424386441707612,
      "num_tokens": 9265520.0,
      "step": 5035
    },
    {
      "entropy": 5.464559745788574,
      "epoch": 0.42098229201470094,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987160777408243,
      "loss": 5.3214,
      "mean_token_accuracy": 0.17781623601913452,
      "num_tokens": 9274916.0,
      "step": 5040
    },
    {
      "entropy": 5.3971545696258545,
      "epoch": 0.421399933177414,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987128999970717,
      "loss": 5.3956,
      "mean_token_accuracy": 0.16885341703891754,
      "num_tokens": 9284684.0,
      "step": 5045
    },
    {
      "entropy": 5.437297439575195,
      "epoch": 0.42181757434012696,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987097183369555,
      "loss": 5.4216,
      "mean_token_accuracy": 0.17992314845323562,
      "num_tokens": 9294020.0,
      "step": 5050
    },
    {
      "entropy": 5.44114055633545,
      "epoch": 0.42223521550283993,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004987065327605315,
      "loss": 5.4109,
      "mean_token_accuracy": 0.17765092402696608,
      "num_tokens": 9303730.0,
      "step": 5055
    },
    {
      "entropy": 5.491128301620483,
      "epoch": 0.42265285666555297,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004987033432678553,
      "loss": 5.3023,
      "mean_token_accuracy": 0.18014814406633378,
      "num_tokens": 9312851.0,
      "step": 5060
    },
    {
      "entropy": 5.286890840530395,
      "epoch": 0.42307049782826595,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004987001498589828,
      "loss": 5.1988,
      "mean_token_accuracy": 0.18571052104234695,
      "num_tokens": 9321143.0,
      "step": 5065
    },
    {
      "entropy": 5.420368099212647,
      "epoch": 0.4234881389909789,
      "grad_norm": 1.140625,
      "learning_rate": 0.00049869695253397,
      "loss": 5.3109,
      "mean_token_accuracy": 0.18008437752723694,
      "num_tokens": 9329843.0,
      "step": 5070
    },
    {
      "entropy": 5.440558433532715,
      "epoch": 0.42390578015369196,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986937512928727,
      "loss": 5.4511,
      "mean_token_accuracy": 0.1702857494354248,
      "num_tokens": 9339549.0,
      "step": 5075
    },
    {
      "entropy": 5.469025373458862,
      "epoch": 0.42432342131640494,
      "grad_norm": 1.0625,
      "learning_rate": 0.000498690546135747,
      "loss": 5.3624,
      "mean_token_accuracy": 0.17415655851364137,
      "num_tokens": 9348220.0,
      "step": 5080
    },
    {
      "entropy": 5.442413854598999,
      "epoch": 0.4247410624791179,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004986873370626491,
      "loss": 5.3316,
      "mean_token_accuracy": 0.17110524326562881,
      "num_tokens": 9357440.0,
      "step": 5085
    },
    {
      "entropy": 5.430461072921753,
      "epoch": 0.42515870364183095,
      "grad_norm": 1.125,
      "learning_rate": 0.000498684124073635,
      "loss": 5.3483,
      "mean_token_accuracy": 0.1753034397959709,
      "num_tokens": 9366457.0,
      "step": 5090
    },
    {
      "entropy": 5.400099229812622,
      "epoch": 0.4255763448045439,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004986809071687612,
      "loss": 5.4408,
      "mean_token_accuracy": 0.17686941027641295,
      "num_tokens": 9375432.0,
      "step": 5095
    },
    {
      "entropy": 5.373173713684082,
      "epoch": 0.4259939859672569,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004986776863480838,
      "loss": 5.2837,
      "mean_token_accuracy": 0.1831916391849518,
      "num_tokens": 9383353.0,
      "step": 5100
    },
    {
      "entropy": 5.457636070251465,
      "epoch": 0.42641162712996994,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986744616116593,
      "loss": 5.3324,
      "mean_token_accuracy": 0.18044039607048035,
      "num_tokens": 9393012.0,
      "step": 5105
    },
    {
      "entropy": 5.423114776611328,
      "epoch": 0.4268292682926829,
      "grad_norm": 1.125,
      "learning_rate": 0.000498671232959544,
      "loss": 5.355,
      "mean_token_accuracy": 0.18421694189310073,
      "num_tokens": 9401394.0,
      "step": 5110
    },
    {
      "entropy": 5.394059324264527,
      "epoch": 0.42724690945539595,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986680003917947,
      "loss": 5.289,
      "mean_token_accuracy": 0.1817804366350174,
      "num_tokens": 9410310.0,
      "step": 5115
    },
    {
      "entropy": 5.39257287979126,
      "epoch": 0.42766455061810893,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986647639084677,
      "loss": 5.2572,
      "mean_token_accuracy": 0.18167548775672912,
      "num_tokens": 9418625.0,
      "step": 5120
    },
    {
      "entropy": 5.394883108139038,
      "epoch": 0.4280821917808219,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00049866152350962,
      "loss": 5.3929,
      "mean_token_accuracy": 0.1777561068534851,
      "num_tokens": 9428722.0,
      "step": 5125
    },
    {
      "entropy": 5.47434630393982,
      "epoch": 0.42849983294353494,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986582791953078,
      "loss": 5.3538,
      "mean_token_accuracy": 0.17843445241451264,
      "num_tokens": 9438447.0,
      "step": 5130
    },
    {
      "entropy": 5.42354383468628,
      "epoch": 0.4289174741062479,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004986550309655884,
      "loss": 5.3001,
      "mean_token_accuracy": 0.1818290963768959,
      "num_tokens": 9447457.0,
      "step": 5135
    },
    {
      "entropy": 5.45627236366272,
      "epoch": 0.4293351152689609,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004986517788205184,
      "loss": 5.4273,
      "mean_token_accuracy": 0.16493106931447982,
      "num_tokens": 9456837.0,
      "step": 5140
    },
    {
      "entropy": 5.535893726348877,
      "epoch": 0.42975275643167393,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986485227601548,
      "loss": 5.4572,
      "mean_token_accuracy": 0.17140540033578872,
      "num_tokens": 9465733.0,
      "step": 5145
    },
    {
      "entropy": 5.390489768981934,
      "epoch": 0.4301703975943869,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986452627845547,
      "loss": 5.3196,
      "mean_token_accuracy": 0.17633309215307236,
      "num_tokens": 9473978.0,
      "step": 5150
    },
    {
      "entropy": 5.351325464248657,
      "epoch": 0.4305880387570999,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000498641998893775,
      "loss": 5.3764,
      "mean_token_accuracy": 0.17647748738527297,
      "num_tokens": 9482914.0,
      "step": 5155
    },
    {
      "entropy": 5.422978019714355,
      "epoch": 0.4310056799198129,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986387310878728,
      "loss": 5.3121,
      "mean_token_accuracy": 0.1877748489379883,
      "num_tokens": 9492060.0,
      "step": 5160
    },
    {
      "entropy": 5.443538570404053,
      "epoch": 0.4314233210825259,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986354593669056,
      "loss": 5.3787,
      "mean_token_accuracy": 0.17866956144571305,
      "num_tokens": 9501186.0,
      "step": 5165
    },
    {
      "entropy": 5.574661493301392,
      "epoch": 0.4318409622452389,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986321837309304,
      "loss": 5.3727,
      "mean_token_accuracy": 0.1690126046538353,
      "num_tokens": 9509727.0,
      "step": 5170
    },
    {
      "entropy": 5.42094612121582,
      "epoch": 0.4322586034079519,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004986289041800046,
      "loss": 5.3256,
      "mean_token_accuracy": 0.17340917736291886,
      "num_tokens": 9519203.0,
      "step": 5175
    },
    {
      "entropy": 5.414110565185547,
      "epoch": 0.4326762445706649,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004986256207141857,
      "loss": 5.3311,
      "mean_token_accuracy": 0.17890969663858414,
      "num_tokens": 9529277.0,
      "step": 5180
    },
    {
      "entropy": 5.462023019790649,
      "epoch": 0.43309388573337787,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986223333335311,
      "loss": 5.3062,
      "mean_token_accuracy": 0.17907025516033173,
      "num_tokens": 9538215.0,
      "step": 5185
    },
    {
      "entropy": 5.25904860496521,
      "epoch": 0.4335115268960909,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004986190420380985,
      "loss": 5.2621,
      "mean_token_accuracy": 0.18161898702383042,
      "num_tokens": 9547570.0,
      "step": 5190
    },
    {
      "entropy": 5.332035541534424,
      "epoch": 0.4339291680588039,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004986157468279453,
      "loss": 5.1566,
      "mean_token_accuracy": 0.18879634886980057,
      "num_tokens": 9556383.0,
      "step": 5195
    },
    {
      "entropy": 5.3860182762146,
      "epoch": 0.43434680922151686,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004986124477031293,
      "loss": 5.3246,
      "mean_token_accuracy": 0.17788449823856353,
      "num_tokens": 9566793.0,
      "step": 5200
    },
    {
      "entropy": 5.480335807800293,
      "epoch": 0.4347644503842299,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986091446637083,
      "loss": 5.3536,
      "mean_token_accuracy": 0.17622597217559816,
      "num_tokens": 9576079.0,
      "step": 5205
    },
    {
      "entropy": 5.371386861801147,
      "epoch": 0.4351820915469429,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049860583770974,
      "loss": 5.3293,
      "mean_token_accuracy": 0.1782429724931717,
      "num_tokens": 9586318.0,
      "step": 5210
    },
    {
      "entropy": 5.322963571548462,
      "epoch": 0.43559973270965585,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004986025268412823,
      "loss": 5.237,
      "mean_token_accuracy": 0.18290005922317504,
      "num_tokens": 9594453.0,
      "step": 5215
    },
    {
      "entropy": 5.249912071228027,
      "epoch": 0.4360173738723689,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004985992120583933,
      "loss": 5.173,
      "mean_token_accuracy": 0.19170199036598207,
      "num_tokens": 9603677.0,
      "step": 5220
    },
    {
      "entropy": 5.418201875686646,
      "epoch": 0.43643501503508186,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000498595893361131,
      "loss": 5.3607,
      "mean_token_accuracy": 0.18029106706380843,
      "num_tokens": 9612353.0,
      "step": 5225
    },
    {
      "entropy": 5.4223198890686035,
      "epoch": 0.43685265619779484,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985925707495533,
      "loss": 5.3662,
      "mean_token_accuracy": 0.18232962340116501,
      "num_tokens": 9621607.0,
      "step": 5230
    },
    {
      "entropy": 5.448773145675659,
      "epoch": 0.4372702973605079,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004985892442237186,
      "loss": 5.3406,
      "mean_token_accuracy": 0.18682206571102142,
      "num_tokens": 9630155.0,
      "step": 5235
    },
    {
      "entropy": 5.41499433517456,
      "epoch": 0.43768793852322085,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004985859137836851,
      "loss": 5.297,
      "mean_token_accuracy": 0.18273041546344757,
      "num_tokens": 9639418.0,
      "step": 5240
    },
    {
      "entropy": 5.419114255905152,
      "epoch": 0.43810557968593383,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004985825794295109,
      "loss": 5.3182,
      "mean_token_accuracy": 0.18026842623949052,
      "num_tokens": 9648030.0,
      "step": 5245
    },
    {
      "entropy": 5.3935081481933596,
      "epoch": 0.43852322084864687,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004985792411612546,
      "loss": 5.2965,
      "mean_token_accuracy": 0.18045262098312378,
      "num_tokens": 9657746.0,
      "step": 5250
    },
    {
      "entropy": 5.3888349533081055,
      "epoch": 0.43894086201135984,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004985758989789746,
      "loss": 5.3331,
      "mean_token_accuracy": 0.18024140894412993,
      "num_tokens": 9667840.0,
      "step": 5255
    },
    {
      "entropy": 5.423379707336426,
      "epoch": 0.4393585031740728,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985725528827293,
      "loss": 5.3848,
      "mean_token_accuracy": 0.1786363258957863,
      "num_tokens": 9677595.0,
      "step": 5260
    },
    {
      "entropy": 5.356011819839478,
      "epoch": 0.43977614433678586,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985692028725774,
      "loss": 5.2595,
      "mean_token_accuracy": 0.1844140723347664,
      "num_tokens": 9685515.0,
      "step": 5265
    },
    {
      "entropy": 5.284841680526734,
      "epoch": 0.44019378549949884,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004985658489485774,
      "loss": 5.2567,
      "mean_token_accuracy": 0.17957810014486314,
      "num_tokens": 9694238.0,
      "step": 5270
    },
    {
      "entropy": 5.479930448532104,
      "epoch": 0.4406114266622118,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985624911107882,
      "loss": 5.3892,
      "mean_token_accuracy": 0.17708906382322312,
      "num_tokens": 9703388.0,
      "step": 5275
    },
    {
      "entropy": 5.435586261749267,
      "epoch": 0.44102906782492485,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004985591293592685,
      "loss": 5.3038,
      "mean_token_accuracy": 0.1767055243253708,
      "num_tokens": 9713085.0,
      "step": 5280
    },
    {
      "entropy": 5.474288082122802,
      "epoch": 0.4414467089876378,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004985557636940771,
      "loss": 5.3355,
      "mean_token_accuracy": 0.1808098629117012,
      "num_tokens": 9721733.0,
      "step": 5285
    },
    {
      "entropy": 5.453846549987793,
      "epoch": 0.4418643501503508,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000498552394115273,
      "loss": 5.4083,
      "mean_token_accuracy": 0.1813710704445839,
      "num_tokens": 9731478.0,
      "step": 5290
    },
    {
      "entropy": 5.362984895706177,
      "epoch": 0.44228199131306384,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004985490206229151,
      "loss": 5.34,
      "mean_token_accuracy": 0.17717154771089555,
      "num_tokens": 9742058.0,
      "step": 5295
    },
    {
      "entropy": 5.367059230804443,
      "epoch": 0.4426996324757768,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004985456432170626,
      "loss": 5.2903,
      "mean_token_accuracy": 0.1790708914399147,
      "num_tokens": 9751034.0,
      "step": 5300
    },
    {
      "entropy": 5.445749044418335,
      "epoch": 0.4431172736384898,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004985422618977744,
      "loss": 5.377,
      "mean_token_accuracy": 0.1745736002922058,
      "num_tokens": 9760692.0,
      "step": 5305
    },
    {
      "entropy": 5.384193801879883,
      "epoch": 0.44353491480120283,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00049853887666511,
      "loss": 5.2298,
      "mean_token_accuracy": 0.1801637202501297,
      "num_tokens": 9769759.0,
      "step": 5310
    },
    {
      "entropy": 5.279079961776733,
      "epoch": 0.4439525559639158,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004985354875191283,
      "loss": 5.2829,
      "mean_token_accuracy": 0.18109699487686157,
      "num_tokens": 9778597.0,
      "step": 5315
    },
    {
      "entropy": 5.284571599960327,
      "epoch": 0.4443701971266288,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004985320944598889,
      "loss": 5.197,
      "mean_token_accuracy": 0.19418197721242905,
      "num_tokens": 9787259.0,
      "step": 5320
    },
    {
      "entropy": 5.396327638626099,
      "epoch": 0.4447878382893418,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004985286974874512,
      "loss": 5.3906,
      "mean_token_accuracy": 0.17595013231039047,
      "num_tokens": 9796297.0,
      "step": 5325
    },
    {
      "entropy": 5.453844833374023,
      "epoch": 0.4452054794520548,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004985252966018745,
      "loss": 5.2087,
      "mean_token_accuracy": 0.18681780248880386,
      "num_tokens": 9805217.0,
      "step": 5330
    },
    {
      "entropy": 5.399526929855346,
      "epoch": 0.4456231206147678,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004985218918032184,
      "loss": 5.356,
      "mean_token_accuracy": 0.17481513768434526,
      "num_tokens": 9814788.0,
      "step": 5335
    },
    {
      "entropy": 5.303859996795654,
      "epoch": 0.4460407617774808,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004985184830915426,
      "loss": 5.222,
      "mean_token_accuracy": 0.17701069712638856,
      "num_tokens": 9823238.0,
      "step": 5340
    },
    {
      "entropy": 5.353955793380737,
      "epoch": 0.4464584029401938,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985150704669067,
      "loss": 5.3696,
      "mean_token_accuracy": 0.17349862605333327,
      "num_tokens": 9832619.0,
      "step": 5345
    },
    {
      "entropy": 5.402960634231567,
      "epoch": 0.44687604410290677,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004985116539293704,
      "loss": 5.2578,
      "mean_token_accuracy": 0.1838441640138626,
      "num_tokens": 9841505.0,
      "step": 5350
    },
    {
      "entropy": 5.365970706939697,
      "epoch": 0.4472936852656198,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985082334789936,
      "loss": 5.3379,
      "mean_token_accuracy": 0.17293965220451354,
      "num_tokens": 9851062.0,
      "step": 5355
    },
    {
      "entropy": 5.319137334823608,
      "epoch": 0.4477113264283328,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004985048091158361,
      "loss": 5.1754,
      "mean_token_accuracy": 0.1796301320195198,
      "num_tokens": 9860080.0,
      "step": 5360
    },
    {
      "entropy": 5.39782829284668,
      "epoch": 0.44812896759104576,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004985013808399578,
      "loss": 5.3652,
      "mean_token_accuracy": 0.17141228467226027,
      "num_tokens": 9870968.0,
      "step": 5365
    },
    {
      "entropy": 5.35687894821167,
      "epoch": 0.4485466087537588,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004984979486514189,
      "loss": 5.2741,
      "mean_token_accuracy": 0.19018529802560807,
      "num_tokens": 9879243.0,
      "step": 5370
    },
    {
      "entropy": 5.345486211776733,
      "epoch": 0.44896424991647177,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984945125502795,
      "loss": 5.2855,
      "mean_token_accuracy": 0.18337069302797318,
      "num_tokens": 9889203.0,
      "step": 5375
    },
    {
      "entropy": 5.429203176498413,
      "epoch": 0.44938189107918475,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984910725365994,
      "loss": 5.3479,
      "mean_token_accuracy": 0.18169282972812653,
      "num_tokens": 9898989.0,
      "step": 5380
    },
    {
      "entropy": 5.344571208953857,
      "epoch": 0.4497995322418978,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004984876286104392,
      "loss": 5.2777,
      "mean_token_accuracy": 0.1815692037343979,
      "num_tokens": 9908030.0,
      "step": 5385
    },
    {
      "entropy": 5.496398973464966,
      "epoch": 0.45021717340461076,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004984841807718591,
      "loss": 5.519,
      "mean_token_accuracy": 0.17015313506126403,
      "num_tokens": 9917331.0,
      "step": 5390
    },
    {
      "entropy": 5.472272443771362,
      "epoch": 0.45063481456732374,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004984807290209193,
      "loss": 5.2568,
      "mean_token_accuracy": 0.1850770518183708,
      "num_tokens": 9925418.0,
      "step": 5395
    },
    {
      "entropy": 5.437618064880371,
      "epoch": 0.45105245573003677,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004984772733576803,
      "loss": 5.3323,
      "mean_token_accuracy": 0.17590898126363755,
      "num_tokens": 9934525.0,
      "step": 5400
    },
    {
      "entropy": 5.3954144477844235,
      "epoch": 0.45147009689274975,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004984738137822028,
      "loss": 5.384,
      "mean_token_accuracy": 0.17028524577617646,
      "num_tokens": 9944087.0,
      "step": 5405
    },
    {
      "entropy": 5.46848349571228,
      "epoch": 0.45188773805546273,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000498470350294547,
      "loss": 5.2981,
      "mean_token_accuracy": 0.18230461627244948,
      "num_tokens": 9953005.0,
      "step": 5410
    },
    {
      "entropy": 5.417088222503662,
      "epoch": 0.45230537921817576,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004984668828947739,
      "loss": 5.2713,
      "mean_token_accuracy": 0.17799917757511138,
      "num_tokens": 9961963.0,
      "step": 5415
    },
    {
      "entropy": 5.3552844524383545,
      "epoch": 0.45272302038088874,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004984634115829439,
      "loss": 5.2903,
      "mean_token_accuracy": 0.18317030221223832,
      "num_tokens": 9970654.0,
      "step": 5420
    },
    {
      "entropy": 5.469633722305298,
      "epoch": 0.4531406615436017,
      "grad_norm": 1.09375,
      "learning_rate": 0.000498459936359118,
      "loss": 5.3608,
      "mean_token_accuracy": 0.18050043284893036,
      "num_tokens": 9979625.0,
      "step": 5425
    },
    {
      "entropy": 5.3770270347595215,
      "epoch": 0.45355830270631475,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004984564572233568,
      "loss": 5.2839,
      "mean_token_accuracy": 0.186361426115036,
      "num_tokens": 9988584.0,
      "step": 5430
    },
    {
      "entropy": 5.2812597274780275,
      "epoch": 0.45397594386902773,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004984529741757216,
      "loss": 5.2125,
      "mean_token_accuracy": 0.1931207850575447,
      "num_tokens": 9997315.0,
      "step": 5435
    },
    {
      "entropy": 5.296917819976807,
      "epoch": 0.4543935850317407,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004984494872162728,
      "loss": 5.3285,
      "mean_token_accuracy": 0.18080750852823257,
      "num_tokens": 10006431.0,
      "step": 5440
    },
    {
      "entropy": 5.411590051651001,
      "epoch": 0.45481122619445374,
      "grad_norm": 1.125,
      "learning_rate": 0.000498445996345072,
      "loss": 5.3017,
      "mean_token_accuracy": 0.18361044079065322,
      "num_tokens": 10015960.0,
      "step": 5445
    },
    {
      "entropy": 5.426028060913086,
      "epoch": 0.4552288673571667,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004984425015621799,
      "loss": 5.2688,
      "mean_token_accuracy": 0.18067145794630052,
      "num_tokens": 10025948.0,
      "step": 5450
    },
    {
      "entropy": 5.4195390224456785,
      "epoch": 0.4556465085198797,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000498439002867658,
      "loss": 5.3295,
      "mean_token_accuracy": 0.17844995409250258,
      "num_tokens": 10034358.0,
      "step": 5455
    },
    {
      "entropy": 5.446758174896241,
      "epoch": 0.45606414968259273,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004984355002615673,
      "loss": 5.3543,
      "mean_token_accuracy": 0.17786922454833984,
      "num_tokens": 10044887.0,
      "step": 5460
    },
    {
      "entropy": 5.2796142578125,
      "epoch": 0.4564817908453057,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004984319937439693,
      "loss": 5.2354,
      "mean_token_accuracy": 0.18258257806301117,
      "num_tokens": 10054026.0,
      "step": 5465
    },
    {
      "entropy": 5.367075252532959,
      "epoch": 0.4568994320080187,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984284833149252,
      "loss": 5.3638,
      "mean_token_accuracy": 0.17856134474277496,
      "num_tokens": 10063089.0,
      "step": 5470
    },
    {
      "entropy": 5.5954488754272464,
      "epoch": 0.4573170731707317,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004984249689744966,
      "loss": 5.3867,
      "mean_token_accuracy": 0.1661436840891838,
      "num_tokens": 10073816.0,
      "step": 5475
    },
    {
      "entropy": 5.362226915359497,
      "epoch": 0.4577347143334447,
      "grad_norm": 1.21875,
      "learning_rate": 0.000498421450722745,
      "loss": 5.3136,
      "mean_token_accuracy": 0.1795751541852951,
      "num_tokens": 10082407.0,
      "step": 5480
    },
    {
      "entropy": 5.411347723007202,
      "epoch": 0.4581523554961577,
      "grad_norm": 1.109375,
      "learning_rate": 0.000498417928559732,
      "loss": 5.3061,
      "mean_token_accuracy": 0.1772495448589325,
      "num_tokens": 10091423.0,
      "step": 5485
    },
    {
      "entropy": 5.3414701461792,
      "epoch": 0.4585699966588707,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984144024855191,
      "loss": 5.2976,
      "mean_token_accuracy": 0.17482586354017257,
      "num_tokens": 10101381.0,
      "step": 5490
    },
    {
      "entropy": 5.3897545337677,
      "epoch": 0.4589876378215837,
      "grad_norm": 1.0,
      "learning_rate": 0.0004984108725001683,
      "loss": 5.363,
      "mean_token_accuracy": 0.17848258167505265,
      "num_tokens": 10111000.0,
      "step": 5495
    },
    {
      "entropy": 5.51628360748291,
      "epoch": 0.45940527898429667,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004984073386037413,
      "loss": 5.3342,
      "mean_token_accuracy": 0.17876617163419722,
      "num_tokens": 10120380.0,
      "step": 5500
    },
    {
      "entropy": 5.328547382354737,
      "epoch": 0.4598229201470097,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004984038007962997,
      "loss": 5.3124,
      "mean_token_accuracy": 0.18223776817321777,
      "num_tokens": 10129236.0,
      "step": 5505
    },
    {
      "entropy": 5.5399939060211185,
      "epoch": 0.4602405613097227,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984002590779059,
      "loss": 5.4564,
      "mean_token_accuracy": 0.1796054482460022,
      "num_tokens": 10138851.0,
      "step": 5510
    },
    {
      "entropy": 5.37919282913208,
      "epoch": 0.46065820247243566,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983967134486215,
      "loss": 5.3018,
      "mean_token_accuracy": 0.18308076858520508,
      "num_tokens": 10148030.0,
      "step": 5515
    },
    {
      "entropy": 5.351871967315674,
      "epoch": 0.4610758436351487,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004983931639085088,
      "loss": 5.2686,
      "mean_token_accuracy": 0.18655598014593125,
      "num_tokens": 10156362.0,
      "step": 5520
    },
    {
      "entropy": 5.436561489105225,
      "epoch": 0.4614934847978617,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983896104576298,
      "loss": 5.3224,
      "mean_token_accuracy": 0.1759396404027939,
      "num_tokens": 10165603.0,
      "step": 5525
    },
    {
      "entropy": 5.40022840499878,
      "epoch": 0.46191112596057465,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000498386053096047,
      "loss": 5.3074,
      "mean_token_accuracy": 0.1788477197289467,
      "num_tokens": 10175577.0,
      "step": 5530
    },
    {
      "entropy": 5.345335578918457,
      "epoch": 0.4623287671232877,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004983824918238221,
      "loss": 5.2218,
      "mean_token_accuracy": 0.18754889219999313,
      "num_tokens": 10184736.0,
      "step": 5535
    },
    {
      "entropy": 5.342322206497192,
      "epoch": 0.46274640828600067,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000498378926641018,
      "loss": 5.316,
      "mean_token_accuracy": 0.18536394238471984,
      "num_tokens": 10194154.0,
      "step": 5540
    },
    {
      "entropy": 5.4188093662261965,
      "epoch": 0.46316404944871364,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004983753575476969,
      "loss": 5.3617,
      "mean_token_accuracy": 0.17856865227222443,
      "num_tokens": 10203773.0,
      "step": 5545
    },
    {
      "entropy": 5.396222543716431,
      "epoch": 0.4635816906114267,
      "grad_norm": 1.125,
      "learning_rate": 0.0004983717845439212,
      "loss": 5.335,
      "mean_token_accuracy": 0.18030931800603867,
      "num_tokens": 10212260.0,
      "step": 5550
    },
    {
      "entropy": 5.3702692031860355,
      "epoch": 0.46399933177413966,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004983682076297536,
      "loss": 5.3828,
      "mean_token_accuracy": 0.17512561082839967,
      "num_tokens": 10221042.0,
      "step": 5555
    },
    {
      "entropy": 5.424762439727783,
      "epoch": 0.46441697293685263,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004983646268052567,
      "loss": 5.2587,
      "mean_token_accuracy": 0.1812494784593582,
      "num_tokens": 10230336.0,
      "step": 5560
    },
    {
      "entropy": 5.431757879257202,
      "epoch": 0.46483461409956567,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004983610420704931,
      "loss": 5.4368,
      "mean_token_accuracy": 0.1741185799241066,
      "num_tokens": 10241655.0,
      "step": 5565
    },
    {
      "entropy": 5.343291521072388,
      "epoch": 0.46525225526227865,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004983574534255256,
      "loss": 5.3419,
      "mean_token_accuracy": 0.18161144405603408,
      "num_tokens": 10250926.0,
      "step": 5570
    },
    {
      "entropy": 5.370347738265991,
      "epoch": 0.4656698964249916,
      "grad_norm": 1.09375,
      "learning_rate": 0.000498353860870417,
      "loss": 5.2968,
      "mean_token_accuracy": 0.1780700668692589,
      "num_tokens": 10260034.0,
      "step": 5575
    },
    {
      "entropy": 5.453782367706299,
      "epoch": 0.46608753758770466,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004983502644052303,
      "loss": 5.4115,
      "mean_token_accuracy": 0.17498396188020707,
      "num_tokens": 10269170.0,
      "step": 5580
    },
    {
      "entropy": 5.44244704246521,
      "epoch": 0.46650517875041764,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004983466640300282,
      "loss": 5.287,
      "mean_token_accuracy": 0.1836057037115097,
      "num_tokens": 10278385.0,
      "step": 5585
    },
    {
      "entropy": 5.429437446594238,
      "epoch": 0.4669228199131306,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004983430597448741,
      "loss": 5.3021,
      "mean_token_accuracy": 0.18313779979944228,
      "num_tokens": 10287678.0,
      "step": 5590
    },
    {
      "entropy": 5.380919170379639,
      "epoch": 0.46734046107584365,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004983394515498308,
      "loss": 5.3274,
      "mean_token_accuracy": 0.17414292246103286,
      "num_tokens": 10296579.0,
      "step": 5595
    },
    {
      "entropy": 5.444692945480346,
      "epoch": 0.46775810223855663,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004983358394449616,
      "loss": 5.3301,
      "mean_token_accuracy": 0.1867544546723366,
      "num_tokens": 10305812.0,
      "step": 5600
    },
    {
      "entropy": 5.416404581069946,
      "epoch": 0.4681757434012696,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004983322234303297,
      "loss": 5.3575,
      "mean_token_accuracy": 0.17334588319063188,
      "num_tokens": 10314735.0,
      "step": 5605
    },
    {
      "entropy": 5.404338455200195,
      "epoch": 0.46859338456398264,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004983286035059984,
      "loss": 5.4523,
      "mean_token_accuracy": 0.174786077439785,
      "num_tokens": 10323415.0,
      "step": 5610
    },
    {
      "entropy": 5.472856664657593,
      "epoch": 0.4690110257266956,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004983249796720311,
      "loss": 5.299,
      "mean_token_accuracy": 0.18797911256551741,
      "num_tokens": 10331670.0,
      "step": 5615
    },
    {
      "entropy": 5.402751874923706,
      "epoch": 0.4694286668894086,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004983213519284912,
      "loss": 5.4216,
      "mean_token_accuracy": 0.17441252171993255,
      "num_tokens": 10340609.0,
      "step": 5620
    },
    {
      "entropy": 5.316213083267212,
      "epoch": 0.46984630805212163,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004983177202754421,
      "loss": 5.1053,
      "mean_token_accuracy": 0.19350895285606384,
      "num_tokens": 10349035.0,
      "step": 5625
    },
    {
      "entropy": 5.323163414001465,
      "epoch": 0.4702639492148346,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004983140847129476,
      "loss": 5.1956,
      "mean_token_accuracy": 0.18771445751190186,
      "num_tokens": 10358005.0,
      "step": 5630
    },
    {
      "entropy": 5.358368301391602,
      "epoch": 0.4706815903775476,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004983104452410713,
      "loss": 5.3316,
      "mean_token_accuracy": 0.17518524080514908,
      "num_tokens": 10366791.0,
      "step": 5635
    },
    {
      "entropy": 5.3897875308990475,
      "epoch": 0.4710992315402606,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004983068018598768,
      "loss": 5.4229,
      "mean_token_accuracy": 0.17349654883146287,
      "num_tokens": 10376485.0,
      "step": 5640
    },
    {
      "entropy": 5.4193414688110355,
      "epoch": 0.4715168727029736,
      "grad_norm": 1.078125,
      "learning_rate": 0.000498303154569428,
      "loss": 5.337,
      "mean_token_accuracy": 0.1830342084169388,
      "num_tokens": 10385962.0,
      "step": 5645
    },
    {
      "entropy": 5.375597143173218,
      "epoch": 0.4719345138656866,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004982995033697887,
      "loss": 5.2611,
      "mean_token_accuracy": 0.18273811936378478,
      "num_tokens": 10394314.0,
      "step": 5650
    },
    {
      "entropy": 5.326365375518799,
      "epoch": 0.4723521550283996,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004982958482610226,
      "loss": 5.2723,
      "mean_token_accuracy": 0.18153406903147698,
      "num_tokens": 10403733.0,
      "step": 5655
    },
    {
      "entropy": 5.369768476486206,
      "epoch": 0.4727697961911126,
      "grad_norm": 1.09375,
      "learning_rate": 0.000498292189243194,
      "loss": 5.3815,
      "mean_token_accuracy": 0.17268814146518707,
      "num_tokens": 10413979.0,
      "step": 5660
    },
    {
      "entropy": 5.455648899078369,
      "epoch": 0.47318743735382557,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000498288526316367,
      "loss": 5.2425,
      "mean_token_accuracy": 0.17913062870502472,
      "num_tokens": 10422630.0,
      "step": 5665
    },
    {
      "entropy": 5.3129785537719725,
      "epoch": 0.4736050785165386,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004982848594806054,
      "loss": 5.2396,
      "mean_token_accuracy": 0.19069661051034928,
      "num_tokens": 10431422.0,
      "step": 5670
    },
    {
      "entropy": 5.407472419738769,
      "epoch": 0.4740227196792516,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982811887359736,
      "loss": 5.3846,
      "mean_token_accuracy": 0.1760818988084793,
      "num_tokens": 10440953.0,
      "step": 5675
    },
    {
      "entropy": 5.336132574081421,
      "epoch": 0.47444036084196456,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982775140825359,
      "loss": 5.2714,
      "mean_token_accuracy": 0.18437761962413787,
      "num_tokens": 10449331.0,
      "step": 5680
    },
    {
      "entropy": 5.3497191905975345,
      "epoch": 0.4748580020046776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982738355203565,
      "loss": 5.2956,
      "mean_token_accuracy": 0.18434551656246184,
      "num_tokens": 10458021.0,
      "step": 5685
    },
    {
      "entropy": 5.471211719512939,
      "epoch": 0.47527564316739057,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004982701530494998,
      "loss": 5.2941,
      "mean_token_accuracy": 0.18104884326457976,
      "num_tokens": 10466743.0,
      "step": 5690
    },
    {
      "entropy": 5.345020341873169,
      "epoch": 0.47569328433010355,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004982664666700304,
      "loss": 5.2955,
      "mean_token_accuracy": 0.18051659911870957,
      "num_tokens": 10475537.0,
      "step": 5695
    },
    {
      "entropy": 5.492947149276733,
      "epoch": 0.4761109254928166,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004982627763820128,
      "loss": 5.4356,
      "mean_token_accuracy": 0.17569617182016373,
      "num_tokens": 10485257.0,
      "step": 5700
    },
    {
      "entropy": 5.471468591690064,
      "epoch": 0.47652856665552956,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004982590821855115,
      "loss": 5.3715,
      "mean_token_accuracy": 0.18662190884351731,
      "num_tokens": 10494449.0,
      "step": 5705
    },
    {
      "entropy": 5.367533349990845,
      "epoch": 0.47694620781824254,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004982553840805912,
      "loss": 5.3658,
      "mean_token_accuracy": 0.17808520942926406,
      "num_tokens": 10504216.0,
      "step": 5710
    },
    {
      "entropy": 5.453322029113769,
      "epoch": 0.4773638489809556,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004982516820673167,
      "loss": 5.3304,
      "mean_token_accuracy": 0.1843886598944664,
      "num_tokens": 10512676.0,
      "step": 5715
    },
    {
      "entropy": 5.36357626914978,
      "epoch": 0.47778149014366855,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004982479761457528,
      "loss": 5.2156,
      "mean_token_accuracy": 0.18708961457014084,
      "num_tokens": 10521587.0,
      "step": 5720
    },
    {
      "entropy": 5.407714462280273,
      "epoch": 0.47819913130638153,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004982442663159643,
      "loss": 5.3579,
      "mean_token_accuracy": 0.17724160701036454,
      "num_tokens": 10530043.0,
      "step": 5725
    },
    {
      "entropy": 5.332644748687744,
      "epoch": 0.47861677246909456,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004982405525780162,
      "loss": 5.195,
      "mean_token_accuracy": 0.18853996247053145,
      "num_tokens": 10539491.0,
      "step": 5730
    },
    {
      "entropy": 5.338552856445313,
      "epoch": 0.47903441363180754,
      "grad_norm": 1.125,
      "learning_rate": 0.0004982368349319736,
      "loss": 5.21,
      "mean_token_accuracy": 0.18114500492811203,
      "num_tokens": 10547893.0,
      "step": 5735
    },
    {
      "entropy": 5.420573902130127,
      "epoch": 0.4794520547945205,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004982331133779014,
      "loss": 5.3839,
      "mean_token_accuracy": 0.1755267933011055,
      "num_tokens": 10557597.0,
      "step": 5740
    },
    {
      "entropy": 5.4395363330841064,
      "epoch": 0.47986969595723356,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004982293879158649,
      "loss": 5.2929,
      "mean_token_accuracy": 0.18169630765914918,
      "num_tokens": 10566349.0,
      "step": 5745
    },
    {
      "entropy": 5.403287315368653,
      "epoch": 0.48028733711994653,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004982256585459293,
      "loss": 5.3656,
      "mean_token_accuracy": 0.18082840144634246,
      "num_tokens": 10574897.0,
      "step": 5750
    },
    {
      "entropy": 5.421693181991577,
      "epoch": 0.4807049782826595,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004982219252681597,
      "loss": 5.279,
      "mean_token_accuracy": 0.1854931116104126,
      "num_tokens": 10583233.0,
      "step": 5755
    },
    {
      "entropy": 5.257638454437256,
      "epoch": 0.48112261944537255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004982181880826218,
      "loss": 5.2115,
      "mean_token_accuracy": 0.1813637986779213,
      "num_tokens": 10592202.0,
      "step": 5760
    },
    {
      "entropy": 5.33853554725647,
      "epoch": 0.4815402606080855,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982144469893807,
      "loss": 5.318,
      "mean_token_accuracy": 0.17970952838659288,
      "num_tokens": 10601221.0,
      "step": 5765
    },
    {
      "entropy": 5.459839248657227,
      "epoch": 0.4819579017707985,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000498210701988502,
      "loss": 5.3788,
      "mean_token_accuracy": 0.17642758935689926,
      "num_tokens": 10610983.0,
      "step": 5770
    },
    {
      "entropy": 5.473546266555786,
      "epoch": 0.48237554293351154,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004982069530800514,
      "loss": 5.3938,
      "mean_token_accuracy": 0.1791031002998352,
      "num_tokens": 10619378.0,
      "step": 5775
    },
    {
      "entropy": 5.433433723449707,
      "epoch": 0.4827931840962245,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982032002640943,
      "loss": 5.2721,
      "mean_token_accuracy": 0.18506969809532164,
      "num_tokens": 10628288.0,
      "step": 5780
    },
    {
      "entropy": 5.428299617767334,
      "epoch": 0.48321082525893755,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004981994435406965,
      "loss": 5.445,
      "mean_token_accuracy": 0.1701508119702339,
      "num_tokens": 10639433.0,
      "step": 5785
    },
    {
      "entropy": 5.355684995651245,
      "epoch": 0.4836284664216505,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004981956829099239,
      "loss": 5.2258,
      "mean_token_accuracy": 0.1857157438993454,
      "num_tokens": 10649654.0,
      "step": 5790
    },
    {
      "entropy": 5.3672160625457765,
      "epoch": 0.4840461075843635,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004981919183718422,
      "loss": 5.3813,
      "mean_token_accuracy": 0.18063102960586547,
      "num_tokens": 10658620.0,
      "step": 5795
    },
    {
      "entropy": 5.387696599960327,
      "epoch": 0.48446374874707654,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981881499265173,
      "loss": 5.2062,
      "mean_token_accuracy": 0.1807701528072357,
      "num_tokens": 10667942.0,
      "step": 5800
    },
    {
      "entropy": 5.285345268249512,
      "epoch": 0.4848813899097895,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004981843775740151,
      "loss": 5.2644,
      "mean_token_accuracy": 0.1784549370408058,
      "num_tokens": 10676895.0,
      "step": 5805
    },
    {
      "entropy": 5.532047176361084,
      "epoch": 0.4852990310725025,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004981806013144017,
      "loss": 5.4255,
      "mean_token_accuracy": 0.1725636303424835,
      "num_tokens": 10685986.0,
      "step": 5810
    },
    {
      "entropy": 5.472192573547363,
      "epoch": 0.48571667223521553,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004981768211477434,
      "loss": 5.3139,
      "mean_token_accuracy": 0.18568273782730102,
      "num_tokens": 10694430.0,
      "step": 5815
    },
    {
      "entropy": 5.282784032821655,
      "epoch": 0.4861343133979285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498173037074106,
      "loss": 5.2105,
      "mean_token_accuracy": 0.17912789583206176,
      "num_tokens": 10702997.0,
      "step": 5820
    },
    {
      "entropy": 5.38949728012085,
      "epoch": 0.4865519545606415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981692490935561,
      "loss": 5.3882,
      "mean_token_accuracy": 0.18065113425254822,
      "num_tokens": 10711652.0,
      "step": 5825
    },
    {
      "entropy": 5.496825456619263,
      "epoch": 0.4869695957233545,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004981654572061597,
      "loss": 5.3689,
      "mean_token_accuracy": 0.17776526212692262,
      "num_tokens": 10721749.0,
      "step": 5830
    },
    {
      "entropy": 5.397292232513427,
      "epoch": 0.4873872368860675,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004981616614119836,
      "loss": 5.2877,
      "mean_token_accuracy": 0.17844431102275848,
      "num_tokens": 10729678.0,
      "step": 5835
    },
    {
      "entropy": 5.380484914779663,
      "epoch": 0.4878048780487805,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004981578617110938,
      "loss": 5.3325,
      "mean_token_accuracy": 0.17817786931991578,
      "num_tokens": 10740276.0,
      "step": 5840
    },
    {
      "entropy": 5.3566937923431395,
      "epoch": 0.4882225192114935,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004981540581035571,
      "loss": 5.3434,
      "mean_token_accuracy": 0.1754148170351982,
      "num_tokens": 10749249.0,
      "step": 5845
    },
    {
      "entropy": 5.469118213653564,
      "epoch": 0.4886401603742065,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049815025058944,
      "loss": 5.2156,
      "mean_token_accuracy": 0.1886957809329033,
      "num_tokens": 10758888.0,
      "step": 5850
    },
    {
      "entropy": 5.273655605316162,
      "epoch": 0.48905780153691947,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004981464391688092,
      "loss": 5.2877,
      "mean_token_accuracy": 0.18605542778968812,
      "num_tokens": 10768156.0,
      "step": 5855
    },
    {
      "entropy": 5.278586053848267,
      "epoch": 0.4894754426996325,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004981426238417312,
      "loss": 5.243,
      "mean_token_accuracy": 0.18633593171834945,
      "num_tokens": 10776762.0,
      "step": 5860
    },
    {
      "entropy": 5.447524738311768,
      "epoch": 0.4898930838623455,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004981388046082731,
      "loss": 5.3322,
      "mean_token_accuracy": 0.17448019236326218,
      "num_tokens": 10785495.0,
      "step": 5865
    },
    {
      "entropy": 5.384194087982178,
      "epoch": 0.49031072502505846,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981349814685016,
      "loss": 5.2171,
      "mean_token_accuracy": 0.1862630292773247,
      "num_tokens": 10794400.0,
      "step": 5870
    },
    {
      "entropy": 5.349345016479492,
      "epoch": 0.4907283661877715,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004981311544224838,
      "loss": 5.274,
      "mean_token_accuracy": 0.18068427443504334,
      "num_tokens": 10803138.0,
      "step": 5875
    },
    {
      "entropy": 5.395100116729736,
      "epoch": 0.49114600735048447,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004981273234702864,
      "loss": 5.3089,
      "mean_token_accuracy": 0.1752660870552063,
      "num_tokens": 10813583.0,
      "step": 5880
    },
    {
      "entropy": 5.316332530975342,
      "epoch": 0.49156364851319745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004981234886119767,
      "loss": 5.2277,
      "mean_token_accuracy": 0.18534980863332748,
      "num_tokens": 10822532.0,
      "step": 5885
    },
    {
      "entropy": 5.319725513458252,
      "epoch": 0.4919812896759105,
      "grad_norm": 1.125,
      "learning_rate": 0.0004981196498476217,
      "loss": 5.3016,
      "mean_token_accuracy": 0.18608538955450057,
      "num_tokens": 10832256.0,
      "step": 5890
    },
    {
      "entropy": 5.3289566993713375,
      "epoch": 0.49239893083862346,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004981158071772887,
      "loss": 5.1902,
      "mean_token_accuracy": 0.18748476803302766,
      "num_tokens": 10840909.0,
      "step": 5895
    },
    {
      "entropy": 5.372947549819946,
      "epoch": 0.49281657200133644,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981119606010449,
      "loss": 5.2664,
      "mean_token_accuracy": 0.18154753595590592,
      "num_tokens": 10849931.0,
      "step": 5900
    },
    {
      "entropy": 5.364373111724854,
      "epoch": 0.4932342131640495,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004981081101189577,
      "loss": 5.3003,
      "mean_token_accuracy": 0.18243797421455382,
      "num_tokens": 10859518.0,
      "step": 5905
    },
    {
      "entropy": 5.352173852920532,
      "epoch": 0.49365185432676245,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004981042557310945,
      "loss": 5.396,
      "mean_token_accuracy": 0.1771364316344261,
      "num_tokens": 10868567.0,
      "step": 5910
    },
    {
      "entropy": 5.324528312683105,
      "epoch": 0.49406949548947543,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004981003974375227,
      "loss": 5.1456,
      "mean_token_accuracy": 0.19510713815689087,
      "num_tokens": 10877463.0,
      "step": 5915
    },
    {
      "entropy": 5.327762222290039,
      "epoch": 0.49448713665218846,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004980965352383099,
      "loss": 5.2654,
      "mean_token_accuracy": 0.1772697314620018,
      "num_tokens": 10886228.0,
      "step": 5920
    },
    {
      "entropy": 5.468494606018067,
      "epoch": 0.49490477781490144,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004980926691335238,
      "loss": 5.3655,
      "mean_token_accuracy": 0.17341220527887344,
      "num_tokens": 10895415.0,
      "step": 5925
    },
    {
      "entropy": 5.3670299530029295,
      "epoch": 0.4953224189776144,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980887991232318,
      "loss": 5.2754,
      "mean_token_accuracy": 0.1696104109287262,
      "num_tokens": 10904146.0,
      "step": 5930
    },
    {
      "entropy": 5.351732397079468,
      "epoch": 0.49574006014032745,
      "grad_norm": 1.078125,
      "learning_rate": 0.000498084925207502,
      "loss": 5.3254,
      "mean_token_accuracy": 0.17885888516902923,
      "num_tokens": 10914027.0,
      "step": 5935
    },
    {
      "entropy": 5.350140905380249,
      "epoch": 0.49615770130304043,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498081047386402,
      "loss": 5.2514,
      "mean_token_accuracy": 0.19121921807527542,
      "num_tokens": 10923065.0,
      "step": 5940
    },
    {
      "entropy": 5.4460278987884525,
      "epoch": 0.4965753424657534,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004980771656599998,
      "loss": 5.407,
      "mean_token_accuracy": 0.1723595902323723,
      "num_tokens": 10932752.0,
      "step": 5945
    },
    {
      "entropy": 5.426205968856811,
      "epoch": 0.49699298362846644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004980732800283633,
      "loss": 5.3437,
      "mean_token_accuracy": 0.18056864738464357,
      "num_tokens": 10942536.0,
      "step": 5950
    },
    {
      "entropy": 5.310159063339233,
      "epoch": 0.4974106247911794,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004980693904915604,
      "loss": 5.2472,
      "mean_token_accuracy": 0.18073009699583054,
      "num_tokens": 10952258.0,
      "step": 5955
    },
    {
      "entropy": 5.361242389678955,
      "epoch": 0.4978282659538924,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004980654970496595,
      "loss": 5.2951,
      "mean_token_accuracy": 0.17866345047950744,
      "num_tokens": 10961590.0,
      "step": 5960
    },
    {
      "entropy": 5.437377309799194,
      "epoch": 0.49824590711660544,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004980615997027284,
      "loss": 5.2668,
      "mean_token_accuracy": 0.18697986453771592,
      "num_tokens": 10971380.0,
      "step": 5965
    },
    {
      "entropy": 5.291711235046387,
      "epoch": 0.4986635482793184,
      "grad_norm": 1.125,
      "learning_rate": 0.0004980576984508356,
      "loss": 5.2741,
      "mean_token_accuracy": 0.1789490357041359,
      "num_tokens": 10979510.0,
      "step": 5970
    },
    {
      "entropy": 5.360995960235596,
      "epoch": 0.4990811894420314,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004980537932940492,
      "loss": 5.2772,
      "mean_token_accuracy": 0.18087197691202164,
      "num_tokens": 10988921.0,
      "step": 5975
    },
    {
      "entropy": 5.389206314086914,
      "epoch": 0.4994988306047444,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980498842324377,
      "loss": 5.2577,
      "mean_token_accuracy": 0.18446117490530015,
      "num_tokens": 10998644.0,
      "step": 5980
    },
    {
      "entropy": 5.250856828689575,
      "epoch": 0.4999164717674574,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004980459712660694,
      "loss": 5.1759,
      "mean_token_accuracy": 0.1854647397994995,
      "num_tokens": 11006975.0,
      "step": 5985
    },
    {
      "entropy": 5.3128009796142575,
      "epoch": 0.5003341129301704,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004980420543950131,
      "loss": 5.233,
      "mean_token_accuracy": 0.1824902191758156,
      "num_tokens": 11016011.0,
      "step": 5990
    },
    {
      "entropy": 5.439278030395508,
      "epoch": 0.5007517540928834,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004980381336193371,
      "loss": 5.4504,
      "mean_token_accuracy": 0.17257942259311676,
      "num_tokens": 11025600.0,
      "step": 5995
    },
    {
      "entropy": 5.371862173080444,
      "epoch": 0.5011693952555963,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004980342089391099,
      "loss": 5.1841,
      "mean_token_accuracy": 0.18732891231775284,
      "num_tokens": 11034747.0,
      "step": 6000
    },
    {
      "epoch": 0.5011693952555963,
      "eval_entropy": 5.168086645437029,
      "eval_loss": 5.272536754608154,
      "eval_mean_token_accuracy": 0.19146194885189163,
      "eval_num_tokens": 11034747.0,
      "eval_runtime": 27.4648,
      "eval_samples_per_second": 1374.96,
      "eval_steps_per_second": 171.893,
      "step": 6000
    },
    {
      "entropy": 5.394799995422363,
      "epoch": 0.5015870364183094,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004980302803544006,
      "loss": 5.3342,
      "mean_token_accuracy": 0.1796693503856659,
      "num_tokens": 11043461.0,
      "step": 6005
    },
    {
      "entropy": 5.376993131637573,
      "epoch": 0.5020046775810224,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004980263478652778,
      "loss": 5.3624,
      "mean_token_accuracy": 0.18350306749343873,
      "num_tokens": 11053632.0,
      "step": 6010
    },
    {
      "entropy": 5.350661516189575,
      "epoch": 0.5024223187437353,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004980224114718102,
      "loss": 5.274,
      "mean_token_accuracy": 0.18089560121297837,
      "num_tokens": 11062620.0,
      "step": 6015
    },
    {
      "entropy": 5.324182462692261,
      "epoch": 0.5028399599064484,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004980184711740669,
      "loss": 5.2863,
      "mean_token_accuracy": 0.17761649787425995,
      "num_tokens": 11071105.0,
      "step": 6020
    },
    {
      "entropy": 5.406068515777588,
      "epoch": 0.5032576010691614,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004980145269721168,
      "loss": 5.2631,
      "mean_token_accuracy": 0.19173841178417206,
      "num_tokens": 11079926.0,
      "step": 6025
    },
    {
      "entropy": 5.341655635833741,
      "epoch": 0.5036752422318743,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498010578866029,
      "loss": 5.2522,
      "mean_token_accuracy": 0.17975347489118576,
      "num_tokens": 11089395.0,
      "step": 6030
    },
    {
      "entropy": 5.341538000106811,
      "epoch": 0.5040928833945874,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004980066268558726,
      "loss": 5.2894,
      "mean_token_accuracy": 0.1858112931251526,
      "num_tokens": 11098464.0,
      "step": 6035
    },
    {
      "entropy": 5.436836290359497,
      "epoch": 0.5045105245573004,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004980026709417167,
      "loss": 5.3628,
      "mean_token_accuracy": 0.18345711678266524,
      "num_tokens": 11107861.0,
      "step": 6040
    },
    {
      "entropy": 5.264323425292969,
      "epoch": 0.5049281657200133,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004979987111236306,
      "loss": 5.2102,
      "mean_token_accuracy": 0.1859588786959648,
      "num_tokens": 11116453.0,
      "step": 6045
    },
    {
      "entropy": 5.344191026687622,
      "epoch": 0.5053458068827263,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004979947474016837,
      "loss": 5.3059,
      "mean_token_accuracy": 0.18284351974725724,
      "num_tokens": 11125166.0,
      "step": 6050
    },
    {
      "entropy": 5.512290000915527,
      "epoch": 0.5057634480454394,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004979907797759453,
      "loss": 5.3684,
      "mean_token_accuracy": 0.17712332904338837,
      "num_tokens": 11134560.0,
      "step": 6055
    },
    {
      "entropy": 5.374829244613648,
      "epoch": 0.5061810892081523,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004979868082464849,
      "loss": 5.2323,
      "mean_token_accuracy": 0.18858323246240616,
      "num_tokens": 11143994.0,
      "step": 6060
    },
    {
      "entropy": 5.292056083679199,
      "epoch": 0.5065987303708653,
      "grad_norm": 1.125,
      "learning_rate": 0.0004979828328133719,
      "loss": 5.2665,
      "mean_token_accuracy": 0.18212510496377946,
      "num_tokens": 11152234.0,
      "step": 6065
    },
    {
      "entropy": 5.347263526916504,
      "epoch": 0.5070163715335784,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004979788534766762,
      "loss": 5.1264,
      "mean_token_accuracy": 0.18932535350322724,
      "num_tokens": 11161334.0,
      "step": 6070
    },
    {
      "entropy": 5.229281902313232,
      "epoch": 0.5074340126962913,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004979748702364672,
      "loss": 5.1679,
      "mean_token_accuracy": 0.17916332483291625,
      "num_tokens": 11170510.0,
      "step": 6075
    },
    {
      "entropy": 5.277205610275269,
      "epoch": 0.5078516538590043,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004979708830928147,
      "loss": 5.1978,
      "mean_token_accuracy": 0.19500905573368071,
      "num_tokens": 11179378.0,
      "step": 6080
    },
    {
      "entropy": 5.34964804649353,
      "epoch": 0.5082692950217174,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004979668920457884,
      "loss": 5.359,
      "mean_token_accuracy": 0.17949633449316024,
      "num_tokens": 11189534.0,
      "step": 6085
    },
    {
      "entropy": 5.319961977005005,
      "epoch": 0.5086869361844303,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004979628970954584,
      "loss": 5.2073,
      "mean_token_accuracy": 0.1925006777048111,
      "num_tokens": 11197827.0,
      "step": 6090
    },
    {
      "entropy": 5.285190200805664,
      "epoch": 0.5091045773471433,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004979588982418944,
      "loss": 5.2304,
      "mean_token_accuracy": 0.18829721659421922,
      "num_tokens": 11206759.0,
      "step": 6095
    },
    {
      "entropy": 5.317176580429077,
      "epoch": 0.5095222185098564,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004979548954851666,
      "loss": 5.1617,
      "mean_token_accuracy": 0.18879591822624206,
      "num_tokens": 11216021.0,
      "step": 6100
    },
    {
      "entropy": 5.352275466918945,
      "epoch": 0.5099398596725693,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497950888825345,
      "loss": 5.2652,
      "mean_token_accuracy": 0.18401817828416825,
      "num_tokens": 11224936.0,
      "step": 6105
    },
    {
      "entropy": 5.258750486373901,
      "epoch": 0.5103575008352823,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004979468782624996,
      "loss": 5.1997,
      "mean_token_accuracy": 0.1853033795952797,
      "num_tokens": 11233905.0,
      "step": 6110
    },
    {
      "entropy": 5.326344060897827,
      "epoch": 0.5107751419979953,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004979428637967009,
      "loss": 5.2003,
      "mean_token_accuracy": 0.1920333117246628,
      "num_tokens": 11242470.0,
      "step": 6115
    },
    {
      "entropy": 5.311832761764526,
      "epoch": 0.5111927831607083,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004979388454280189,
      "loss": 5.2477,
      "mean_token_accuracy": 0.17805823534727097,
      "num_tokens": 11251861.0,
      "step": 6120
    },
    {
      "entropy": 5.367962598800659,
      "epoch": 0.5116104243234213,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000497934823156524,
      "loss": 5.2545,
      "mean_token_accuracy": 0.18427685350179673,
      "num_tokens": 11260830.0,
      "step": 6125
    },
    {
      "entropy": 5.274614429473877,
      "epoch": 0.5120280654861343,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004979307969822869,
      "loss": 5.2527,
      "mean_token_accuracy": 0.18828731626272202,
      "num_tokens": 11269906.0,
      "step": 6130
    },
    {
      "entropy": 5.392090368270874,
      "epoch": 0.5124457066488474,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004979267669053776,
      "loss": 5.3299,
      "mean_token_accuracy": 0.18325559794902802,
      "num_tokens": 11279471.0,
      "step": 6135
    },
    {
      "entropy": 5.494892406463623,
      "epoch": 0.5128633478115603,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004979227329258671,
      "loss": 5.375,
      "mean_token_accuracy": 0.17832857221364976,
      "num_tokens": 11288586.0,
      "step": 6140
    },
    {
      "entropy": 5.371882963180542,
      "epoch": 0.5132809889742733,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004979186950438256,
      "loss": 5.3656,
      "mean_token_accuracy": 0.18487958014011383,
      "num_tokens": 11297768.0,
      "step": 6145
    },
    {
      "entropy": 5.271016025543213,
      "epoch": 0.5136986301369864,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004979146532593243,
      "loss": 5.2111,
      "mean_token_accuracy": 0.187505404651165,
      "num_tokens": 11307096.0,
      "step": 6150
    },
    {
      "entropy": 5.2447319507598875,
      "epoch": 0.5141162712996993,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004979106075724334,
      "loss": 5.0824,
      "mean_token_accuracy": 0.1980063021183014,
      "num_tokens": 11315518.0,
      "step": 6155
    },
    {
      "entropy": 5.287977123260498,
      "epoch": 0.5145339124624123,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004979065579832242,
      "loss": 5.1794,
      "mean_token_accuracy": 0.18654987812042237,
      "num_tokens": 11324232.0,
      "step": 6160
    },
    {
      "entropy": 5.291946649551392,
      "epoch": 0.5149515536251253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004979025044917673,
      "loss": 5.2933,
      "mean_token_accuracy": 0.1816413149237633,
      "num_tokens": 11332923.0,
      "step": 6165
    },
    {
      "entropy": 5.396589136123657,
      "epoch": 0.5153691947878383,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004978984470981339,
      "loss": 5.299,
      "mean_token_accuracy": 0.18228267282247543,
      "num_tokens": 11342350.0,
      "step": 6170
    },
    {
      "entropy": 5.327940893173218,
      "epoch": 0.5157868359505513,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004978943858023946,
      "loss": 5.1379,
      "mean_token_accuracy": 0.1896440178155899,
      "num_tokens": 11350982.0,
      "step": 6175
    },
    {
      "entropy": 5.320274305343628,
      "epoch": 0.5162044771132643,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000497890320604621,
      "loss": 5.2553,
      "mean_token_accuracy": 0.18778664916753768,
      "num_tokens": 11360637.0,
      "step": 6180
    },
    {
      "entropy": 5.281010341644287,
      "epoch": 0.5166221182759773,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000497886251504884,
      "loss": 5.1994,
      "mean_token_accuracy": 0.1879116863012314,
      "num_tokens": 11368783.0,
      "step": 6185
    },
    {
      "entropy": 5.260682153701782,
      "epoch": 0.5170397594386903,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004978821785032549,
      "loss": 5.199,
      "mean_token_accuracy": 0.18889974802732468,
      "num_tokens": 11378359.0,
      "step": 6190
    },
    {
      "entropy": 5.3671022891998295,
      "epoch": 0.5174574006014033,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004978781015998048,
      "loss": 5.3692,
      "mean_token_accuracy": 0.18173590898513795,
      "num_tokens": 11387235.0,
      "step": 6195
    },
    {
      "entropy": 5.400910234451294,
      "epoch": 0.5178750417641163,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004978740207946056,
      "loss": 5.2559,
      "mean_token_accuracy": 0.18128463476896287,
      "num_tokens": 11396400.0,
      "step": 6200
    },
    {
      "entropy": 5.402005577087403,
      "epoch": 0.5182926829268293,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004978699360877281,
      "loss": 5.3207,
      "mean_token_accuracy": 0.18878876864910127,
      "num_tokens": 11405589.0,
      "step": 6205
    },
    {
      "entropy": 5.29880199432373,
      "epoch": 0.5187103240895423,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004978658474792442,
      "loss": 5.2803,
      "mean_token_accuracy": 0.17831768691539765,
      "num_tokens": 11414983.0,
      "step": 6210
    },
    {
      "entropy": 5.375881719589233,
      "epoch": 0.5191279652522552,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004978617549692253,
      "loss": 5.3074,
      "mean_token_accuracy": 0.1887017771601677,
      "num_tokens": 11423172.0,
      "step": 6215
    },
    {
      "entropy": 5.417163419723511,
      "epoch": 0.5195456064149683,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004978576585577432,
      "loss": 5.2478,
      "mean_token_accuracy": 0.18309810310602187,
      "num_tokens": 11432758.0,
      "step": 6220
    },
    {
      "entropy": 5.248141956329346,
      "epoch": 0.5199632475776813,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004978535582448695,
      "loss": 5.2349,
      "mean_token_accuracy": 0.18524384051561354,
      "num_tokens": 11441754.0,
      "step": 6225
    },
    {
      "entropy": 5.318865442276001,
      "epoch": 0.5203808887403942,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004978494540306761,
      "loss": 5.2414,
      "mean_token_accuracy": 0.18242405503988265,
      "num_tokens": 11451176.0,
      "step": 6230
    },
    {
      "entropy": 5.402262163162232,
      "epoch": 0.5207985299031073,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004978453459152346,
      "loss": 5.3693,
      "mean_token_accuracy": 0.18198805004358293,
      "num_tokens": 11461353.0,
      "step": 6235
    },
    {
      "entropy": 5.390100288391113,
      "epoch": 0.5212161710658203,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004978412338986172,
      "loss": 5.3194,
      "mean_token_accuracy": 0.17699251919984818,
      "num_tokens": 11470103.0,
      "step": 6240
    },
    {
      "entropy": 5.353792905807495,
      "epoch": 0.5216338122285332,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004978371179808958,
      "loss": 5.2242,
      "mean_token_accuracy": 0.19344995766878129,
      "num_tokens": 11480348.0,
      "step": 6245
    },
    {
      "entropy": 5.366096448898316,
      "epoch": 0.5220514533912463,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004978329981621424,
      "loss": 5.2926,
      "mean_token_accuracy": 0.18386424630880355,
      "num_tokens": 11490836.0,
      "step": 6250
    },
    {
      "entropy": 5.267794942855835,
      "epoch": 0.5224690945539593,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000497828874442429,
      "loss": 5.1759,
      "mean_token_accuracy": 0.19093482792377472,
      "num_tokens": 11500344.0,
      "step": 6255
    },
    {
      "entropy": 5.4199565887451175,
      "epoch": 0.5228867357166722,
      "grad_norm": 1.125,
      "learning_rate": 0.000497824746821828,
      "loss": 5.3225,
      "mean_token_accuracy": 0.18004805445671082,
      "num_tokens": 11509624.0,
      "step": 6260
    },
    {
      "entropy": 5.363396215438843,
      "epoch": 0.5233043768793852,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004978206153004117,
      "loss": 5.3097,
      "mean_token_accuracy": 0.17866567373275757,
      "num_tokens": 11519915.0,
      "step": 6265
    },
    {
      "entropy": 5.339620590209961,
      "epoch": 0.5237220180420983,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004978164798782523,
      "loss": 5.1342,
      "mean_token_accuracy": 0.18377067893743515,
      "num_tokens": 11528918.0,
      "step": 6270
    },
    {
      "entropy": 5.312790298461914,
      "epoch": 0.5241396592048112,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004978123405554221,
      "loss": 5.2752,
      "mean_token_accuracy": 0.18144809901714326,
      "num_tokens": 11537396.0,
      "step": 6275
    },
    {
      "entropy": 5.415223217010498,
      "epoch": 0.5245573003675242,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004978081973319938,
      "loss": 5.4156,
      "mean_token_accuracy": 0.17622718214988708,
      "num_tokens": 11547196.0,
      "step": 6280
    },
    {
      "entropy": 5.396162748336792,
      "epoch": 0.5249749415302373,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004978040502080398,
      "loss": 5.2548,
      "mean_token_accuracy": 0.18347851037979127,
      "num_tokens": 11556126.0,
      "step": 6285
    },
    {
      "entropy": 5.36080379486084,
      "epoch": 0.5253925826929502,
      "grad_norm": 1.125,
      "learning_rate": 0.0004977998991836327,
      "loss": 5.3083,
      "mean_token_accuracy": 0.1859940469264984,
      "num_tokens": 11565622.0,
      "step": 6290
    },
    {
      "entropy": 5.297771167755127,
      "epoch": 0.5258102238556632,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004977957442588451,
      "loss": 5.1626,
      "mean_token_accuracy": 0.18722766786813735,
      "num_tokens": 11576052.0,
      "step": 6295
    },
    {
      "entropy": 5.355345773696899,
      "epoch": 0.5262278650183763,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004977915854337498,
      "loss": 5.3068,
      "mean_token_accuracy": 0.16857851147651673,
      "num_tokens": 11585907.0,
      "step": 6300
    },
    {
      "entropy": 5.287508487701416,
      "epoch": 0.5266455061810892,
      "grad_norm": 1.125,
      "learning_rate": 0.0004977874227084197,
      "loss": 5.1824,
      "mean_token_accuracy": 0.18631986975669862,
      "num_tokens": 11594802.0,
      "step": 6305
    },
    {
      "entropy": 5.292425870895386,
      "epoch": 0.5270631473438022,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004977832560829277,
      "loss": 5.2174,
      "mean_token_accuracy": 0.19151604771614075,
      "num_tokens": 11604026.0,
      "step": 6310
    },
    {
      "entropy": 5.32039098739624,
      "epoch": 0.5274807885065153,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004977790855573464,
      "loss": 5.3018,
      "mean_token_accuracy": 0.1752871811389923,
      "num_tokens": 11614209.0,
      "step": 6315
    },
    {
      "entropy": 5.335412788391113,
      "epoch": 0.5278984296692282,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004977749111317492,
      "loss": 5.1697,
      "mean_token_accuracy": 0.18593906611204147,
      "num_tokens": 11622492.0,
      "step": 6320
    },
    {
      "entropy": 5.362774753570557,
      "epoch": 0.5283160708319412,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000497770732806209,
      "loss": 5.207,
      "mean_token_accuracy": 0.18361861407756805,
      "num_tokens": 11631106.0,
      "step": 6325
    },
    {
      "entropy": 5.334519529342652,
      "epoch": 0.5287337119946542,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000497766550580799,
      "loss": 5.2684,
      "mean_token_accuracy": 0.18772526681423188,
      "num_tokens": 11639748.0,
      "step": 6330
    },
    {
      "entropy": 5.312885999679565,
      "epoch": 0.5291513531573672,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004977623644555923,
      "loss": 5.2546,
      "mean_token_accuracy": 0.18425588458776473,
      "num_tokens": 11648524.0,
      "step": 6335
    },
    {
      "entropy": 5.372761058807373,
      "epoch": 0.5295689943200802,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004977581744306624,
      "loss": 5.248,
      "mean_token_accuracy": 0.178777277469635,
      "num_tokens": 11658439.0,
      "step": 6340
    },
    {
      "entropy": 5.339153194427491,
      "epoch": 0.5299866354827932,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004977539805060824,
      "loss": 5.2525,
      "mean_token_accuracy": 0.18067725151777267,
      "num_tokens": 11666319.0,
      "step": 6345
    },
    {
      "entropy": 5.301879787445069,
      "epoch": 0.5304042766455062,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004977497826819259,
      "loss": 5.2733,
      "mean_token_accuracy": 0.1856699615716934,
      "num_tokens": 11675506.0,
      "step": 6350
    },
    {
      "entropy": 5.319541597366333,
      "epoch": 0.5308219178082192,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004977455809582664,
      "loss": 5.2294,
      "mean_token_accuracy": 0.1835037186741829,
      "num_tokens": 11685613.0,
      "step": 6355
    },
    {
      "entropy": 5.319072628021241,
      "epoch": 0.5312395589709322,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004977413753351772,
      "loss": 5.2734,
      "mean_token_accuracy": 0.1810879200696945,
      "num_tokens": 11694896.0,
      "step": 6360
    },
    {
      "entropy": 5.491897201538086,
      "epoch": 0.5316572001336451,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004977371658127323,
      "loss": 5.3726,
      "mean_token_accuracy": 0.17695481777191163,
      "num_tokens": 11703599.0,
      "step": 6365
    },
    {
      "entropy": 5.344759941101074,
      "epoch": 0.5320748412963582,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004977329523910051,
      "loss": 5.1779,
      "mean_token_accuracy": 0.1957557275891304,
      "num_tokens": 11712155.0,
      "step": 6370
    },
    {
      "entropy": 5.273065137863159,
      "epoch": 0.5324924824590712,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004977287350700695,
      "loss": 5.2773,
      "mean_token_accuracy": 0.18458784073591233,
      "num_tokens": 11721224.0,
      "step": 6375
    },
    {
      "entropy": 5.3125457763671875,
      "epoch": 0.5329101236217841,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004977245138499993,
      "loss": 5.1416,
      "mean_token_accuracy": 0.18925199061632156,
      "num_tokens": 11729733.0,
      "step": 6380
    },
    {
      "entropy": 5.3117241859436035,
      "epoch": 0.5333277647844972,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004977202887308684,
      "loss": 5.3434,
      "mean_token_accuracy": 0.17879771590232849,
      "num_tokens": 11738840.0,
      "step": 6385
    },
    {
      "entropy": 5.356008720397949,
      "epoch": 0.5337454059472102,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004977160597127507,
      "loss": 5.3318,
      "mean_token_accuracy": 0.1799089327454567,
      "num_tokens": 11748632.0,
      "step": 6390
    },
    {
      "entropy": 5.310668516159057,
      "epoch": 0.5341630471099231,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004977118267957204,
      "loss": 5.1149,
      "mean_token_accuracy": 0.19112634360790254,
      "num_tokens": 11756707.0,
      "step": 6395
    },
    {
      "entropy": 5.355916213989258,
      "epoch": 0.5345806882726362,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004977075899798513,
      "loss": 5.2832,
      "mean_token_accuracy": 0.18032287508249284,
      "num_tokens": 11765346.0,
      "step": 6400
    },
    {
      "entropy": 5.340433788299561,
      "epoch": 0.5349983294353492,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000497703349265218,
      "loss": 5.2021,
      "mean_token_accuracy": 0.1888291731476784,
      "num_tokens": 11773663.0,
      "step": 6405
    },
    {
      "entropy": 5.372768115997315,
      "epoch": 0.5354159705980621,
      "grad_norm": 1.125,
      "learning_rate": 0.0004976991046518943,
      "loss": 5.3208,
      "mean_token_accuracy": 0.1827616810798645,
      "num_tokens": 11783782.0,
      "step": 6410
    },
    {
      "entropy": 5.398755311965942,
      "epoch": 0.5358336117607752,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004976948561399549,
      "loss": 5.2534,
      "mean_token_accuracy": 0.18580945432186127,
      "num_tokens": 11792241.0,
      "step": 6415
    },
    {
      "entropy": 5.3338744163513185,
      "epoch": 0.5362512529234882,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004976906037294738,
      "loss": 5.3103,
      "mean_token_accuracy": 0.17886969894170762,
      "num_tokens": 11801576.0,
      "step": 6420
    },
    {
      "entropy": 5.31033148765564,
      "epoch": 0.5366688940862011,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004976863474205257,
      "loss": 5.1901,
      "mean_token_accuracy": 0.188408163189888,
      "num_tokens": 11810710.0,
      "step": 6425
    },
    {
      "entropy": 5.397537183761597,
      "epoch": 0.5370865352489141,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004976820872131851,
      "loss": 5.3281,
      "mean_token_accuracy": 0.17591421604156493,
      "num_tokens": 11820099.0,
      "step": 6430
    },
    {
      "entropy": 5.3122954845428465,
      "epoch": 0.5375041764116272,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004976778231075266,
      "loss": 5.1716,
      "mean_token_accuracy": 0.18763360679149627,
      "num_tokens": 11828846.0,
      "step": 6435
    },
    {
      "entropy": 5.272968387603759,
      "epoch": 0.5379218175743401,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976735551036245,
      "loss": 5.1926,
      "mean_token_accuracy": 0.1852931261062622,
      "num_tokens": 11837935.0,
      "step": 6440
    },
    {
      "entropy": 5.34056282043457,
      "epoch": 0.5383394587370531,
      "grad_norm": 1.140625,
      "learning_rate": 0.000497669283201554,
      "loss": 5.1929,
      "mean_token_accuracy": 0.1879839390516281,
      "num_tokens": 11846984.0,
      "step": 6445
    },
    {
      "entropy": 5.253386402130127,
      "epoch": 0.5387570998997662,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004976650074013895,
      "loss": 5.1703,
      "mean_token_accuracy": 0.18641676306724547,
      "num_tokens": 11856114.0,
      "step": 6450
    },
    {
      "entropy": 5.310311222076416,
      "epoch": 0.5391747410624791,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004976607277032061,
      "loss": 5.3302,
      "mean_token_accuracy": 0.18253034204244614,
      "num_tokens": 11866623.0,
      "step": 6455
    },
    {
      "entropy": 5.37350287437439,
      "epoch": 0.5395923822251921,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004976564441070788,
      "loss": 5.2374,
      "mean_token_accuracy": 0.18385156393051147,
      "num_tokens": 11876445.0,
      "step": 6460
    },
    {
      "entropy": 5.384360313415527,
      "epoch": 0.5400100233879052,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976521566130823,
      "loss": 5.267,
      "mean_token_accuracy": 0.17601316720247268,
      "num_tokens": 11885411.0,
      "step": 6465
    },
    {
      "entropy": 5.3778462409973145,
      "epoch": 0.5404276645506181,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004976478652212918,
      "loss": 5.2688,
      "mean_token_accuracy": 0.18532623797655107,
      "num_tokens": 11894803.0,
      "step": 6470
    },
    {
      "entropy": 5.297641801834106,
      "epoch": 0.5408453057133311,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004976435699317824,
      "loss": 5.2744,
      "mean_token_accuracy": 0.18821530789136887,
      "num_tokens": 11904299.0,
      "step": 6475
    },
    {
      "entropy": 5.235662508010864,
      "epoch": 0.5412629468760441,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004976392707446293,
      "loss": 5.1785,
      "mean_token_accuracy": 0.1875820815563202,
      "num_tokens": 11914197.0,
      "step": 6480
    },
    {
      "entropy": 5.400727891921997,
      "epoch": 0.5416805880387571,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976349676599078,
      "loss": 5.3529,
      "mean_token_accuracy": 0.1850871503353119,
      "num_tokens": 11923610.0,
      "step": 6485
    },
    {
      "entropy": 5.395964241027832,
      "epoch": 0.5420982292014701,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976306606776932,
      "loss": 5.2011,
      "mean_token_accuracy": 0.18822776824235915,
      "num_tokens": 11932896.0,
      "step": 6490
    },
    {
      "entropy": 5.292629766464233,
      "epoch": 0.5425158703641831,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004976263497980609,
      "loss": 5.2748,
      "mean_token_accuracy": 0.1863493502140045,
      "num_tokens": 11941343.0,
      "step": 6495
    },
    {
      "entropy": 5.30104546546936,
      "epoch": 0.5429335115268961,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004976220350210864,
      "loss": 5.2425,
      "mean_token_accuracy": 0.1853145882487297,
      "num_tokens": 11950412.0,
      "step": 6500
    },
    {
      "entropy": 5.289846134185791,
      "epoch": 0.5433511526896091,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004976177163468451,
      "loss": 5.1474,
      "mean_token_accuracy": 0.19622760713100434,
      "num_tokens": 11959021.0,
      "step": 6505
    },
    {
      "entropy": 5.367030763626099,
      "epoch": 0.5437687938523221,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004976133937754128,
      "loss": 5.2782,
      "mean_token_accuracy": 0.18347511291503907,
      "num_tokens": 11968725.0,
      "step": 6510
    },
    {
      "entropy": 5.335864305496216,
      "epoch": 0.544186435015035,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004976090673068651,
      "loss": 5.2373,
      "mean_token_accuracy": 0.17939867675304413,
      "num_tokens": 11977775.0,
      "step": 6515
    },
    {
      "entropy": 5.379273796081543,
      "epoch": 0.5446040761777481,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004976047369412778,
      "loss": 5.3014,
      "mean_token_accuracy": 0.180289189517498,
      "num_tokens": 11988010.0,
      "step": 6520
    },
    {
      "entropy": 5.261785268783569,
      "epoch": 0.5450217173404611,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004976004026787266,
      "loss": 5.163,
      "mean_token_accuracy": 0.1917942926287651,
      "num_tokens": 11997284.0,
      "step": 6525
    },
    {
      "entropy": 5.341743993759155,
      "epoch": 0.545439358503174,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004975960645192873,
      "loss": 5.2673,
      "mean_token_accuracy": 0.183339224755764,
      "num_tokens": 12005516.0,
      "step": 6530
    },
    {
      "entropy": 5.3298993587493895,
      "epoch": 0.5458569996658871,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004975917224630359,
      "loss": 5.1775,
      "mean_token_accuracy": 0.186410990357399,
      "num_tokens": 12014236.0,
      "step": 6535
    },
    {
      "entropy": 5.3112791061401365,
      "epoch": 0.5462746408286001,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004975873765100487,
      "loss": 5.2576,
      "mean_token_accuracy": 0.18704894930124283,
      "num_tokens": 12022201.0,
      "step": 6540
    },
    {
      "entropy": 5.334233808517456,
      "epoch": 0.546692281991313,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004975830266604013,
      "loss": 5.3534,
      "mean_token_accuracy": 0.1755892276763916,
      "num_tokens": 12032421.0,
      "step": 6545
    },
    {
      "entropy": 5.419464063644409,
      "epoch": 0.5471099231540261,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004975786729141702,
      "loss": 5.2537,
      "mean_token_accuracy": 0.18279391378164292,
      "num_tokens": 12041443.0,
      "step": 6550
    },
    {
      "entropy": 5.299913644790649,
      "epoch": 0.5475275643167391,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004975743152714314,
      "loss": 5.1747,
      "mean_token_accuracy": 0.18709085285663604,
      "num_tokens": 12050753.0,
      "step": 6555
    },
    {
      "entropy": 5.299720525741577,
      "epoch": 0.547945205479452,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004975699537322616,
      "loss": 5.2164,
      "mean_token_accuracy": 0.184294630587101,
      "num_tokens": 12059886.0,
      "step": 6560
    },
    {
      "entropy": 5.289112997055054,
      "epoch": 0.5483628466421651,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004975655882967366,
      "loss": 5.2209,
      "mean_token_accuracy": 0.1795896425843239,
      "num_tokens": 12068695.0,
      "step": 6565
    },
    {
      "entropy": 5.289681816101075,
      "epoch": 0.5487804878048781,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004975612189649331,
      "loss": 5.1453,
      "mean_token_accuracy": 0.18942635357379914,
      "num_tokens": 12077552.0,
      "step": 6570
    },
    {
      "entropy": 5.463022994995117,
      "epoch": 0.549198128967591,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004975568457369276,
      "loss": 5.3123,
      "mean_token_accuracy": 0.18367442786693572,
      "num_tokens": 12085994.0,
      "step": 6575
    },
    {
      "entropy": 5.388846778869629,
      "epoch": 0.549615770130304,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004975524686127966,
      "loss": 5.1717,
      "mean_token_accuracy": 0.1829482540488243,
      "num_tokens": 12095342.0,
      "step": 6580
    },
    {
      "entropy": 5.213991975784301,
      "epoch": 0.5500334112930171,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004975480875926168,
      "loss": 5.2461,
      "mean_token_accuracy": 0.18362866789102555,
      "num_tokens": 12104578.0,
      "step": 6585
    },
    {
      "entropy": 5.146052742004395,
      "epoch": 0.55045105245573,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004975437026764648,
      "loss": 5.0599,
      "mean_token_accuracy": 0.19359901547431946,
      "num_tokens": 12113614.0,
      "step": 6590
    },
    {
      "entropy": 5.4128223896026615,
      "epoch": 0.550868693618443,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004975393138644175,
      "loss": 5.3453,
      "mean_token_accuracy": 0.18435679525136947,
      "num_tokens": 12123328.0,
      "step": 6595
    },
    {
      "entropy": 5.384577131271362,
      "epoch": 0.5512863347811561,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004975349211565515,
      "loss": 5.2947,
      "mean_token_accuracy": 0.1866936355829239,
      "num_tokens": 12133011.0,
      "step": 6600
    },
    {
      "entropy": 5.469799757003784,
      "epoch": 0.551703975943869,
      "grad_norm": 1.03125,
      "learning_rate": 0.000497530524552944,
      "loss": 5.413,
      "mean_token_accuracy": 0.17317648231983185,
      "num_tokens": 12142398.0,
      "step": 6605
    },
    {
      "entropy": 5.417175436019898,
      "epoch": 0.552121617106582,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004975261240536716,
      "loss": 5.4295,
      "mean_token_accuracy": 0.16581220924854279,
      "num_tokens": 12152210.0,
      "step": 6610
    },
    {
      "entropy": 5.388317823410034,
      "epoch": 0.5525392582692951,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004975217196588117,
      "loss": 5.2156,
      "mean_token_accuracy": 0.1858538344502449,
      "num_tokens": 12162083.0,
      "step": 6615
    },
    {
      "entropy": 5.433413791656494,
      "epoch": 0.552956899432008,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004975173113684413,
      "loss": 5.2999,
      "mean_token_accuracy": 0.1742522731423378,
      "num_tokens": 12172024.0,
      "step": 6620
    },
    {
      "entropy": 5.428104543685913,
      "epoch": 0.553374540594721,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004975128991826374,
      "loss": 5.3976,
      "mean_token_accuracy": 0.17468141615390778,
      "num_tokens": 12181474.0,
      "step": 6625
    },
    {
      "entropy": 5.425901985168457,
      "epoch": 0.553792181757434,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004975084831014774,
      "loss": 5.3075,
      "mean_token_accuracy": 0.1809348151087761,
      "num_tokens": 12190622.0,
      "step": 6630
    },
    {
      "entropy": 5.315758991241455,
      "epoch": 0.554209822920147,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004975040631250386,
      "loss": 5.3024,
      "mean_token_accuracy": 0.18531003296375276,
      "num_tokens": 12199282.0,
      "step": 6635
    },
    {
      "entropy": 5.305345439910889,
      "epoch": 0.55462746408286,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004974996392533983,
      "loss": 5.216,
      "mean_token_accuracy": 0.18667261004447938,
      "num_tokens": 12208536.0,
      "step": 6640
    },
    {
      "entropy": 5.39209303855896,
      "epoch": 0.555045105245573,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497495211486634,
      "loss": 5.2755,
      "mean_token_accuracy": 0.1802435338497162,
      "num_tokens": 12217716.0,
      "step": 6645
    },
    {
      "entropy": 5.259994697570801,
      "epoch": 0.555462746408286,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004974907798248232,
      "loss": 5.1512,
      "mean_token_accuracy": 0.18910439759492875,
      "num_tokens": 12226214.0,
      "step": 6650
    },
    {
      "entropy": 5.279370307922363,
      "epoch": 0.555880387570999,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004974863442680434,
      "loss": 5.1201,
      "mean_token_accuracy": 0.19275906234979628,
      "num_tokens": 12235005.0,
      "step": 6655
    },
    {
      "entropy": 5.25799880027771,
      "epoch": 0.556298028733712,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004974819048163725,
      "loss": 5.2161,
      "mean_token_accuracy": 0.18881827145814895,
      "num_tokens": 12243085.0,
      "step": 6660
    },
    {
      "entropy": 5.421213436126709,
      "epoch": 0.556715669896425,
      "grad_norm": 1.09375,
      "learning_rate": 0.000497477461469888,
      "loss": 5.3475,
      "mean_token_accuracy": 0.17473243325948715,
      "num_tokens": 12252826.0,
      "step": 6665
    },
    {
      "entropy": 5.40093355178833,
      "epoch": 0.557133311059138,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004974730142286677,
      "loss": 5.2829,
      "mean_token_accuracy": 0.17674487233161926,
      "num_tokens": 12261166.0,
      "step": 6670
    },
    {
      "entropy": 5.298817110061646,
      "epoch": 0.557550952221851,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004974685630927895,
      "loss": 5.2826,
      "mean_token_accuracy": 0.18788901716470718,
      "num_tokens": 12270695.0,
      "step": 6675
    },
    {
      "entropy": 5.434196329116821,
      "epoch": 0.557968593384564,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004974641080623314,
      "loss": 5.3661,
      "mean_token_accuracy": 0.1773197114467621,
      "num_tokens": 12280486.0,
      "step": 6680
    },
    {
      "entropy": 5.424243497848511,
      "epoch": 0.558386234547277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004974596491373711,
      "loss": 5.2092,
      "mean_token_accuracy": 0.1849601835012436,
      "num_tokens": 12289246.0,
      "step": 6685
    },
    {
      "entropy": 5.122715139389038,
      "epoch": 0.55880387570999,
      "grad_norm": 1.25,
      "learning_rate": 0.0004974551863179871,
      "loss": 5.1871,
      "mean_token_accuracy": 0.18086159825325013,
      "num_tokens": 12297562.0,
      "step": 6690
    },
    {
      "entropy": 5.3659802913665775,
      "epoch": 0.5592215168727029,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004974507196042572,
      "loss": 5.2891,
      "mean_token_accuracy": 0.1803388088941574,
      "num_tokens": 12307647.0,
      "step": 6695
    },
    {
      "entropy": 5.42486891746521,
      "epoch": 0.559639158035416,
      "grad_norm": 1.125,
      "learning_rate": 0.0004974462489962597,
      "loss": 5.282,
      "mean_token_accuracy": 0.18235939145088195,
      "num_tokens": 12317840.0,
      "step": 6700
    },
    {
      "entropy": 5.329939699172973,
      "epoch": 0.560056799198129,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004974417744940729,
      "loss": 5.2677,
      "mean_token_accuracy": 0.18844126164913177,
      "num_tokens": 12327340.0,
      "step": 6705
    },
    {
      "entropy": 5.260721683502197,
      "epoch": 0.5604744403608419,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000497437296097775,
      "loss": 5.2154,
      "mean_token_accuracy": 0.18876503109931947,
      "num_tokens": 12336269.0,
      "step": 6710
    },
    {
      "entropy": 5.442825603485107,
      "epoch": 0.560892081523555,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004974328138074445,
      "loss": 5.331,
      "mean_token_accuracy": 0.17961659133434296,
      "num_tokens": 12345197.0,
      "step": 6715
    },
    {
      "entropy": 5.43170132637024,
      "epoch": 0.561309722686268,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004974283276231599,
      "loss": 5.2684,
      "mean_token_accuracy": 0.183046917617321,
      "num_tokens": 12355312.0,
      "step": 6720
    },
    {
      "entropy": 5.3082623958587645,
      "epoch": 0.5617273638489809,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004974238375449996,
      "loss": 5.2583,
      "mean_token_accuracy": 0.18221138715744017,
      "num_tokens": 12364354.0,
      "step": 6725
    },
    {
      "entropy": 5.375804042816162,
      "epoch": 0.562145005011694,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004974193435730424,
      "loss": 5.3878,
      "mean_token_accuracy": 0.17645538002252578,
      "num_tokens": 12373512.0,
      "step": 6730
    },
    {
      "entropy": 5.373493480682373,
      "epoch": 0.562562646174407,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004974148457073668,
      "loss": 5.3206,
      "mean_token_accuracy": 0.1795360788702965,
      "num_tokens": 12383087.0,
      "step": 6735
    },
    {
      "entropy": 5.34245867729187,
      "epoch": 0.5629802873371199,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004974103439480517,
      "loss": 5.1756,
      "mean_token_accuracy": 0.19321381896734238,
      "num_tokens": 12392097.0,
      "step": 6740
    },
    {
      "entropy": 5.346031045913696,
      "epoch": 0.563397928499833,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004974058382951757,
      "loss": 5.2188,
      "mean_token_accuracy": 0.1921417385339737,
      "num_tokens": 12401806.0,
      "step": 6745
    },
    {
      "entropy": 5.3388477802276615,
      "epoch": 0.563815569662546,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004974013287488177,
      "loss": 5.2208,
      "mean_token_accuracy": 0.1802361339330673,
      "num_tokens": 12412111.0,
      "step": 6750
    },
    {
      "entropy": 5.190551090240478,
      "epoch": 0.5642332108252589,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004973968153090569,
      "loss": 5.0697,
      "mean_token_accuracy": 0.19084409177303313,
      "num_tokens": 12421451.0,
      "step": 6755
    },
    {
      "entropy": 5.178379869461059,
      "epoch": 0.5646508519879719,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004973922979759722,
      "loss": 5.0914,
      "mean_token_accuracy": 0.19183809757232667,
      "num_tokens": 12430535.0,
      "step": 6760
    },
    {
      "entropy": 5.332199239730835,
      "epoch": 0.565068493150685,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004973877767496424,
      "loss": 5.2687,
      "mean_token_accuracy": 0.17987142205238343,
      "num_tokens": 12439727.0,
      "step": 6765
    },
    {
      "entropy": 5.314920043945312,
      "epoch": 0.5654861343133979,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004973832516301472,
      "loss": 5.1334,
      "mean_token_accuracy": 0.19156450480222703,
      "num_tokens": 12448889.0,
      "step": 6770
    },
    {
      "entropy": 5.208768129348755,
      "epoch": 0.5659037754761109,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004973787226175652,
      "loss": 5.1141,
      "mean_token_accuracy": 0.19008472561836243,
      "num_tokens": 12457529.0,
      "step": 6775
    },
    {
      "entropy": 5.298174905776977,
      "epoch": 0.566321416638824,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004973741897119761,
      "loss": 5.2319,
      "mean_token_accuracy": 0.186704021692276,
      "num_tokens": 12466379.0,
      "step": 6780
    },
    {
      "entropy": 5.298062181472778,
      "epoch": 0.5667390578015369,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004973696529134592,
      "loss": 5.2886,
      "mean_token_accuracy": 0.18568971008062363,
      "num_tokens": 12475317.0,
      "step": 6785
    },
    {
      "entropy": 5.388056898117066,
      "epoch": 0.5671566989642499,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004973651122220938,
      "loss": 5.3005,
      "mean_token_accuracy": 0.17674677073955536,
      "num_tokens": 12483882.0,
      "step": 6790
    },
    {
      "entropy": 5.393543100357055,
      "epoch": 0.567574340126963,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004973605676379595,
      "loss": 5.2769,
      "mean_token_accuracy": 0.1825995609164238,
      "num_tokens": 12493782.0,
      "step": 6795
    },
    {
      "entropy": 5.22726674079895,
      "epoch": 0.5679919812896759,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004973560191611357,
      "loss": 5.0696,
      "mean_token_accuracy": 0.19600007981061934,
      "num_tokens": 12501940.0,
      "step": 6800
    },
    {
      "entropy": 5.2739979267120365,
      "epoch": 0.5684096224523889,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004973514667917021,
      "loss": 5.2244,
      "mean_token_accuracy": 0.1818484529852867,
      "num_tokens": 12511566.0,
      "step": 6805
    },
    {
      "entropy": 5.299200105667114,
      "epoch": 0.5688272636151019,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004973469105297386,
      "loss": 5.2246,
      "mean_token_accuracy": 0.19029798060655595,
      "num_tokens": 12519902.0,
      "step": 6810
    },
    {
      "entropy": 5.405635023117066,
      "epoch": 0.5692449047778149,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004973423503753247,
      "loss": 5.3314,
      "mean_token_accuracy": 0.17589733600616456,
      "num_tokens": 12529011.0,
      "step": 6815
    },
    {
      "entropy": 5.315787696838379,
      "epoch": 0.5696625459405279,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004973377863285403,
      "loss": 5.2656,
      "mean_token_accuracy": 0.1805856019258499,
      "num_tokens": 12538181.0,
      "step": 6820
    },
    {
      "entropy": 5.35498456954956,
      "epoch": 0.5700801871032409,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004973332183894653,
      "loss": 5.231,
      "mean_token_accuracy": 0.17893521636724471,
      "num_tokens": 12545935.0,
      "step": 6825
    },
    {
      "entropy": 5.29422960281372,
      "epoch": 0.5704978282659539,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004973286465581798,
      "loss": 5.2257,
      "mean_token_accuracy": 0.1820117339491844,
      "num_tokens": 12555586.0,
      "step": 6830
    },
    {
      "entropy": 5.323349094390869,
      "epoch": 0.5709154694286669,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004973240708347637,
      "loss": 5.2415,
      "mean_token_accuracy": 0.18298058658838273,
      "num_tokens": 12564478.0,
      "step": 6835
    },
    {
      "entropy": 5.304536437988281,
      "epoch": 0.5713331105913799,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004973194912192971,
      "loss": 5.1674,
      "mean_token_accuracy": 0.1874542161822319,
      "num_tokens": 12573769.0,
      "step": 6840
    },
    {
      "entropy": 5.286143493652344,
      "epoch": 0.5717507517540928,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004973149077118602,
      "loss": 5.2004,
      "mean_token_accuracy": 0.189128577709198,
      "num_tokens": 12583291.0,
      "step": 6845
    },
    {
      "entropy": 5.273022413253784,
      "epoch": 0.5721683929168059,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004973103203125333,
      "loss": 5.1564,
      "mean_token_accuracy": 0.19136483520269393,
      "num_tokens": 12591828.0,
      "step": 6850
    },
    {
      "entropy": 5.224130153656006,
      "epoch": 0.5725860340795189,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004973057290213965,
      "loss": 5.2357,
      "mean_token_accuracy": 0.18667172342538835,
      "num_tokens": 12600417.0,
      "step": 6855
    },
    {
      "entropy": 5.288603210449219,
      "epoch": 0.5730036752422318,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004973011338385304,
      "loss": 5.2072,
      "mean_token_accuracy": 0.19128816276788713,
      "num_tokens": 12609555.0,
      "step": 6860
    },
    {
      "entropy": 5.465093469619751,
      "epoch": 0.5734213164049449,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004972965347640154,
      "loss": 5.344,
      "mean_token_accuracy": 0.1806751385331154,
      "num_tokens": 12618666.0,
      "step": 6865
    },
    {
      "entropy": 5.46686863899231,
      "epoch": 0.5738389575676579,
      "grad_norm": 1.09375,
      "learning_rate": 0.000497291931797932,
      "loss": 5.3224,
      "mean_token_accuracy": 0.1820882961153984,
      "num_tokens": 12627206.0,
      "step": 6870
    },
    {
      "entropy": 5.374486637115479,
      "epoch": 0.5742565987303708,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004972873249403608,
      "loss": 5.27,
      "mean_token_accuracy": 0.18061840683221816,
      "num_tokens": 12636782.0,
      "step": 6875
    },
    {
      "entropy": 5.309173727035523,
      "epoch": 0.5746742398930839,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004972827141913822,
      "loss": 5.2141,
      "mean_token_accuracy": 0.1896350860595703,
      "num_tokens": 12645895.0,
      "step": 6880
    },
    {
      "entropy": 5.220979022979736,
      "epoch": 0.5750918810557969,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004972780995510773,
      "loss": 5.1448,
      "mean_token_accuracy": 0.1859116241335869,
      "num_tokens": 12655744.0,
      "step": 6885
    },
    {
      "entropy": 5.219499254226685,
      "epoch": 0.5755095222185098,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004972734810195266,
      "loss": 5.1873,
      "mean_token_accuracy": 0.19034142792224884,
      "num_tokens": 12665498.0,
      "step": 6890
    },
    {
      "entropy": 5.347544240951538,
      "epoch": 0.5759271633812229,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004972688585968112,
      "loss": 5.1988,
      "mean_token_accuracy": 0.18442365378141404,
      "num_tokens": 12674218.0,
      "step": 6895
    },
    {
      "entropy": 5.3160968780517575,
      "epoch": 0.5763448045439359,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004972642322830118,
      "loss": 5.2868,
      "mean_token_accuracy": 0.1894842803478241,
      "num_tokens": 12683060.0,
      "step": 6900
    },
    {
      "entropy": 5.311458444595337,
      "epoch": 0.5767624457066488,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004972596020782096,
      "loss": 5.2216,
      "mean_token_accuracy": 0.18776786774396897,
      "num_tokens": 12693145.0,
      "step": 6905
    },
    {
      "entropy": 5.243571090698242,
      "epoch": 0.5771800868693618,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004972549679824853,
      "loss": 5.1923,
      "mean_token_accuracy": 0.17975606322288512,
      "num_tokens": 12701439.0,
      "step": 6910
    },
    {
      "entropy": 5.400462102890015,
      "epoch": 0.5775977280320749,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004972503299959204,
      "loss": 5.3282,
      "mean_token_accuracy": 0.17578370869159698,
      "num_tokens": 12710582.0,
      "step": 6915
    },
    {
      "entropy": 5.386904525756836,
      "epoch": 0.5780153691947878,
      "grad_norm": 1.109375,
      "learning_rate": 0.000497245688118596,
      "loss": 5.1998,
      "mean_token_accuracy": 0.184463831782341,
      "num_tokens": 12719685.0,
      "step": 6920
    },
    {
      "entropy": 5.32031364440918,
      "epoch": 0.5784330103575008,
      "grad_norm": 1.125,
      "learning_rate": 0.0004972410423505932,
      "loss": 5.2624,
      "mean_token_accuracy": 0.18566998988389968,
      "num_tokens": 12729632.0,
      "step": 6925
    },
    {
      "entropy": 5.217878770828247,
      "epoch": 0.5788506515202139,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004972363926919935,
      "loss": 5.1036,
      "mean_token_accuracy": 0.19576410204172134,
      "num_tokens": 12738900.0,
      "step": 6930
    },
    {
      "entropy": 5.214184093475342,
      "epoch": 0.5792682926829268,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004972317391428782,
      "loss": 5.0763,
      "mean_token_accuracy": 0.19970481544733049,
      "num_tokens": 12748195.0,
      "step": 6935
    },
    {
      "entropy": 5.331734228134155,
      "epoch": 0.5796859338456398,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004972270817033288,
      "loss": 5.2268,
      "mean_token_accuracy": 0.18134150207042693,
      "num_tokens": 12757038.0,
      "step": 6940
    },
    {
      "entropy": 5.19258189201355,
      "epoch": 0.5801035750083529,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004972224203734269,
      "loss": 5.1753,
      "mean_token_accuracy": 0.19638993591070175,
      "num_tokens": 12765871.0,
      "step": 6945
    },
    {
      "entropy": 5.282092189788818,
      "epoch": 0.5805212161710658,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497217755153254,
      "loss": 5.1589,
      "mean_token_accuracy": 0.19112114757299423,
      "num_tokens": 12774776.0,
      "step": 6950
    },
    {
      "entropy": 5.372214317321777,
      "epoch": 0.5809388573337788,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004972130860428918,
      "loss": 5.2495,
      "mean_token_accuracy": 0.18278664201498032,
      "num_tokens": 12783815.0,
      "step": 6955
    },
    {
      "entropy": 5.4111034870147705,
      "epoch": 0.5813564984964918,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004972084130424222,
      "loss": 5.29,
      "mean_token_accuracy": 0.17568725496530532,
      "num_tokens": 12793690.0,
      "step": 6960
    },
    {
      "entropy": 5.3078378200531,
      "epoch": 0.5817741396592048,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004972037361519267,
      "loss": 5.2524,
      "mean_token_accuracy": 0.18332116156816483,
      "num_tokens": 12802525.0,
      "step": 6965
    },
    {
      "entropy": 5.360084676742554,
      "epoch": 0.5821917808219178,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004971990553714874,
      "loss": 5.2274,
      "mean_token_accuracy": 0.18500502705574035,
      "num_tokens": 12811246.0,
      "step": 6970
    },
    {
      "entropy": 5.355574703216552,
      "epoch": 0.5826094219846308,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004971943707011861,
      "loss": 5.2554,
      "mean_token_accuracy": 0.1834030732512474,
      "num_tokens": 12820477.0,
      "step": 6975
    },
    {
      "entropy": 5.310442113876343,
      "epoch": 0.5830270631473438,
      "grad_norm": 1.15625,
      "learning_rate": 0.000497189682141105,
      "loss": 5.1185,
      "mean_token_accuracy": 0.19551344811916352,
      "num_tokens": 12829544.0,
      "step": 6980
    },
    {
      "entropy": 5.269707918167114,
      "epoch": 0.5834447043100568,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004971849896913259,
      "loss": 5.2829,
      "mean_token_accuracy": 0.1951313778758049,
      "num_tokens": 12839388.0,
      "step": 6985
    },
    {
      "entropy": 5.24746265411377,
      "epoch": 0.5838623454727698,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004971802933519312,
      "loss": 5.0883,
      "mean_token_accuracy": 0.1925328031182289,
      "num_tokens": 12848191.0,
      "step": 6990
    },
    {
      "entropy": 5.24632158279419,
      "epoch": 0.5842799866354828,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004971755931230031,
      "loss": 5.2668,
      "mean_token_accuracy": 0.18232031911611557,
      "num_tokens": 12859239.0,
      "step": 6995
    },
    {
      "entropy": 5.420159435272216,
      "epoch": 0.5846976277981958,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004971708890046236,
      "loss": 5.3088,
      "mean_token_accuracy": 0.18971134573221207,
      "num_tokens": 12868335.0,
      "step": 7000
    },
    {
      "entropy": 5.269492864608765,
      "epoch": 0.5851152689609088,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004971661809968755,
      "loss": 5.1487,
      "mean_token_accuracy": 0.1872716426849365,
      "num_tokens": 12876939.0,
      "step": 7005
    },
    {
      "entropy": 5.19940676689148,
      "epoch": 0.5855329101236217,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004971614690998409,
      "loss": 5.1511,
      "mean_token_accuracy": 0.1863880440592766,
      "num_tokens": 12885856.0,
      "step": 7010
    },
    {
      "entropy": 5.435565090179443,
      "epoch": 0.5859505512863348,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004971567533136024,
      "loss": 5.3365,
      "mean_token_accuracy": 0.17623012959957124,
      "num_tokens": 12895451.0,
      "step": 7015
    },
    {
      "entropy": 5.384980773925781,
      "epoch": 0.5863681924490478,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004971520336382424,
      "loss": 5.1984,
      "mean_token_accuracy": 0.18783926218748093,
      "num_tokens": 12904796.0,
      "step": 7020
    },
    {
      "entropy": 5.320826721191406,
      "epoch": 0.5867858336117607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004971473100738437,
      "loss": 5.3358,
      "mean_token_accuracy": 0.17882730811834335,
      "num_tokens": 12913668.0,
      "step": 7025
    },
    {
      "entropy": 5.3277966499328615,
      "epoch": 0.5872034747744738,
      "grad_norm": 1.1875,
      "learning_rate": 0.000497142582620489,
      "loss": 5.1788,
      "mean_token_accuracy": 0.18854000717401503,
      "num_tokens": 12922469.0,
      "step": 7030
    },
    {
      "entropy": 5.193450784683227,
      "epoch": 0.5876211159371868,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004971378512782609,
      "loss": 5.1896,
      "mean_token_accuracy": 0.18770741820335388,
      "num_tokens": 12931070.0,
      "step": 7035
    },
    {
      "entropy": 5.296234560012818,
      "epoch": 0.5880387570998997,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004971331160472424,
      "loss": 5.2357,
      "mean_token_accuracy": 0.19035572558641434,
      "num_tokens": 12939397.0,
      "step": 7040
    },
    {
      "entropy": 5.422897005081177,
      "epoch": 0.5884563982626128,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004971283769275163,
      "loss": 5.3066,
      "mean_token_accuracy": 0.17893508523702623,
      "num_tokens": 12948951.0,
      "step": 7045
    },
    {
      "entropy": 5.32398829460144,
      "epoch": 0.5888740394253258,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004971236339191656,
      "loss": 5.2309,
      "mean_token_accuracy": 0.1897008717060089,
      "num_tokens": 12958431.0,
      "step": 7050
    },
    {
      "entropy": 5.408656358718872,
      "epoch": 0.5892916805880387,
      "grad_norm": 1.125,
      "learning_rate": 0.0004971188870222734,
      "loss": 5.2422,
      "mean_token_accuracy": 0.19711191207170486,
      "num_tokens": 12966521.0,
      "step": 7055
    },
    {
      "entropy": 5.28595757484436,
      "epoch": 0.5897093217507517,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004971141362369226,
      "loss": 5.2156,
      "mean_token_accuracy": 0.18485396653413771,
      "num_tokens": 12976458.0,
      "step": 7060
    },
    {
      "entropy": 5.255414915084839,
      "epoch": 0.5901269629134648,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004971093815631965,
      "loss": 5.1757,
      "mean_token_accuracy": 0.19233511239290238,
      "num_tokens": 12985715.0,
      "step": 7065
    },
    {
      "entropy": 5.2574474811553955,
      "epoch": 0.5905446040761777,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004971046230011783,
      "loss": 5.3092,
      "mean_token_accuracy": 0.18003213554620742,
      "num_tokens": 12995640.0,
      "step": 7070
    },
    {
      "entropy": 5.328100156784058,
      "epoch": 0.5909622452388907,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004970998605509514,
      "loss": 5.1575,
      "mean_token_accuracy": 0.18547921776771545,
      "num_tokens": 13004645.0,
      "step": 7075
    },
    {
      "entropy": 5.327472066879272,
      "epoch": 0.5913798864016038,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497095094212599,
      "loss": 5.2345,
      "mean_token_accuracy": 0.18197911977767944,
      "num_tokens": 13013433.0,
      "step": 7080
    },
    {
      "entropy": 5.399848365783692,
      "epoch": 0.5917975275643167,
      "grad_norm": 1.125,
      "learning_rate": 0.0004970903239862047,
      "loss": 5.3034,
      "mean_token_accuracy": 0.18082935661077498,
      "num_tokens": 13023098.0,
      "step": 7085
    },
    {
      "entropy": 5.346718597412109,
      "epoch": 0.5922151687270297,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004970855498718519,
      "loss": 5.1017,
      "mean_token_accuracy": 0.1969370573759079,
      "num_tokens": 13031715.0,
      "step": 7090
    },
    {
      "entropy": 5.270420169830322,
      "epoch": 0.5926328098897428,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004970807718696243,
      "loss": 5.1948,
      "mean_token_accuracy": 0.18444593250751495,
      "num_tokens": 13040847.0,
      "step": 7095
    },
    {
      "entropy": 5.278219938278198,
      "epoch": 0.5930504510524557,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004970759899796055,
      "loss": 5.2049,
      "mean_token_accuracy": 0.1857482075691223,
      "num_tokens": 13050502.0,
      "step": 7100
    },
    {
      "entropy": 5.30429048538208,
      "epoch": 0.5934680922151687,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004970712042018791,
      "loss": 5.16,
      "mean_token_accuracy": 0.18868696242570876,
      "num_tokens": 13059729.0,
      "step": 7105
    },
    {
      "entropy": 5.34501051902771,
      "epoch": 0.5938857333778818,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004970664145365289,
      "loss": 5.2663,
      "mean_token_accuracy": 0.1812531977891922,
      "num_tokens": 13069123.0,
      "step": 7110
    },
    {
      "entropy": 5.319238185882568,
      "epoch": 0.5943033745405947,
      "grad_norm": 1.125,
      "learning_rate": 0.000497061620983639,
      "loss": 5.1588,
      "mean_token_accuracy": 0.1846571832895279,
      "num_tokens": 13077777.0,
      "step": 7115
    },
    {
      "entropy": 5.386619281768799,
      "epoch": 0.5947210157033077,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004970568235432931,
      "loss": 5.3285,
      "mean_token_accuracy": 0.17931535094976425,
      "num_tokens": 13086576.0,
      "step": 7120
    },
    {
      "entropy": 5.321637153625488,
      "epoch": 0.5951386568660207,
      "grad_norm": 1.125,
      "learning_rate": 0.0004970520222155751,
      "loss": 5.1253,
      "mean_token_accuracy": 0.18916917145252227,
      "num_tokens": 13095166.0,
      "step": 7125
    },
    {
      "entropy": 5.322514963150025,
      "epoch": 0.5955562980287337,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004970472170005692,
      "loss": 5.2062,
      "mean_token_accuracy": 0.18322862684726715,
      "num_tokens": 13103692.0,
      "step": 7130
    },
    {
      "entropy": 5.261594581604004,
      "epoch": 0.5959739391914467,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004970424078983596,
      "loss": 5.1294,
      "mean_token_accuracy": 0.18912494629621507,
      "num_tokens": 13111899.0,
      "step": 7135
    },
    {
      "entropy": 5.301736068725586,
      "epoch": 0.5963915803541597,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004970375949090302,
      "loss": 5.1457,
      "mean_token_accuracy": 0.19724455028772353,
      "num_tokens": 13120345.0,
      "step": 7140
    },
    {
      "entropy": 5.303208208084106,
      "epoch": 0.5968092215168727,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004970327780326655,
      "loss": 5.1662,
      "mean_token_accuracy": 0.1907811388373375,
      "num_tokens": 13129035.0,
      "step": 7145
    },
    {
      "entropy": 5.288313198089599,
      "epoch": 0.5972268626795857,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004970279572693498,
      "loss": 5.202,
      "mean_token_accuracy": 0.18506258428096772,
      "num_tokens": 13138956.0,
      "step": 7150
    },
    {
      "entropy": 5.338060474395752,
      "epoch": 0.5976445038422987,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004970231326191674,
      "loss": 5.2023,
      "mean_token_accuracy": 0.18579529523849486,
      "num_tokens": 13148605.0,
      "step": 7155
    },
    {
      "entropy": 5.278689813613892,
      "epoch": 0.5980621450050116,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004970183040822029,
      "loss": 5.2392,
      "mean_token_accuracy": 0.18053558766841887,
      "num_tokens": 13158125.0,
      "step": 7160
    },
    {
      "entropy": 5.243731737136841,
      "epoch": 0.5984797861677247,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004970134716585407,
      "loss": 5.1428,
      "mean_token_accuracy": 0.1900031641125679,
      "num_tokens": 13167404.0,
      "step": 7165
    },
    {
      "entropy": 5.294559335708618,
      "epoch": 0.5988974273304377,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004970086353482654,
      "loss": 5.1731,
      "mean_token_accuracy": 0.19000328779220582,
      "num_tokens": 13176467.0,
      "step": 7170
    },
    {
      "entropy": 5.217110300064087,
      "epoch": 0.5993150684931506,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004970037951514617,
      "loss": 5.1097,
      "mean_token_accuracy": 0.19307963848114013,
      "num_tokens": 13185201.0,
      "step": 7175
    },
    {
      "entropy": 5.23073673248291,
      "epoch": 0.5997327096558637,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004969989510682143,
      "loss": 5.1817,
      "mean_token_accuracy": 0.19369222819805146,
      "num_tokens": 13194871.0,
      "step": 7180
    },
    {
      "entropy": 5.34356837272644,
      "epoch": 0.6001503508185767,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004969941030986082,
      "loss": 5.2411,
      "mean_token_accuracy": 0.18253419399261475,
      "num_tokens": 13205006.0,
      "step": 7185
    },
    {
      "entropy": 5.272913312911987,
      "epoch": 0.6005679919812896,
      "grad_norm": 1.109375,
      "learning_rate": 0.000496989251242728,
      "loss": 5.1738,
      "mean_token_accuracy": 0.18916500508785247,
      "num_tokens": 13214093.0,
      "step": 7190
    },
    {
      "entropy": 5.267279148101807,
      "epoch": 0.6009856331440027,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004969843955006588,
      "loss": 5.2125,
      "mean_token_accuracy": 0.19354642778635026,
      "num_tokens": 13223178.0,
      "step": 7195
    },
    {
      "entropy": 5.257905721664429,
      "epoch": 0.6014032743067157,
      "grad_norm": 1.125,
      "learning_rate": 0.0004969795358724855,
      "loss": 5.125,
      "mean_token_accuracy": 0.19590703397989273,
      "num_tokens": 13232357.0,
      "step": 7200
    },
    {
      "entropy": 5.307076072692871,
      "epoch": 0.6018209154694286,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004969746723582932,
      "loss": 5.2645,
      "mean_token_accuracy": 0.181342650949955,
      "num_tokens": 13242538.0,
      "step": 7205
    },
    {
      "entropy": 5.454485702514648,
      "epoch": 0.6022385566321417,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004969698049581672,
      "loss": 5.3288,
      "mean_token_accuracy": 0.17714773416519164,
      "num_tokens": 13252410.0,
      "step": 7210
    },
    {
      "entropy": 5.3239843368530275,
      "epoch": 0.6026561977948547,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004969649336721924,
      "loss": 5.2551,
      "mean_token_accuracy": 0.18909587562084199,
      "num_tokens": 13262291.0,
      "step": 7215
    },
    {
      "entropy": 5.291239643096924,
      "epoch": 0.6030738389575676,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004969600585004544,
      "loss": 5.3155,
      "mean_token_accuracy": 0.17985448390245437,
      "num_tokens": 13271952.0,
      "step": 7220
    },
    {
      "entropy": 5.377409601211548,
      "epoch": 0.6034914801202806,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004969551794430383,
      "loss": 5.2516,
      "mean_token_accuracy": 0.18127916753292084,
      "num_tokens": 13281898.0,
      "step": 7225
    },
    {
      "entropy": 5.3245580196380615,
      "epoch": 0.6039091212829937,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004969502965000297,
      "loss": 5.2475,
      "mean_token_accuracy": 0.18747837841510773,
      "num_tokens": 13290205.0,
      "step": 7230
    },
    {
      "entropy": 5.371450138092041,
      "epoch": 0.6043267624457066,
      "grad_norm": 1.234375,
      "learning_rate": 0.000496945409671514,
      "loss": 5.264,
      "mean_token_accuracy": 0.17748096883296965,
      "num_tokens": 13299420.0,
      "step": 7235
    },
    {
      "entropy": 5.299730682373047,
      "epoch": 0.6047444036084196,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004969405189575768,
      "loss": 5.1459,
      "mean_token_accuracy": 0.19273400604724883,
      "num_tokens": 13307931.0,
      "step": 7240
    },
    {
      "entropy": 5.288811588287354,
      "epoch": 0.6051620447711327,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004969356243583036,
      "loss": 5.1471,
      "mean_token_accuracy": 0.18483478277921678,
      "num_tokens": 13316646.0,
      "step": 7245
    },
    {
      "entropy": 5.221371078491211,
      "epoch": 0.6055796859338456,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004969307258737802,
      "loss": 5.2255,
      "mean_token_accuracy": 0.18441895246505738,
      "num_tokens": 13325564.0,
      "step": 7250
    },
    {
      "entropy": 5.361169385910034,
      "epoch": 0.6059973270965586,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004969258235040921,
      "loss": 5.224,
      "mean_token_accuracy": 0.18220609426498413,
      "num_tokens": 13335396.0,
      "step": 7255
    },
    {
      "entropy": 5.309096097946167,
      "epoch": 0.6064149682592717,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004969209172493255,
      "loss": 5.1604,
      "mean_token_accuracy": 0.18890001624822617,
      "num_tokens": 13344124.0,
      "step": 7260
    },
    {
      "entropy": 5.197010850906372,
      "epoch": 0.6068326094219846,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004969160071095661,
      "loss": 5.0791,
      "mean_token_accuracy": 0.19340737760066987,
      "num_tokens": 13353125.0,
      "step": 7265
    },
    {
      "entropy": 5.3060260772705075,
      "epoch": 0.6072502505846976,
      "grad_norm": 1.125,
      "learning_rate": 0.0004969110930848999,
      "loss": 5.1914,
      "mean_token_accuracy": 0.18631464689970018,
      "num_tokens": 13361388.0,
      "step": 7270
    },
    {
      "entropy": 5.3180506229400635,
      "epoch": 0.6076678917474106,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004969061751754129,
      "loss": 5.3241,
      "mean_token_accuracy": 0.18138295859098436,
      "num_tokens": 13371061.0,
      "step": 7275
    },
    {
      "entropy": 5.2150026798248295,
      "epoch": 0.6080855329101236,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004969012533811911,
      "loss": 5.0505,
      "mean_token_accuracy": 0.1936565563082695,
      "num_tokens": 13380077.0,
      "step": 7280
    },
    {
      "entropy": 5.37867226600647,
      "epoch": 0.6085031740728366,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004968963277023208,
      "loss": 5.2728,
      "mean_token_accuracy": 0.18226235508918762,
      "num_tokens": 13390091.0,
      "step": 7285
    },
    {
      "entropy": 5.338456916809082,
      "epoch": 0.6089208152355496,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004968913981388882,
      "loss": 5.1096,
      "mean_token_accuracy": 0.19249432682991027,
      "num_tokens": 13399793.0,
      "step": 7290
    },
    {
      "entropy": 5.247054529190064,
      "epoch": 0.6093384563982626,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004968864646909797,
      "loss": 5.228,
      "mean_token_accuracy": 0.19297747313976288,
      "num_tokens": 13408771.0,
      "step": 7295
    },
    {
      "entropy": 5.349658203125,
      "epoch": 0.6097560975609756,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004968815273586814,
      "loss": 5.3082,
      "mean_token_accuracy": 0.179022678732872,
      "num_tokens": 13417569.0,
      "step": 7300
    },
    {
      "entropy": 5.310270833969116,
      "epoch": 0.6101737387236886,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004968765861420799,
      "loss": 5.1827,
      "mean_token_accuracy": 0.18481265157461166,
      "num_tokens": 13427296.0,
      "step": 7305
    },
    {
      "entropy": 5.387631130218506,
      "epoch": 0.6105913798864016,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004968716410412617,
      "loss": 5.2748,
      "mean_token_accuracy": 0.1780461221933365,
      "num_tokens": 13436489.0,
      "step": 7310
    },
    {
      "entropy": 5.335109424591065,
      "epoch": 0.6110090210491146,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004968666920563134,
      "loss": 5.2593,
      "mean_token_accuracy": 0.1812914326786995,
      "num_tokens": 13445783.0,
      "step": 7315
    },
    {
      "entropy": 5.333179616928101,
      "epoch": 0.6114266622118276,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004968617391873216,
      "loss": 5.2886,
      "mean_token_accuracy": 0.1859467312693596,
      "num_tokens": 13454862.0,
      "step": 7320
    },
    {
      "entropy": 5.3446472644805905,
      "epoch": 0.6118443033745405,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004968567824343729,
      "loss": 5.2537,
      "mean_token_accuracy": 0.18590015321969985,
      "num_tokens": 13463758.0,
      "step": 7325
    },
    {
      "entropy": 5.388862609863281,
      "epoch": 0.6122619445372536,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004968518217975542,
      "loss": 5.2573,
      "mean_token_accuracy": 0.18363479375839234,
      "num_tokens": 13472808.0,
      "step": 7330
    },
    {
      "entropy": 5.257959079742432,
      "epoch": 0.6126795856999666,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004968468572769524,
      "loss": 5.159,
      "mean_token_accuracy": 0.19224871397018434,
      "num_tokens": 13482550.0,
      "step": 7335
    },
    {
      "entropy": 5.274133014678955,
      "epoch": 0.6130972268626795,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004968418888726543,
      "loss": 5.2659,
      "mean_token_accuracy": 0.1815333530306816,
      "num_tokens": 13491970.0,
      "step": 7340
    },
    {
      "entropy": 5.294647169113159,
      "epoch": 0.6135148680253926,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004968369165847469,
      "loss": 5.1567,
      "mean_token_accuracy": 0.1917921334505081,
      "num_tokens": 13500699.0,
      "step": 7345
    },
    {
      "entropy": 5.257726860046387,
      "epoch": 0.6139325091881056,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004968319404133173,
      "loss": 5.0976,
      "mean_token_accuracy": 0.18986501544713974,
      "num_tokens": 13509108.0,
      "step": 7350
    },
    {
      "entropy": 5.279509687423706,
      "epoch": 0.6143501503508185,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004968269603584525,
      "loss": 5.1395,
      "mean_token_accuracy": 0.19250394254922867,
      "num_tokens": 13517529.0,
      "step": 7355
    },
    {
      "entropy": 5.177189016342163,
      "epoch": 0.6147677915135316,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004968219764202397,
      "loss": 5.0102,
      "mean_token_accuracy": 0.20139864683151246,
      "num_tokens": 13526379.0,
      "step": 7360
    },
    {
      "entropy": 5.291180562973023,
      "epoch": 0.6151854326762446,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004968169885987663,
      "loss": 5.2717,
      "mean_token_accuracy": 0.17230814099311828,
      "num_tokens": 13535873.0,
      "step": 7365
    },
    {
      "entropy": 5.307261562347412,
      "epoch": 0.6156030738389575,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004968119968941193,
      "loss": 5.2178,
      "mean_token_accuracy": 0.18773140162229537,
      "num_tokens": 13545514.0,
      "step": 7370
    },
    {
      "entropy": 5.291108083724976,
      "epoch": 0.6160207150016705,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004968070013063866,
      "loss": 5.1478,
      "mean_token_accuracy": 0.18375722020864488,
      "num_tokens": 13554226.0,
      "step": 7375
    },
    {
      "entropy": 5.328939390182495,
      "epoch": 0.6164383561643836,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004968020018356552,
      "loss": 5.2306,
      "mean_token_accuracy": 0.1805404931306839,
      "num_tokens": 13563554.0,
      "step": 7380
    },
    {
      "entropy": 5.196860980987549,
      "epoch": 0.6168559973270965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004967969984820128,
      "loss": 5.0066,
      "mean_token_accuracy": 0.20545026659965515,
      "num_tokens": 13572128.0,
      "step": 7385
    },
    {
      "entropy": 5.334772062301636,
      "epoch": 0.6172736384898095,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004967919912455469,
      "loss": 5.2553,
      "mean_token_accuracy": 0.18795283138751984,
      "num_tokens": 13582285.0,
      "step": 7390
    },
    {
      "entropy": 5.281139945983886,
      "epoch": 0.6176912796525226,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004967869801263454,
      "loss": 5.1325,
      "mean_token_accuracy": 0.1949436455965042,
      "num_tokens": 13590969.0,
      "step": 7395
    },
    {
      "entropy": 5.300789642333984,
      "epoch": 0.6181089208152355,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004967819651244957,
      "loss": 5.2401,
      "mean_token_accuracy": 0.1911582663655281,
      "num_tokens": 13600523.0,
      "step": 7400
    },
    {
      "entropy": 5.322118473052979,
      "epoch": 0.6185265619779485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004967769462400857,
      "loss": 5.1526,
      "mean_token_accuracy": 0.1932110920548439,
      "num_tokens": 13609203.0,
      "step": 7405
    },
    {
      "entropy": 5.232994413375854,
      "epoch": 0.6189442031406616,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004967719234732033,
      "loss": 5.1856,
      "mean_token_accuracy": 0.18712807595729827,
      "num_tokens": 13618175.0,
      "step": 7410
    },
    {
      "entropy": 5.3347086906433105,
      "epoch": 0.6193618443033745,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004967668968239365,
      "loss": 5.3388,
      "mean_token_accuracy": 0.18564439415931702,
      "num_tokens": 13627245.0,
      "step": 7415
    },
    {
      "entropy": 5.450446367263794,
      "epoch": 0.6197794854660875,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004967618662923732,
      "loss": 5.3435,
      "mean_token_accuracy": 0.18550013601779938,
      "num_tokens": 13636910.0,
      "step": 7420
    },
    {
      "entropy": 5.332142734527588,
      "epoch": 0.6201971266288006,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004967568318786016,
      "loss": 5.2331,
      "mean_token_accuracy": 0.18843764215707778,
      "num_tokens": 13646072.0,
      "step": 7425
    },
    {
      "entropy": 5.355451202392578,
      "epoch": 0.6206147677915135,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004967517935827097,
      "loss": 5.2408,
      "mean_token_accuracy": 0.18733312636613847,
      "num_tokens": 13655823.0,
      "step": 7430
    },
    {
      "entropy": 5.280789518356324,
      "epoch": 0.6210324089542265,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004967467514047856,
      "loss": 5.2718,
      "mean_token_accuracy": 0.18302804231643677,
      "num_tokens": 13665038.0,
      "step": 7435
    },
    {
      "entropy": 5.395793724060058,
      "epoch": 0.6214500501169395,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004967417053449178,
      "loss": 5.2694,
      "mean_token_accuracy": 0.1843681290745735,
      "num_tokens": 13673925.0,
      "step": 7440
    },
    {
      "entropy": 5.338581800460815,
      "epoch": 0.6218676912796525,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004967366554031945,
      "loss": 5.178,
      "mean_token_accuracy": 0.18515721708536148,
      "num_tokens": 13682252.0,
      "step": 7445
    },
    {
      "entropy": 5.216605138778687,
      "epoch": 0.6222853324423655,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004967316015797042,
      "loss": 5.1968,
      "mean_token_accuracy": 0.1913221836090088,
      "num_tokens": 13691462.0,
      "step": 7450
    },
    {
      "entropy": 5.4017653465271,
      "epoch": 0.6227029736050785,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004967265438745352,
      "loss": 5.2517,
      "mean_token_accuracy": 0.1839205339550972,
      "num_tokens": 13701370.0,
      "step": 7455
    },
    {
      "entropy": 5.366975736618042,
      "epoch": 0.6231206147677915,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004967214822877763,
      "loss": 5.2277,
      "mean_token_accuracy": 0.18677881509065627,
      "num_tokens": 13709673.0,
      "step": 7460
    },
    {
      "entropy": 5.234641790390015,
      "epoch": 0.6235382559305045,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004967164168195159,
      "loss": 5.2688,
      "mean_token_accuracy": 0.18868394643068315,
      "num_tokens": 13719947.0,
      "step": 7465
    },
    {
      "entropy": 5.3808183670043945,
      "epoch": 0.6239558970932175,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004967113474698428,
      "loss": 5.257,
      "mean_token_accuracy": 0.18341430723667146,
      "num_tokens": 13728813.0,
      "step": 7470
    },
    {
      "entropy": 5.396779632568359,
      "epoch": 0.6243735382559304,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004967062742388456,
      "loss": 5.2901,
      "mean_token_accuracy": 0.18841298520565034,
      "num_tokens": 13738319.0,
      "step": 7475
    },
    {
      "entropy": 5.248705005645752,
      "epoch": 0.6247911794186435,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004967011971266132,
      "loss": 5.0932,
      "mean_token_accuracy": 0.191838076710701,
      "num_tokens": 13747337.0,
      "step": 7480
    },
    {
      "entropy": 5.271488285064697,
      "epoch": 0.6252088205813565,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004966961161332346,
      "loss": 5.1089,
      "mean_token_accuracy": 0.18855925649404526,
      "num_tokens": 13756050.0,
      "step": 7485
    },
    {
      "entropy": 5.308868312835694,
      "epoch": 0.6256264617440696,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004966910312587985,
      "loss": 5.2168,
      "mean_token_accuracy": 0.1839320629835129,
      "num_tokens": 13765297.0,
      "step": 7490
    },
    {
      "entropy": 5.336758756637574,
      "epoch": 0.6260441029067825,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004966859425033941,
      "loss": 5.2085,
      "mean_token_accuracy": 0.1943068489432335,
      "num_tokens": 13773958.0,
      "step": 7495
    },
    {
      "entropy": 5.285659694671631,
      "epoch": 0.6264617440694955,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004966808498671104,
      "loss": 5.271,
      "mean_token_accuracy": 0.1855810433626175,
      "num_tokens": 13784037.0,
      "step": 7500
    },
    {
      "entropy": 5.382388496398926,
      "epoch": 0.6268793852322085,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004966757533500367,
      "loss": 5.2457,
      "mean_token_accuracy": 0.19026362299919128,
      "num_tokens": 13793709.0,
      "step": 7505
    },
    {
      "entropy": 5.288311243057251,
      "epoch": 0.6272970263949215,
      "grad_norm": 1.203125,
      "learning_rate": 0.000496670652952262,
      "loss": 5.1078,
      "mean_token_accuracy": 0.19725475013256072,
      "num_tokens": 13801466.0,
      "step": 7510
    },
    {
      "entropy": 5.331857442855835,
      "epoch": 0.6277146675576345,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004966655486738757,
      "loss": 5.2512,
      "mean_token_accuracy": 0.17607493251562117,
      "num_tokens": 13811903.0,
      "step": 7515
    },
    {
      "entropy": 5.3850696086883545,
      "epoch": 0.6281323087203475,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004966604405149672,
      "loss": 5.2609,
      "mean_token_accuracy": 0.19172122031450273,
      "num_tokens": 13820861.0,
      "step": 7520
    },
    {
      "entropy": 5.385110235214233,
      "epoch": 0.6285499498830605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004966553284756257,
      "loss": 5.3254,
      "mean_token_accuracy": 0.1790284112095833,
      "num_tokens": 13831049.0,
      "step": 7525
    },
    {
      "entropy": 5.344508504867553,
      "epoch": 0.6289675910457735,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004966502125559411,
      "loss": 5.2478,
      "mean_token_accuracy": 0.18893121927976608,
      "num_tokens": 13840628.0,
      "step": 7530
    },
    {
      "entropy": 5.25383415222168,
      "epoch": 0.6293852322084865,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004966450927560026,
      "loss": 5.1073,
      "mean_token_accuracy": 0.18811701685190202,
      "num_tokens": 13849424.0,
      "step": 7535
    },
    {
      "entropy": 5.24720778465271,
      "epoch": 0.6298028733711994,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004966399690759,
      "loss": 5.1524,
      "mean_token_accuracy": 0.19245729148387908,
      "num_tokens": 13859216.0,
      "step": 7540
    },
    {
      "entropy": 5.361807584762573,
      "epoch": 0.6302205145339125,
      "grad_norm": 1.21875,
      "learning_rate": 0.000496634841515723,
      "loss": 5.2818,
      "mean_token_accuracy": 0.1917789936065674,
      "num_tokens": 13867638.0,
      "step": 7545
    },
    {
      "entropy": 5.4235604763031,
      "epoch": 0.6306381556966255,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004966297100755613,
      "loss": 5.2996,
      "mean_token_accuracy": 0.1831221789121628,
      "num_tokens": 13877016.0,
      "step": 7550
    },
    {
      "entropy": 5.312173700332641,
      "epoch": 0.6310557968593384,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004966245747555047,
      "loss": 5.2891,
      "mean_token_accuracy": 0.17328477948904036,
      "num_tokens": 13887305.0,
      "step": 7555
    },
    {
      "entropy": 5.295931434631347,
      "epoch": 0.6314734380220515,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004966194355556432,
      "loss": 5.1256,
      "mean_token_accuracy": 0.19361737221479416,
      "num_tokens": 13896003.0,
      "step": 7560
    },
    {
      "entropy": 5.282864570617676,
      "epoch": 0.6318910791847645,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004966142924760666,
      "loss": 5.2229,
      "mean_token_accuracy": 0.18436953872442247,
      "num_tokens": 13904964.0,
      "step": 7565
    },
    {
      "entropy": 5.317825841903686,
      "epoch": 0.6323087203474774,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004966091455168652,
      "loss": 5.2212,
      "mean_token_accuracy": 0.18204531371593474,
      "num_tokens": 13915049.0,
      "step": 7570
    },
    {
      "entropy": 5.303839635848999,
      "epoch": 0.6327263615101905,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004966039946781289,
      "loss": 5.1524,
      "mean_token_accuracy": 0.19663185626268387,
      "num_tokens": 13924938.0,
      "step": 7575
    },
    {
      "entropy": 5.43185510635376,
      "epoch": 0.6331440026729035,
      "grad_norm": 1.078125,
      "learning_rate": 0.000496598839959948,
      "loss": 5.3817,
      "mean_token_accuracy": 0.16907385885715484,
      "num_tokens": 13935050.0,
      "step": 7580
    },
    {
      "entropy": 5.339563465118408,
      "epoch": 0.6335616438356164,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004965936813624126,
      "loss": 5.1755,
      "mean_token_accuracy": 0.19084191918373108,
      "num_tokens": 13943576.0,
      "step": 7585
    },
    {
      "entropy": 5.241533613204956,
      "epoch": 0.6339792849983295,
      "grad_norm": 1.09375,
      "learning_rate": 0.000496588518885613,
      "loss": 5.2051,
      "mean_token_accuracy": 0.18564321249723434,
      "num_tokens": 13953232.0,
      "step": 7590
    },
    {
      "entropy": 5.343335962295532,
      "epoch": 0.6343969261610425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004965833525296397,
      "loss": 5.3758,
      "mean_token_accuracy": 0.17697664722800255,
      "num_tokens": 13962121.0,
      "step": 7595
    },
    {
      "entropy": 5.385099267959594,
      "epoch": 0.6348145673237554,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004965781822945831,
      "loss": 5.2501,
      "mean_token_accuracy": 0.17955022156238556,
      "num_tokens": 13971681.0,
      "step": 7600
    },
    {
      "entropy": 5.30552110671997,
      "epoch": 0.6352322084864684,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004965730081805338,
      "loss": 5.1102,
      "mean_token_accuracy": 0.19973503351211547,
      "num_tokens": 13980118.0,
      "step": 7605
    },
    {
      "entropy": 5.297439622879028,
      "epoch": 0.6356498496491815,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004965678301875821,
      "loss": 5.1404,
      "mean_token_accuracy": 0.18801899403333663,
      "num_tokens": 13989755.0,
      "step": 7610
    },
    {
      "entropy": 5.222371053695679,
      "epoch": 0.6360674908118944,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000496562648315819,
      "loss": 5.155,
      "mean_token_accuracy": 0.18301116526126862,
      "num_tokens": 13999289.0,
      "step": 7615
    },
    {
      "entropy": 5.316830015182495,
      "epoch": 0.6364851319746074,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000496557462565335,
      "loss": 5.2644,
      "mean_token_accuracy": 0.18290550112724305,
      "num_tokens": 14009101.0,
      "step": 7620
    },
    {
      "entropy": 5.377586317062378,
      "epoch": 0.6369027731373205,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004965522729362209,
      "loss": 5.139,
      "mean_token_accuracy": 0.18653845489025117,
      "num_tokens": 14018197.0,
      "step": 7625
    },
    {
      "entropy": 5.2109386920928955,
      "epoch": 0.6373204143000334,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004965470794285676,
      "loss": 5.1341,
      "mean_token_accuracy": 0.19235458523035048,
      "num_tokens": 14028108.0,
      "step": 7630
    },
    {
      "entropy": 5.31895809173584,
      "epoch": 0.6377380554627464,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000496541882042466,
      "loss": 5.2855,
      "mean_token_accuracy": 0.1735656425356865,
      "num_tokens": 14038636.0,
      "step": 7635
    },
    {
      "entropy": 5.3375874042510985,
      "epoch": 0.6381556966254595,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004965366807780071,
      "loss": 5.2375,
      "mean_token_accuracy": 0.1839364290237427,
      "num_tokens": 14048502.0,
      "step": 7640
    },
    {
      "entropy": 5.372132015228272,
      "epoch": 0.6385733377881724,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004965314756352818,
      "loss": 5.3236,
      "mean_token_accuracy": 0.17816595435142518,
      "num_tokens": 14058654.0,
      "step": 7645
    },
    {
      "entropy": 5.328364324569702,
      "epoch": 0.6389909789508854,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004965262666143816,
      "loss": 5.1575,
      "mean_token_accuracy": 0.19192207008600234,
      "num_tokens": 14068614.0,
      "step": 7650
    },
    {
      "entropy": 5.280334568023681,
      "epoch": 0.6394086201135984,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004965210537153973,
      "loss": 5.216,
      "mean_token_accuracy": 0.1896479234099388,
      "num_tokens": 14077550.0,
      "step": 7655
    },
    {
      "entropy": 5.306500101089478,
      "epoch": 0.6398262612763114,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004965158369384204,
      "loss": 5.262,
      "mean_token_accuracy": 0.18305613249540328,
      "num_tokens": 14086610.0,
      "step": 7660
    },
    {
      "entropy": 5.260514593124389,
      "epoch": 0.6402439024390244,
      "grad_norm": 1.203125,
      "learning_rate": 0.000496510616283542,
      "loss": 5.2614,
      "mean_token_accuracy": 0.18237021267414094,
      "num_tokens": 14095795.0,
      "step": 7665
    },
    {
      "entropy": 5.321788883209228,
      "epoch": 0.6406615436017374,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004965053917508537,
      "loss": 5.1405,
      "mean_token_accuracy": 0.1904669240117073,
      "num_tokens": 14104779.0,
      "step": 7670
    },
    {
      "entropy": 5.297952651977539,
      "epoch": 0.6410791847644504,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004965001633404469,
      "loss": 5.0558,
      "mean_token_accuracy": 0.19697895497083664,
      "num_tokens": 14113299.0,
      "step": 7675
    },
    {
      "entropy": 5.252028608322144,
      "epoch": 0.6414968259271634,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004964949310524131,
      "loss": 5.2322,
      "mean_token_accuracy": 0.19213329404592513,
      "num_tokens": 14121969.0,
      "step": 7680
    },
    {
      "entropy": 5.353702354431152,
      "epoch": 0.6419144670898764,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004964896948868439,
      "loss": 5.1617,
      "mean_token_accuracy": 0.18738498985767366,
      "num_tokens": 14130850.0,
      "step": 7685
    },
    {
      "entropy": 5.330335378646851,
      "epoch": 0.6423321082525894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004964844548438311,
      "loss": 5.2063,
      "mean_token_accuracy": 0.18880516290664673,
      "num_tokens": 14140142.0,
      "step": 7690
    },
    {
      "entropy": 5.123462390899658,
      "epoch": 0.6427497494153024,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004964792109234662,
      "loss": 5.093,
      "mean_token_accuracy": 0.19274820387363434,
      "num_tokens": 14148335.0,
      "step": 7695
    },
    {
      "entropy": 5.360991096496582,
      "epoch": 0.6431673905780154,
      "grad_norm": 1.125,
      "learning_rate": 0.0004964739631258411,
      "loss": 5.2553,
      "mean_token_accuracy": 0.1787935361266136,
      "num_tokens": 14157566.0,
      "step": 7700
    },
    {
      "entropy": 5.3845436573028564,
      "epoch": 0.6435850317407283,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004964687114510478,
      "loss": 5.261,
      "mean_token_accuracy": 0.17988270670175552,
      "num_tokens": 14166037.0,
      "step": 7705
    },
    {
      "entropy": 5.191599130630493,
      "epoch": 0.6440026729034414,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000496463455899178,
      "loss": 5.1262,
      "mean_token_accuracy": 0.19823317974805832,
      "num_tokens": 14174735.0,
      "step": 7710
    },
    {
      "entropy": 5.350530290603638,
      "epoch": 0.6444203140661544,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004964581964703239,
      "loss": 5.2549,
      "mean_token_accuracy": 0.18151563853025438,
      "num_tokens": 14185130.0,
      "step": 7715
    },
    {
      "entropy": 5.4578296661376955,
      "epoch": 0.6448379552288673,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004964529331645776,
      "loss": 5.2584,
      "mean_token_accuracy": 0.18424693793058394,
      "num_tokens": 14194676.0,
      "step": 7720
    },
    {
      "entropy": 5.360512399673462,
      "epoch": 0.6452555963915804,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000496447665982031,
      "loss": 5.2263,
      "mean_token_accuracy": 0.18780234456062317,
      "num_tokens": 14204218.0,
      "step": 7725
    },
    {
      "entropy": 5.230155420303345,
      "epoch": 0.6456732375542934,
      "grad_norm": 1.125,
      "learning_rate": 0.0004964423949227765,
      "loss": 5.2559,
      "mean_token_accuracy": 0.18467172980308533,
      "num_tokens": 14213176.0,
      "step": 7730
    },
    {
      "entropy": 5.373636198043823,
      "epoch": 0.6460908787170063,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004964371199869062,
      "loss": 5.3205,
      "mean_token_accuracy": 0.18102581351995467,
      "num_tokens": 14222490.0,
      "step": 7735
    },
    {
      "entropy": 5.3416601657867435,
      "epoch": 0.6465085198797194,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004964318411745127,
      "loss": 5.2803,
      "mean_token_accuracy": 0.17828164547681807,
      "num_tokens": 14231335.0,
      "step": 7740
    },
    {
      "entropy": 5.359674739837646,
      "epoch": 0.6469261610424324,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004964265584856883,
      "loss": 5.2303,
      "mean_token_accuracy": 0.18459064662456512,
      "num_tokens": 14240583.0,
      "step": 7745
    },
    {
      "entropy": 5.392180061340332,
      "epoch": 0.6473438022051453,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004964212719205254,
      "loss": 5.2008,
      "mean_token_accuracy": 0.19037179052829742,
      "num_tokens": 14250087.0,
      "step": 7750
    },
    {
      "entropy": 5.374785852432251,
      "epoch": 0.6477614433678583,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004964159814791167,
      "loss": 5.3015,
      "mean_token_accuracy": 0.17718515545129776,
      "num_tokens": 14261076.0,
      "step": 7755
    },
    {
      "entropy": 5.3138974666595455,
      "epoch": 0.6481790845305714,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004964106871615547,
      "loss": 5.1898,
      "mean_token_accuracy": 0.18958569467067718,
      "num_tokens": 14270234.0,
      "step": 7760
    },
    {
      "entropy": 5.254745769500732,
      "epoch": 0.6485967256932843,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000496405388967932,
      "loss": 5.1477,
      "mean_token_accuracy": 0.18739570677280426,
      "num_tokens": 14279154.0,
      "step": 7765
    },
    {
      "entropy": 5.262589073181152,
      "epoch": 0.6490143668559973,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004964000868983416,
      "loss": 5.2272,
      "mean_token_accuracy": 0.18639917820692062,
      "num_tokens": 14289626.0,
      "step": 7770
    },
    {
      "entropy": 5.267973899841309,
      "epoch": 0.6494320080187104,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004963947809528761,
      "loss": 5.1356,
      "mean_token_accuracy": 0.19295641034841537,
      "num_tokens": 14299647.0,
      "step": 7775
    },
    {
      "entropy": 5.436268663406372,
      "epoch": 0.6498496491814233,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004963894711316285,
      "loss": 5.3369,
      "mean_token_accuracy": 0.179313787817955,
      "num_tokens": 14308944.0,
      "step": 7780
    },
    {
      "entropy": 5.38254828453064,
      "epoch": 0.6502672903441363,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004963841574346917,
      "loss": 5.2986,
      "mean_token_accuracy": 0.1824707269668579,
      "num_tokens": 14318168.0,
      "step": 7785
    },
    {
      "entropy": 5.289823293685913,
      "epoch": 0.6506849315068494,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004963788398621587,
      "loss": 5.1312,
      "mean_token_accuracy": 0.1992276653647423,
      "num_tokens": 14327249.0,
      "step": 7790
    },
    {
      "entropy": 5.281346654891967,
      "epoch": 0.6511025726695623,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004963735184141227,
      "loss": 5.2197,
      "mean_token_accuracy": 0.18996517956256867,
      "num_tokens": 14336055.0,
      "step": 7795
    },
    {
      "entropy": 5.295825099945068,
      "epoch": 0.6515202138322753,
      "grad_norm": 1.125,
      "learning_rate": 0.0004963681930906768,
      "loss": 5.1459,
      "mean_token_accuracy": 0.19329801350831985,
      "num_tokens": 14344862.0,
      "step": 7800
    },
    {
      "entropy": 5.403456354141236,
      "epoch": 0.6519378549949884,
      "grad_norm": 1.125,
      "learning_rate": 0.0004963628638919141,
      "loss": 5.3726,
      "mean_token_accuracy": 0.17764141261577607,
      "num_tokens": 14354284.0,
      "step": 7805
    },
    {
      "entropy": 5.354050016403198,
      "epoch": 0.6523554961577013,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004963575308179281,
      "loss": 5.1875,
      "mean_token_accuracy": 0.18953971862792968,
      "num_tokens": 14363615.0,
      "step": 7810
    },
    {
      "entropy": 5.330633354187012,
      "epoch": 0.6527731373204143,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004963521938688122,
      "loss": 5.1425,
      "mean_token_accuracy": 0.18493614941835404,
      "num_tokens": 14372805.0,
      "step": 7815
    },
    {
      "entropy": 5.274392747879029,
      "epoch": 0.6531907784831273,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004963468530446596,
      "loss": 5.2607,
      "mean_token_accuracy": 0.17995638400316238,
      "num_tokens": 14382334.0,
      "step": 7820
    },
    {
      "entropy": 5.328124380111694,
      "epoch": 0.6536084196458403,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004963415083455639,
      "loss": 5.1736,
      "mean_token_accuracy": 0.18850196301937103,
      "num_tokens": 14391360.0,
      "step": 7825
    },
    {
      "entropy": 5.347755670547485,
      "epoch": 0.6540260608085533,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004963361597716186,
      "loss": 5.3621,
      "mean_token_accuracy": 0.1800881505012512,
      "num_tokens": 14401232.0,
      "step": 7830
    },
    {
      "entropy": 5.426602840423584,
      "epoch": 0.6544437019712663,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004963308073229175,
      "loss": 5.2234,
      "mean_token_accuracy": 0.1826469749212265,
      "num_tokens": 14409773.0,
      "step": 7835
    },
    {
      "entropy": 5.385968399047852,
      "epoch": 0.6548613431339793,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004963254509995542,
      "loss": 5.2057,
      "mean_token_accuracy": 0.18334679454565048,
      "num_tokens": 14419511.0,
      "step": 7840
    },
    {
      "entropy": 5.2667396545410154,
      "epoch": 0.6552789842966923,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004963200908016224,
      "loss": 5.1925,
      "mean_token_accuracy": 0.19224025309085846,
      "num_tokens": 14428611.0,
      "step": 7845
    },
    {
      "entropy": 5.1868668556213375,
      "epoch": 0.6556966254594053,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000496314726729216,
      "loss": 5.0746,
      "mean_token_accuracy": 0.19552791267633438,
      "num_tokens": 14436721.0,
      "step": 7850
    },
    {
      "entropy": 5.320091581344604,
      "epoch": 0.6561142666221182,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004963093587824289,
      "loss": 5.2332,
      "mean_token_accuracy": 0.18761514872312546,
      "num_tokens": 14445766.0,
      "step": 7855
    },
    {
      "entropy": 5.244475412368774,
      "epoch": 0.6565319077848313,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004963039869613551,
      "loss": 5.1005,
      "mean_token_accuracy": 0.18802379816770554,
      "num_tokens": 14453857.0,
      "step": 7860
    },
    {
      "entropy": 5.245399618148804,
      "epoch": 0.6569495489475443,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004962986112660887,
      "loss": 5.2395,
      "mean_token_accuracy": 0.19586975872516632,
      "num_tokens": 14462356.0,
      "step": 7865
    },
    {
      "entropy": 5.287431621551514,
      "epoch": 0.6573671901102572,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004962932316967237,
      "loss": 5.1637,
      "mean_token_accuracy": 0.19043235331773758,
      "num_tokens": 14471848.0,
      "step": 7870
    },
    {
      "entropy": 5.263445663452148,
      "epoch": 0.6577848312729703,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004962878482533544,
      "loss": 5.2027,
      "mean_token_accuracy": 0.18838344365358353,
      "num_tokens": 14481128.0,
      "step": 7875
    },
    {
      "entropy": 5.280665159225464,
      "epoch": 0.6582024724356833,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004962824609360748,
      "loss": 5.0895,
      "mean_token_accuracy": 0.19354743063449859,
      "num_tokens": 14490071.0,
      "step": 7880
    },
    {
      "entropy": 5.344894695281982,
      "epoch": 0.6586201135983962,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004962770697449794,
      "loss": 5.2365,
      "mean_token_accuracy": 0.18232745081186294,
      "num_tokens": 14499162.0,
      "step": 7885
    },
    {
      "entropy": 5.327264404296875,
      "epoch": 0.6590377547611093,
      "grad_norm": 1.25,
      "learning_rate": 0.0004962716746801625,
      "loss": 5.1882,
      "mean_token_accuracy": 0.18955582082271577,
      "num_tokens": 14508525.0,
      "step": 7890
    },
    {
      "entropy": 5.304840087890625,
      "epoch": 0.6594553959238223,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004962662757417187,
      "loss": 5.157,
      "mean_token_accuracy": 0.1902075305581093,
      "num_tokens": 14516885.0,
      "step": 7895
    },
    {
      "entropy": 5.261585330963134,
      "epoch": 0.6598730370865352,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004962608729297424,
      "loss": 5.1358,
      "mean_token_accuracy": 0.19048814475536346,
      "num_tokens": 14525116.0,
      "step": 7900
    },
    {
      "entropy": 5.250884675979615,
      "epoch": 0.6602906782492483,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004962554662443281,
      "loss": 5.2641,
      "mean_token_accuracy": 0.1920710727572441,
      "num_tokens": 14534452.0,
      "step": 7905
    },
    {
      "entropy": 5.284350824356079,
      "epoch": 0.6607083194119613,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004962500556855706,
      "loss": 5.1212,
      "mean_token_accuracy": 0.19259440004825593,
      "num_tokens": 14544040.0,
      "step": 7910
    },
    {
      "entropy": 5.422929334640503,
      "epoch": 0.6611259605746742,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004962446412535645,
      "loss": 5.2602,
      "mean_token_accuracy": 0.18493396043777466,
      "num_tokens": 14553343.0,
      "step": 7915
    },
    {
      "entropy": 5.316207981109619,
      "epoch": 0.6615436017373872,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004962392229484046,
      "loss": 5.1587,
      "mean_token_accuracy": 0.18950279653072358,
      "num_tokens": 14562657.0,
      "step": 7920
    },
    {
      "entropy": 5.242904329299927,
      "epoch": 0.6619612429001003,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000496233800770186,
      "loss": 5.154,
      "mean_token_accuracy": 0.18897778391838074,
      "num_tokens": 14572167.0,
      "step": 7925
    },
    {
      "entropy": 5.254485607147217,
      "epoch": 0.6623788840628132,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004962283747190033,
      "loss": 5.164,
      "mean_token_accuracy": 0.18617894649505615,
      "num_tokens": 14580934.0,
      "step": 7930
    },
    {
      "entropy": 5.275124263763428,
      "epoch": 0.6627965252255262,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004962229447949515,
      "loss": 5.154,
      "mean_token_accuracy": 0.19313859790563584,
      "num_tokens": 14589579.0,
      "step": 7935
    },
    {
      "entropy": 5.327301549911499,
      "epoch": 0.6632141663882393,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004962175109981259,
      "loss": 5.2216,
      "mean_token_accuracy": 0.1868191123008728,
      "num_tokens": 14599289.0,
      "step": 7940
    },
    {
      "entropy": 5.343547105789185,
      "epoch": 0.6636318075509522,
      "grad_norm": 1.25,
      "learning_rate": 0.0004962120733286215,
      "loss": 5.1827,
      "mean_token_accuracy": 0.19341417402029037,
      "num_tokens": 14608322.0,
      "step": 7945
    },
    {
      "entropy": 5.324182224273682,
      "epoch": 0.6640494487136652,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004962066317865334,
      "loss": 5.2062,
      "mean_token_accuracy": 0.190206441283226,
      "num_tokens": 14617591.0,
      "step": 7950
    },
    {
      "entropy": 5.283283615112305,
      "epoch": 0.6644670898763783,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000496201186371957,
      "loss": 5.1554,
      "mean_token_accuracy": 0.1912426620721817,
      "num_tokens": 14625809.0,
      "step": 7955
    },
    {
      "entropy": 5.375922298431396,
      "epoch": 0.6648847310390912,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004961957370849875,
      "loss": 5.2284,
      "mean_token_accuracy": 0.18226083368062973,
      "num_tokens": 14635061.0,
      "step": 7960
    },
    {
      "entropy": 5.26524715423584,
      "epoch": 0.6653023722018042,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004961902839257205,
      "loss": 5.2736,
      "mean_token_accuracy": 0.18430304676294326,
      "num_tokens": 14645104.0,
      "step": 7965
    },
    {
      "entropy": 5.364509105682373,
      "epoch": 0.6657200133645172,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004961848268942513,
      "loss": 5.2402,
      "mean_token_accuracy": 0.18456349819898604,
      "num_tokens": 14653984.0,
      "step": 7970
    },
    {
      "entropy": 5.258089303970337,
      "epoch": 0.6661376545272302,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004961793659906754,
      "loss": 5.184,
      "mean_token_accuracy": 0.18901885002851487,
      "num_tokens": 14663315.0,
      "step": 7975
    },
    {
      "entropy": 5.215296411514283,
      "epoch": 0.6665552956899432,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004961739012150885,
      "loss": 5.1774,
      "mean_token_accuracy": 0.1944261074066162,
      "num_tokens": 14672608.0,
      "step": 7980
    },
    {
      "entropy": 5.38453106880188,
      "epoch": 0.6669729368526562,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004961684325675862,
      "loss": 5.2197,
      "mean_token_accuracy": 0.19073372185230256,
      "num_tokens": 14682031.0,
      "step": 7985
    },
    {
      "entropy": 5.307742977142334,
      "epoch": 0.6673905780153692,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004961629600482643,
      "loss": 5.242,
      "mean_token_accuracy": 0.18530760258436202,
      "num_tokens": 14691894.0,
      "step": 7990
    },
    {
      "entropy": 5.276402330398559,
      "epoch": 0.6678082191780822,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004961574836572186,
      "loss": 5.1867,
      "mean_token_accuracy": 0.18451732099056245,
      "num_tokens": 14700508.0,
      "step": 7995
    },
    {
      "entropy": 5.397201347351074,
      "epoch": 0.6682258603407952,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004961520033945448,
      "loss": 5.2214,
      "mean_token_accuracy": 0.1825764000415802,
      "num_tokens": 14709241.0,
      "step": 8000
    },
    {
      "entropy": 5.342011737823486,
      "epoch": 0.6686435015035082,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004961465192603391,
      "loss": 5.2368,
      "mean_token_accuracy": 0.18244830667972564,
      "num_tokens": 14719300.0,
      "step": 8005
    },
    {
      "entropy": 5.3934918403625485,
      "epoch": 0.6690611426662212,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004961410312546974,
      "loss": 5.2947,
      "mean_token_accuracy": 0.18987406194210052,
      "num_tokens": 14729717.0,
      "step": 8010
    },
    {
      "entropy": 5.3500213623046875,
      "epoch": 0.6694787838289342,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004961355393777158,
      "loss": 5.1179,
      "mean_token_accuracy": 0.1889166057109833,
      "num_tokens": 14738757.0,
      "step": 8015
    },
    {
      "entropy": 5.0954750061035154,
      "epoch": 0.6698964249916471,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004961300436294902,
      "loss": 5.1069,
      "mean_token_accuracy": 0.18967511951923371,
      "num_tokens": 14747171.0,
      "step": 8020
    },
    {
      "entropy": 5.1857727527618405,
      "epoch": 0.6703140661543602,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004961245440101171,
      "loss": 5.1301,
      "mean_token_accuracy": 0.1838504657149315,
      "num_tokens": 14756064.0,
      "step": 8025
    },
    {
      "entropy": 5.3562424659729,
      "epoch": 0.6707317073170732,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004961190405196928,
      "loss": 5.1366,
      "mean_token_accuracy": 0.1907202944159508,
      "num_tokens": 14765156.0,
      "step": 8030
    },
    {
      "entropy": 5.4175444602966305,
      "epoch": 0.6711493484797861,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004961135331583135,
      "loss": 5.205,
      "mean_token_accuracy": 0.19196956157684325,
      "num_tokens": 14774059.0,
      "step": 8035
    },
    {
      "entropy": 5.201346492767334,
      "epoch": 0.6715669896424992,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004961080219260755,
      "loss": 5.115,
      "mean_token_accuracy": 0.19651183634996414,
      "num_tokens": 14783500.0,
      "step": 8040
    },
    {
      "entropy": 5.31565113067627,
      "epoch": 0.6719846308052122,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004961025068230757,
      "loss": 5.2174,
      "mean_token_accuracy": 0.18489293456077577,
      "num_tokens": 14792901.0,
      "step": 8045
    },
    {
      "entropy": 5.413379955291748,
      "epoch": 0.6724022719679251,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004960969878494103,
      "loss": 5.1998,
      "mean_token_accuracy": 0.1820690855383873,
      "num_tokens": 14801619.0,
      "step": 8050
    },
    {
      "entropy": 5.240681743621826,
      "epoch": 0.6728199131306382,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004960914650051758,
      "loss": 5.1637,
      "mean_token_accuracy": 0.1892152801156044,
      "num_tokens": 14811128.0,
      "step": 8055
    },
    {
      "entropy": 5.2843005657196045,
      "epoch": 0.6732375542933512,
      "grad_norm": 1.125,
      "learning_rate": 0.0004960859382904691,
      "loss": 5.2199,
      "mean_token_accuracy": 0.18780788034200668,
      "num_tokens": 14819772.0,
      "step": 8060
    },
    {
      "entropy": 5.304719734191894,
      "epoch": 0.6736551954560641,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000496080407705387,
      "loss": 5.0928,
      "mean_token_accuracy": 0.19334981739521026,
      "num_tokens": 14828139.0,
      "step": 8065
    },
    {
      "entropy": 5.315226221084595,
      "epoch": 0.6740728366187771,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004960748732500263,
      "loss": 5.2266,
      "mean_token_accuracy": 0.18278479874134063,
      "num_tokens": 14837121.0,
      "step": 8070
    },
    {
      "entropy": 5.302050685882568,
      "epoch": 0.6744904777814902,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004960693349244838,
      "loss": 5.1677,
      "mean_token_accuracy": 0.188950976729393,
      "num_tokens": 14845973.0,
      "step": 8075
    },
    {
      "entropy": 5.295114278793335,
      "epoch": 0.6749081189442031,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004960637927288566,
      "loss": 5.2624,
      "mean_token_accuracy": 0.18287192285060883,
      "num_tokens": 14856072.0,
      "step": 8080
    },
    {
      "entropy": 5.262533712387085,
      "epoch": 0.6753257601069161,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004960582466632414,
      "loss": 5.1554,
      "mean_token_accuracy": 0.1857127159833908,
      "num_tokens": 14864845.0,
      "step": 8085
    },
    {
      "entropy": 5.317600965499878,
      "epoch": 0.6757434012696292,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004960526967277356,
      "loss": 5.1702,
      "mean_token_accuracy": 0.18602113127708436,
      "num_tokens": 14874170.0,
      "step": 8090
    },
    {
      "entropy": 5.296185922622681,
      "epoch": 0.6761610424323421,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004960471429224363,
      "loss": 5.1112,
      "mean_token_accuracy": 0.19489070773124695,
      "num_tokens": 14882618.0,
      "step": 8095
    },
    {
      "entropy": 5.308361148834228,
      "epoch": 0.6765786835950551,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004960415852474406,
      "loss": 5.2508,
      "mean_token_accuracy": 0.17853918075561523,
      "num_tokens": 14893258.0,
      "step": 8100
    },
    {
      "entropy": 5.336450529098511,
      "epoch": 0.6769963247577682,
      "grad_norm": 1.078125,
      "learning_rate": 0.000496036023702846,
      "loss": 5.3112,
      "mean_token_accuracy": 0.17795838117599488,
      "num_tokens": 14904424.0,
      "step": 8105
    },
    {
      "entropy": 5.40800838470459,
      "epoch": 0.6774139659204811,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004960304582887497,
      "loss": 5.2794,
      "mean_token_accuracy": 0.1789371445775032,
      "num_tokens": 14913935.0,
      "step": 8110
    },
    {
      "entropy": 5.236117124557495,
      "epoch": 0.6778316070831941,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004960248890052493,
      "loss": 5.0956,
      "mean_token_accuracy": 0.1956703245639801,
      "num_tokens": 14923352.0,
      "step": 8115
    },
    {
      "entropy": 5.184804248809814,
      "epoch": 0.6782492482459072,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000496019315852442,
      "loss": 5.0684,
      "mean_token_accuracy": 0.19307749718427658,
      "num_tokens": 14932080.0,
      "step": 8120
    },
    {
      "entropy": 5.289352321624756,
      "epoch": 0.6786668894086201,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004960137388304255,
      "loss": 5.221,
      "mean_token_accuracy": 0.18349912762641907,
      "num_tokens": 14939747.0,
      "step": 8125
    },
    {
      "entropy": 5.277930116653442,
      "epoch": 0.6790845305713331,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004960081579392975,
      "loss": 5.2157,
      "mean_token_accuracy": 0.18977949172258377,
      "num_tokens": 14948639.0,
      "step": 8130
    },
    {
      "entropy": 5.346895742416382,
      "epoch": 0.6795021717340461,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004960025731791558,
      "loss": 5.1724,
      "mean_token_accuracy": 0.1869726985692978,
      "num_tokens": 14957831.0,
      "step": 8135
    },
    {
      "entropy": 5.384111928939819,
      "epoch": 0.6799198128967591,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004959969845500979,
      "loss": 5.334,
      "mean_token_accuracy": 0.1735308885574341,
      "num_tokens": 14968732.0,
      "step": 8140
    },
    {
      "entropy": 5.318551921844483,
      "epoch": 0.6803374540594721,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004959913920522218,
      "loss": 5.1665,
      "mean_token_accuracy": 0.19190097600221634,
      "num_tokens": 14978684.0,
      "step": 8145
    },
    {
      "entropy": 5.502088594436645,
      "epoch": 0.6807550952221851,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004959857956856254,
      "loss": 5.3792,
      "mean_token_accuracy": 0.18139556646347046,
      "num_tokens": 14988036.0,
      "step": 8150
    },
    {
      "entropy": 5.332143402099609,
      "epoch": 0.6811727363848981,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004959801954504067,
      "loss": 5.1558,
      "mean_token_accuracy": 0.194850791990757,
      "num_tokens": 14996773.0,
      "step": 8155
    },
    {
      "entropy": 5.29870080947876,
      "epoch": 0.6815903775476111,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004959745913466635,
      "loss": 5.2494,
      "mean_token_accuracy": 0.18419882506132126,
      "num_tokens": 15005552.0,
      "step": 8160
    },
    {
      "entropy": 5.280312871932983,
      "epoch": 0.6820080187103241,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004959689833744942,
      "loss": 5.1666,
      "mean_token_accuracy": 0.19198163002729415,
      "num_tokens": 15014026.0,
      "step": 8165
    },
    {
      "entropy": 5.363230514526367,
      "epoch": 0.682425659873037,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004959633715339969,
      "loss": 5.211,
      "mean_token_accuracy": 0.18872150033712387,
      "num_tokens": 15023607.0,
      "step": 8170
    },
    {
      "entropy": 5.294639158248901,
      "epoch": 0.6828433010357501,
      "grad_norm": 1.125,
      "learning_rate": 0.0004959577558252698,
      "loss": 5.2251,
      "mean_token_accuracy": 0.1872321143746376,
      "num_tokens": 15033102.0,
      "step": 8175
    },
    {
      "entropy": 5.307678127288819,
      "epoch": 0.6832609421984631,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004959521362484112,
      "loss": 5.1896,
      "mean_token_accuracy": 0.191020168364048,
      "num_tokens": 15042506.0,
      "step": 8180
    },
    {
      "entropy": 5.313725519180298,
      "epoch": 0.683678583361176,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004959465128035195,
      "loss": 5.1572,
      "mean_token_accuracy": 0.19131357371807098,
      "num_tokens": 15052483.0,
      "step": 8185
    },
    {
      "entropy": 5.27556676864624,
      "epoch": 0.6840962245238891,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004959408854906932,
      "loss": 5.1567,
      "mean_token_accuracy": 0.19765861928462983,
      "num_tokens": 15060769.0,
      "step": 8190
    },
    {
      "entropy": 5.3392613410949705,
      "epoch": 0.6845138656866021,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004959352543100306,
      "loss": 5.278,
      "mean_token_accuracy": 0.18380562216043472,
      "num_tokens": 15070398.0,
      "step": 8195
    },
    {
      "entropy": 5.349657106399536,
      "epoch": 0.684931506849315,
      "grad_norm": 1.125,
      "learning_rate": 0.0004959296192616305,
      "loss": 5.3061,
      "mean_token_accuracy": 0.18381480127573013,
      "num_tokens": 15079939.0,
      "step": 8200
    },
    {
      "entropy": 5.3600842475891115,
      "epoch": 0.6853491480120281,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004959239803455916,
      "loss": 5.1668,
      "mean_token_accuracy": 0.19448400437831878,
      "num_tokens": 15089391.0,
      "step": 8205
    },
    {
      "entropy": 5.251515674591064,
      "epoch": 0.6857667891747411,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004959183375620123,
      "loss": 5.1647,
      "mean_token_accuracy": 0.1877202644944191,
      "num_tokens": 15098875.0,
      "step": 8210
    },
    {
      "entropy": 5.311760473251343,
      "epoch": 0.686184430337454,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004959126909109918,
      "loss": 5.1946,
      "mean_token_accuracy": 0.18449883610010148,
      "num_tokens": 15108359.0,
      "step": 8215
    },
    {
      "entropy": 5.388326025009155,
      "epoch": 0.686602071500167,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004959070403926286,
      "loss": 5.3304,
      "mean_token_accuracy": 0.17387497872114183,
      "num_tokens": 15117957.0,
      "step": 8220
    },
    {
      "entropy": 5.376242208480835,
      "epoch": 0.6870197126628801,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004959013860070217,
      "loss": 5.2341,
      "mean_token_accuracy": 0.18277179151773454,
      "num_tokens": 15127913.0,
      "step": 8225
    },
    {
      "entropy": 5.3696400165557865,
      "epoch": 0.687437353825593,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004958957277542701,
      "loss": 5.2022,
      "mean_token_accuracy": 0.18735553920269013,
      "num_tokens": 15137588.0,
      "step": 8230
    },
    {
      "entropy": 5.3270057201385494,
      "epoch": 0.687854994988306,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004958900656344731,
      "loss": 5.2407,
      "mean_token_accuracy": 0.1827906534075737,
      "num_tokens": 15146147.0,
      "step": 8235
    },
    {
      "entropy": 5.247193956375122,
      "epoch": 0.6882726361510191,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004958843996477296,
      "loss": 5.121,
      "mean_token_accuracy": 0.19691976755857468,
      "num_tokens": 15156448.0,
      "step": 8240
    },
    {
      "entropy": 5.236945533752442,
      "epoch": 0.688690277313732,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004958787297941386,
      "loss": 5.166,
      "mean_token_accuracy": 0.19662522971630098,
      "num_tokens": 15165447.0,
      "step": 8245
    },
    {
      "entropy": 5.32114315032959,
      "epoch": 0.689107918476445,
      "grad_norm": 1.125,
      "learning_rate": 0.0004958730560737998,
      "loss": 5.2185,
      "mean_token_accuracy": 0.18877577781677246,
      "num_tokens": 15175046.0,
      "step": 8250
    },
    {
      "entropy": 5.3226992130279545,
      "epoch": 0.6895255596391581,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004958673784868121,
      "loss": 5.1568,
      "mean_token_accuracy": 0.19063031673431396,
      "num_tokens": 15184142.0,
      "step": 8255
    },
    {
      "entropy": 5.261743783950806,
      "epoch": 0.689943200801871,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004958616970332752,
      "loss": 5.1447,
      "mean_token_accuracy": 0.18793658018112183,
      "num_tokens": 15194101.0,
      "step": 8260
    },
    {
      "entropy": 5.241338300704956,
      "epoch": 0.690360841964584,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004958560117132884,
      "loss": 5.1632,
      "mean_token_accuracy": 0.1834757149219513,
      "num_tokens": 15203083.0,
      "step": 8265
    },
    {
      "entropy": 5.291897964477539,
      "epoch": 0.6907784831272971,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004958503225269513,
      "loss": 5.1542,
      "mean_token_accuracy": 0.19320226311683655,
      "num_tokens": 15210957.0,
      "step": 8270
    },
    {
      "entropy": 5.360385274887085,
      "epoch": 0.69119612429001,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004958446294743633,
      "loss": 5.2125,
      "mean_token_accuracy": 0.18672982901334761,
      "num_tokens": 15219144.0,
      "step": 8275
    },
    {
      "entropy": 5.328982257843018,
      "epoch": 0.691613765452723,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004958389325556245,
      "loss": 5.224,
      "mean_token_accuracy": 0.18534116446971893,
      "num_tokens": 15227633.0,
      "step": 8280
    },
    {
      "entropy": 5.376241874694824,
      "epoch": 0.692031406615436,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004958332317708341,
      "loss": 5.1659,
      "mean_token_accuracy": 0.1915632367134094,
      "num_tokens": 15235673.0,
      "step": 8285
    },
    {
      "entropy": 5.23722357749939,
      "epoch": 0.692449047778149,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004958275271200924,
      "loss": 5.1204,
      "mean_token_accuracy": 0.19583763778209687,
      "num_tokens": 15245164.0,
      "step": 8290
    },
    {
      "entropy": 5.217330074310302,
      "epoch": 0.692866688940862,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004958218186034988,
      "loss": 5.218,
      "mean_token_accuracy": 0.1901165008544922,
      "num_tokens": 15254654.0,
      "step": 8295
    },
    {
      "entropy": 5.287178087234497,
      "epoch": 0.693284330103575,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004958161062211537,
      "loss": 5.0609,
      "mean_token_accuracy": 0.19153740555047988,
      "num_tokens": 15264042.0,
      "step": 8300
    },
    {
      "entropy": 5.319094657897949,
      "epoch": 0.693701971266288,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004958103899731566,
      "loss": 5.3547,
      "mean_token_accuracy": 0.18011727482080458,
      "num_tokens": 15273215.0,
      "step": 8305
    },
    {
      "entropy": 5.276962995529175,
      "epoch": 0.694119612429001,
      "grad_norm": 1.125,
      "learning_rate": 0.0004958046698596081,
      "loss": 5.0908,
      "mean_token_accuracy": 0.19372446835041046,
      "num_tokens": 15281441.0,
      "step": 8310
    },
    {
      "entropy": 5.338572978973389,
      "epoch": 0.694537253591714,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004957989458806079,
      "loss": 5.216,
      "mean_token_accuracy": 0.18333572298288345,
      "num_tokens": 15290967.0,
      "step": 8315
    },
    {
      "entropy": 5.3263952255249025,
      "epoch": 0.694954894754427,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004957932180362564,
      "loss": 5.1989,
      "mean_token_accuracy": 0.18652485460042953,
      "num_tokens": 15301332.0,
      "step": 8320
    },
    {
      "entropy": 5.389624834060669,
      "epoch": 0.69537253591714,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004957874863266538,
      "loss": 5.2136,
      "mean_token_accuracy": 0.1886873036623001,
      "num_tokens": 15309547.0,
      "step": 8325
    },
    {
      "entropy": 5.417660903930664,
      "epoch": 0.695790177079853,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004957817507519006,
      "loss": 5.245,
      "mean_token_accuracy": 0.18211768120527266,
      "num_tokens": 15318841.0,
      "step": 8330
    },
    {
      "entropy": 5.354889583587647,
      "epoch": 0.696207818242566,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004957760113120971,
      "loss": 5.2924,
      "mean_token_accuracy": 0.1856311857700348,
      "num_tokens": 15327893.0,
      "step": 8335
    },
    {
      "entropy": 5.211272096633911,
      "epoch": 0.696625459405279,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004957702680073437,
      "loss": 5.1382,
      "mean_token_accuracy": 0.18985779732465743,
      "num_tokens": 15336320.0,
      "step": 8340
    },
    {
      "entropy": 5.325290536880493,
      "epoch": 0.697043100567992,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004957645208377411,
      "loss": 5.2085,
      "mean_token_accuracy": 0.18871812969446183,
      "num_tokens": 15346002.0,
      "step": 8345
    },
    {
      "entropy": 5.269504594802856,
      "epoch": 0.6974607417307049,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004957587698033898,
      "loss": 5.1459,
      "mean_token_accuracy": 0.1908067524433136,
      "num_tokens": 15355670.0,
      "step": 8350
    },
    {
      "entropy": 5.211881399154663,
      "epoch": 0.697878382893418,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004957530149043905,
      "loss": 5.0686,
      "mean_token_accuracy": 0.19455941915512084,
      "num_tokens": 15364207.0,
      "step": 8355
    },
    {
      "entropy": 5.260488557815552,
      "epoch": 0.698296024056131,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004957472561408438,
      "loss": 5.1657,
      "mean_token_accuracy": 0.19283827394247055,
      "num_tokens": 15373021.0,
      "step": 8360
    },
    {
      "entropy": 5.3370285987854,
      "epoch": 0.6987136652188439,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004957414935128509,
      "loss": 5.1794,
      "mean_token_accuracy": 0.19300519675016403,
      "num_tokens": 15382023.0,
      "step": 8365
    },
    {
      "entropy": 5.157944536209106,
      "epoch": 0.699131306381557,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004957357270205123,
      "loss": 5.1004,
      "mean_token_accuracy": 0.19824046194553374,
      "num_tokens": 15390829.0,
      "step": 8370
    },
    {
      "entropy": 5.320179271697998,
      "epoch": 0.69954894754427,
      "grad_norm": 1.125,
      "learning_rate": 0.0004957299566639292,
      "loss": 5.1777,
      "mean_token_accuracy": 0.18232600688934325,
      "num_tokens": 15400415.0,
      "step": 8375
    },
    {
      "entropy": 5.257931232452393,
      "epoch": 0.6999665887069829,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004957241824432025,
      "loss": 5.2009,
      "mean_token_accuracy": 0.18950661569833754,
      "num_tokens": 15408839.0,
      "step": 8380
    },
    {
      "entropy": 5.245182609558105,
      "epoch": 0.700384229869696,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004957184043584333,
      "loss": 5.0573,
      "mean_token_accuracy": 0.19407836645841597,
      "num_tokens": 15416989.0,
      "step": 8385
    },
    {
      "entropy": 5.273245906829834,
      "epoch": 0.700801871032409,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004957126224097227,
      "loss": 5.1053,
      "mean_token_accuracy": 0.19343802630901336,
      "num_tokens": 15425404.0,
      "step": 8390
    },
    {
      "entropy": 5.206933212280274,
      "epoch": 0.7012195121951219,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000495706836597172,
      "loss": 5.0341,
      "mean_token_accuracy": 0.20115216672420502,
      "num_tokens": 15433672.0,
      "step": 8395
    },
    {
      "entropy": 5.1540303230285645,
      "epoch": 0.7016371533578349,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004957010469208825,
      "loss": 5.0976,
      "mean_token_accuracy": 0.1932443007826805,
      "num_tokens": 15442345.0,
      "step": 8400
    },
    {
      "entropy": 5.235260820388794,
      "epoch": 0.702054794520548,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004956952533809554,
      "loss": 5.1279,
      "mean_token_accuracy": 0.19401500672101973,
      "num_tokens": 15451891.0,
      "step": 8405
    },
    {
      "entropy": 5.343758869171142,
      "epoch": 0.7024724356832609,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004956894559774923,
      "loss": 5.2138,
      "mean_token_accuracy": 0.18704367130994798,
      "num_tokens": 15462192.0,
      "step": 8410
    },
    {
      "entropy": 5.3598902225494385,
      "epoch": 0.7028900768459739,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004956836547105947,
      "loss": 5.294,
      "mean_token_accuracy": 0.182576522231102,
      "num_tokens": 15471723.0,
      "step": 8415
    },
    {
      "entropy": 5.315639162063599,
      "epoch": 0.703307718008687,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004956778495803641,
      "loss": 5.0883,
      "mean_token_accuracy": 0.19733315110206603,
      "num_tokens": 15479621.0,
      "step": 8420
    },
    {
      "entropy": 5.274339628219605,
      "epoch": 0.7037253591713999,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000495672040586902,
      "loss": 5.1801,
      "mean_token_accuracy": 0.18573080152273178,
      "num_tokens": 15489180.0,
      "step": 8425
    },
    {
      "entropy": 5.183819150924682,
      "epoch": 0.7041430003341129,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004956662277303103,
      "loss": 5.1504,
      "mean_token_accuracy": 0.1898150399327278,
      "num_tokens": 15498155.0,
      "step": 8430
    },
    {
      "entropy": 5.42573618888855,
      "epoch": 0.704560641496826,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004956604110106905,
      "loss": 5.2491,
      "mean_token_accuracy": 0.18965250998735428,
      "num_tokens": 15507944.0,
      "step": 8435
    },
    {
      "entropy": 5.2981466293334964,
      "epoch": 0.7049782826595389,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004956545904281447,
      "loss": 5.2097,
      "mean_token_accuracy": 0.1860172912478447,
      "num_tokens": 15516800.0,
      "step": 8440
    },
    {
      "entropy": 5.25133867263794,
      "epoch": 0.7053959238222519,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004956487659827748,
      "loss": 5.1572,
      "mean_token_accuracy": 0.18447698205709456,
      "num_tokens": 15525824.0,
      "step": 8445
    },
    {
      "entropy": 5.3067363739013675,
      "epoch": 0.705813564984965,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004956429376746825,
      "loss": 5.1804,
      "mean_token_accuracy": 0.19260694533586503,
      "num_tokens": 15534954.0,
      "step": 8450
    },
    {
      "entropy": 5.251150894165039,
      "epoch": 0.7062312061476779,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004956371055039701,
      "loss": 5.1375,
      "mean_token_accuracy": 0.20133766233921052,
      "num_tokens": 15543439.0,
      "step": 8455
    },
    {
      "entropy": 5.237839937210083,
      "epoch": 0.7066488473103909,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004956312694707395,
      "loss": 5.1059,
      "mean_token_accuracy": 0.19485290348529816,
      "num_tokens": 15552768.0,
      "step": 8460
    },
    {
      "entropy": 5.35594367980957,
      "epoch": 0.7070664884731039,
      "grad_norm": 1.1875,
      "learning_rate": 0.000495625429575093,
      "loss": 5.3146,
      "mean_token_accuracy": 0.18608065098524093,
      "num_tokens": 15562193.0,
      "step": 8465
    },
    {
      "entropy": 5.347560262680053,
      "epoch": 0.7074841296358169,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004956195858171327,
      "loss": 5.1624,
      "mean_token_accuracy": 0.19208250790834427,
      "num_tokens": 15571403.0,
      "step": 8470
    },
    {
      "entropy": 5.253741312026977,
      "epoch": 0.7079017707985299,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004956137381969611,
      "loss": 5.1244,
      "mean_token_accuracy": 0.19098494946956635,
      "num_tokens": 15580477.0,
      "step": 8475
    },
    {
      "entropy": 5.242291355133057,
      "epoch": 0.7083194119612429,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004956078867146805,
      "loss": 5.2093,
      "mean_token_accuracy": 0.19199638515710832,
      "num_tokens": 15590106.0,
      "step": 8480
    },
    {
      "entropy": 5.384551525115967,
      "epoch": 0.7087370531239559,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004956020313703933,
      "loss": 5.1951,
      "mean_token_accuracy": 0.1891821801662445,
      "num_tokens": 15598939.0,
      "step": 8485
    },
    {
      "entropy": 5.258433818817139,
      "epoch": 0.7091546942866689,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004955961721642018,
      "loss": 5.1698,
      "mean_token_accuracy": 0.18928396254777907,
      "num_tokens": 15607439.0,
      "step": 8490
    },
    {
      "entropy": 5.258288764953614,
      "epoch": 0.7095723354493819,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000495590309096209,
      "loss": 5.2011,
      "mean_token_accuracy": 0.18600573688745498,
      "num_tokens": 15616887.0,
      "step": 8495
    },
    {
      "entropy": 5.27916522026062,
      "epoch": 0.7099899766120948,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004955844421665171,
      "loss": 5.1435,
      "mean_token_accuracy": 0.19364818334579467,
      "num_tokens": 15626399.0,
      "step": 8500
    },
    {
      "entropy": 5.242170810699463,
      "epoch": 0.7104076177748079,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004955785713752292,
      "loss": 5.165,
      "mean_token_accuracy": 0.18425308167934418,
      "num_tokens": 15635719.0,
      "step": 8505
    },
    {
      "entropy": 5.2916895866394045,
      "epoch": 0.7108252589375209,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004955726967224479,
      "loss": 5.1352,
      "mean_token_accuracy": 0.19397978931665422,
      "num_tokens": 15644504.0,
      "step": 8510
    },
    {
      "entropy": 5.242740106582642,
      "epoch": 0.7112429001002338,
      "grad_norm": 1.125,
      "learning_rate": 0.000495566818208276,
      "loss": 5.0882,
      "mean_token_accuracy": 0.19888446629047393,
      "num_tokens": 15653842.0,
      "step": 8515
    },
    {
      "entropy": 5.276203393936157,
      "epoch": 0.7116605412629469,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004955609358328163,
      "loss": 5.2757,
      "mean_token_accuracy": 0.1843859702348709,
      "num_tokens": 15664235.0,
      "step": 8520
    },
    {
      "entropy": 5.311121940612793,
      "epoch": 0.7120781824256599,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004955550495961722,
      "loss": 5.1211,
      "mean_token_accuracy": 0.19658908993005753,
      "num_tokens": 15673113.0,
      "step": 8525
    },
    {
      "entropy": 5.311741638183594,
      "epoch": 0.7124958235883728,
      "grad_norm": 1.125,
      "learning_rate": 0.0004955491594984462,
      "loss": 5.1894,
      "mean_token_accuracy": 0.19137491434812545,
      "num_tokens": 15682609.0,
      "step": 8530
    },
    {
      "entropy": 5.231457662582398,
      "epoch": 0.7129134647510859,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004955432655397419,
      "loss": 5.2471,
      "mean_token_accuracy": 0.18539536893367767,
      "num_tokens": 15691414.0,
      "step": 8535
    },
    {
      "entropy": 5.317198991775513,
      "epoch": 0.7133311059137989,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004955373677201621,
      "loss": 5.1667,
      "mean_token_accuracy": 0.18964322805404663,
      "num_tokens": 15700690.0,
      "step": 8540
    },
    {
      "entropy": 5.28790979385376,
      "epoch": 0.7137487470765118,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004955314660398105,
      "loss": 5.0807,
      "mean_token_accuracy": 0.19457218945026397,
      "num_tokens": 15710110.0,
      "step": 8545
    },
    {
      "entropy": 5.256107902526855,
      "epoch": 0.7141663882392248,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004955255604987899,
      "loss": 5.1268,
      "mean_token_accuracy": 0.18826364278793334,
      "num_tokens": 15719182.0,
      "step": 8550
    },
    {
      "entropy": 5.258205127716065,
      "epoch": 0.7145840294019379,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000495519651097204,
      "loss": 5.1771,
      "mean_token_accuracy": 0.19180652499198914,
      "num_tokens": 15729045.0,
      "step": 8555
    },
    {
      "entropy": 5.24560055732727,
      "epoch": 0.7150016705646508,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004955137378351561,
      "loss": 5.1445,
      "mean_token_accuracy": 0.19240989536046982,
      "num_tokens": 15738312.0,
      "step": 8560
    },
    {
      "entropy": 5.219484615325928,
      "epoch": 0.7154193117273638,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004955078207127498,
      "loss": 5.1576,
      "mean_token_accuracy": 0.19288695603609085,
      "num_tokens": 15747852.0,
      "step": 8565
    },
    {
      "entropy": 5.335433578491211,
      "epoch": 0.7158369528900769,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004955018997300886,
      "loss": 5.1732,
      "mean_token_accuracy": 0.19398155361413955,
      "num_tokens": 15757593.0,
      "step": 8570
    },
    {
      "entropy": 5.322363948822021,
      "epoch": 0.7162545940527898,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004954959748872764,
      "loss": 5.1552,
      "mean_token_accuracy": 0.19538125097751619,
      "num_tokens": 15767740.0,
      "step": 8575
    },
    {
      "entropy": 5.239190530776978,
      "epoch": 0.7166722352155028,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004954900461844167,
      "loss": 5.1197,
      "mean_token_accuracy": 0.20100190937519075,
      "num_tokens": 15776306.0,
      "step": 8580
    },
    {
      "entropy": 5.238163805007934,
      "epoch": 0.7170898763782159,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004954841136216132,
      "loss": 5.1283,
      "mean_token_accuracy": 0.1930987611413002,
      "num_tokens": 15784674.0,
      "step": 8585
    },
    {
      "entropy": 5.297255373001098,
      "epoch": 0.7175075175409288,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004954781771989701,
      "loss": 5.145,
      "mean_token_accuracy": 0.19163012653589248,
      "num_tokens": 15794262.0,
      "step": 8590
    },
    {
      "entropy": 5.283347129821777,
      "epoch": 0.7179251587036418,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000495472236916591,
      "loss": 5.2225,
      "mean_token_accuracy": 0.18070160895586013,
      "num_tokens": 15805853.0,
      "step": 8595
    },
    {
      "entropy": 5.293683576583862,
      "epoch": 0.7183427998663549,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004954662927745799,
      "loss": 5.2048,
      "mean_token_accuracy": 0.19223368018865586,
      "num_tokens": 15815426.0,
      "step": 8600
    },
    {
      "entropy": 5.289098787307739,
      "epoch": 0.7187604410290678,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004954603447730411,
      "loss": 5.1729,
      "mean_token_accuracy": 0.19176294207572936,
      "num_tokens": 15825121.0,
      "step": 8605
    },
    {
      "entropy": 5.1976173400878904,
      "epoch": 0.7191780821917808,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004954543929120786,
      "loss": 5.1277,
      "mean_token_accuracy": 0.1959634006023407,
      "num_tokens": 15834649.0,
      "step": 8610
    },
    {
      "entropy": 5.219401264190674,
      "epoch": 0.7195957233544938,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004954484371917967,
      "loss": 5.1773,
      "mean_token_accuracy": 0.18379189521074296,
      "num_tokens": 15844583.0,
      "step": 8615
    },
    {
      "entropy": 5.313754844665527,
      "epoch": 0.7200133645172068,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004954424776122993,
      "loss": 5.1123,
      "mean_token_accuracy": 0.19071279764175414,
      "num_tokens": 15853169.0,
      "step": 8620
    },
    {
      "entropy": 5.2378050804138185,
      "epoch": 0.7204310056799198,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004954365141736913,
      "loss": 5.1412,
      "mean_token_accuracy": 0.19123534709215165,
      "num_tokens": 15862139.0,
      "step": 8625
    },
    {
      "entropy": 5.376718235015869,
      "epoch": 0.7208486468426328,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004954305468760765,
      "loss": 5.2345,
      "mean_token_accuracy": 0.1848651185631752,
      "num_tokens": 15871415.0,
      "step": 8630
    },
    {
      "entropy": 5.185443925857544,
      "epoch": 0.7212662880053458,
      "grad_norm": 1.125,
      "learning_rate": 0.0004954245757195598,
      "loss": 5.0595,
      "mean_token_accuracy": 0.20612120032310485,
      "num_tokens": 15880581.0,
      "step": 8635
    },
    {
      "entropy": 5.2180744171142575,
      "epoch": 0.7216839291680588,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004954186007042454,
      "loss": 5.1278,
      "mean_token_accuracy": 0.18776457458734513,
      "num_tokens": 15889142.0,
      "step": 8640
    },
    {
      "entropy": 5.246136474609375,
      "epoch": 0.7221015703307718,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004954126218302382,
      "loss": 5.143,
      "mean_token_accuracy": 0.19433717876672746,
      "num_tokens": 15898058.0,
      "step": 8645
    },
    {
      "entropy": 5.1824267387390135,
      "epoch": 0.7225192114934847,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004954066390976427,
      "loss": 5.078,
      "mean_token_accuracy": 0.19284000992774963,
      "num_tokens": 15907196.0,
      "step": 8650
    },
    {
      "entropy": 5.234852409362793,
      "epoch": 0.7229368526561978,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004954006525065639,
      "loss": 5.0927,
      "mean_token_accuracy": 0.1937478795647621,
      "num_tokens": 15916291.0,
      "step": 8655
    },
    {
      "entropy": 5.262243700027466,
      "epoch": 0.7233544938189108,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004953946620571061,
      "loss": 5.1262,
      "mean_token_accuracy": 0.18791772276163102,
      "num_tokens": 15924846.0,
      "step": 8660
    },
    {
      "entropy": 5.194462299346924,
      "epoch": 0.7237721349816237,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004953886677493746,
      "loss": 5.1045,
      "mean_token_accuracy": 0.1966923326253891,
      "num_tokens": 15934406.0,
      "step": 8665
    },
    {
      "entropy": 5.302021646499634,
      "epoch": 0.7241897761443368,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004953826695834741,
      "loss": 5.1039,
      "mean_token_accuracy": 0.19082928746938704,
      "num_tokens": 15943481.0,
      "step": 8670
    },
    {
      "entropy": 5.361473083496094,
      "epoch": 0.7246074173070498,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004953766675595098,
      "loss": 5.2415,
      "mean_token_accuracy": 0.18488221615552902,
      "num_tokens": 15952787.0,
      "step": 8675
    },
    {
      "entropy": 5.24317774772644,
      "epoch": 0.7250250584697627,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004953706616775866,
      "loss": 5.1775,
      "mean_token_accuracy": 0.19052420258522035,
      "num_tokens": 15962295.0,
      "step": 8680
    },
    {
      "entropy": 5.322543287277222,
      "epoch": 0.7254426996324758,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004953646519378097,
      "loss": 5.3223,
      "mean_token_accuracy": 0.19563299417495728,
      "num_tokens": 15971367.0,
      "step": 8685
    },
    {
      "entropy": 5.19252119064331,
      "epoch": 0.7258603407951888,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004953586383402844,
      "loss": 5.1053,
      "mean_token_accuracy": 0.19953658282756806,
      "num_tokens": 15981535.0,
      "step": 8690
    },
    {
      "entropy": 5.28391318321228,
      "epoch": 0.7262779819579017,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004953526208851157,
      "loss": 5.2329,
      "mean_token_accuracy": 0.18641686141490937,
      "num_tokens": 15991067.0,
      "step": 8695
    },
    {
      "entropy": 5.215488290786743,
      "epoch": 0.7266956231206148,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004953465995724092,
      "loss": 5.0331,
      "mean_token_accuracy": 0.1963928982615471,
      "num_tokens": 15999310.0,
      "step": 8700
    },
    {
      "entropy": 5.211255645751953,
      "epoch": 0.7271132642833278,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004953405744022705,
      "loss": 5.0615,
      "mean_token_accuracy": 0.19351856261491776,
      "num_tokens": 16008574.0,
      "step": 8705
    },
    {
      "entropy": 5.2873673915863035,
      "epoch": 0.7275309054460407,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004953345453748045,
      "loss": 5.1352,
      "mean_token_accuracy": 0.1901836633682251,
      "num_tokens": 16017208.0,
      "step": 8710
    },
    {
      "entropy": 5.331624221801758,
      "epoch": 0.7279485466087537,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004953285124901171,
      "loss": 5.2795,
      "mean_token_accuracy": 0.18591621071100234,
      "num_tokens": 16027009.0,
      "step": 8715
    },
    {
      "entropy": 5.237117099761963,
      "epoch": 0.7283661877714668,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000495322475748314,
      "loss": 5.1043,
      "mean_token_accuracy": 0.20092488080263138,
      "num_tokens": 16035019.0,
      "step": 8720
    },
    {
      "entropy": 5.221333694458008,
      "epoch": 0.7287838289341797,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004953164351495007,
      "loss": 5.0904,
      "mean_token_accuracy": 0.198664990067482,
      "num_tokens": 16044373.0,
      "step": 8725
    },
    {
      "entropy": 5.257413673400879,
      "epoch": 0.7292014700968927,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004953103906937831,
      "loss": 5.1284,
      "mean_token_accuracy": 0.19224874079227447,
      "num_tokens": 16052888.0,
      "step": 8730
    },
    {
      "entropy": 5.236996507644653,
      "epoch": 0.7296191112596058,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004953043423812669,
      "loss": 5.118,
      "mean_token_accuracy": 0.1905507206916809,
      "num_tokens": 16062137.0,
      "step": 8735
    },
    {
      "entropy": 5.299497938156128,
      "epoch": 0.7300367524223187,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004952982902120579,
      "loss": 5.2945,
      "mean_token_accuracy": 0.18409110754728317,
      "num_tokens": 16072755.0,
      "step": 8740
    },
    {
      "entropy": 5.306175708770752,
      "epoch": 0.7304543935850317,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004952922341862622,
      "loss": 5.1622,
      "mean_token_accuracy": 0.1882558822631836,
      "num_tokens": 16081069.0,
      "step": 8745
    },
    {
      "entropy": 5.331820440292359,
      "epoch": 0.7308720347477448,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004952861743039858,
      "loss": 5.1903,
      "mean_token_accuracy": 0.1941478058695793,
      "num_tokens": 16089938.0,
      "step": 8750
    },
    {
      "entropy": 5.21022253036499,
      "epoch": 0.7312896759104577,
      "grad_norm": 1.25,
      "learning_rate": 0.0004952801105653347,
      "loss": 5.1661,
      "mean_token_accuracy": 0.19628458470106125,
      "num_tokens": 16099395.0,
      "step": 8755
    },
    {
      "entropy": 5.175596570968628,
      "epoch": 0.7317073170731707,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004952740429704152,
      "loss": 5.0466,
      "mean_token_accuracy": 0.20078281313180923,
      "num_tokens": 16108196.0,
      "step": 8760
    },
    {
      "entropy": 5.236020612716675,
      "epoch": 0.7321249582358837,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004952679715193334,
      "loss": 5.1502,
      "mean_token_accuracy": 0.19143548905849456,
      "num_tokens": 16118059.0,
      "step": 8765
    },
    {
      "entropy": 5.30237340927124,
      "epoch": 0.7325425993985967,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004952618962121956,
      "loss": 5.2095,
      "mean_token_accuracy": 0.18966662883758545,
      "num_tokens": 16126981.0,
      "step": 8770
    },
    {
      "entropy": 5.2454451560974125,
      "epoch": 0.7329602405613097,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004952558170491082,
      "loss": 5.0849,
      "mean_token_accuracy": 0.19678558707237243,
      "num_tokens": 16136164.0,
      "step": 8775
    },
    {
      "entropy": 5.283247566223144,
      "epoch": 0.7333778817240227,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004952497340301776,
      "loss": 5.2405,
      "mean_token_accuracy": 0.18060089349746705,
      "num_tokens": 16145739.0,
      "step": 8780
    },
    {
      "entropy": 5.313655376434326,
      "epoch": 0.7337955228867357,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004952436471555103,
      "loss": 5.1639,
      "mean_token_accuracy": 0.18898289054632186,
      "num_tokens": 16155092.0,
      "step": 8785
    },
    {
      "entropy": 5.204944849014282,
      "epoch": 0.7342131640494487,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004952375564252128,
      "loss": 5.065,
      "mean_token_accuracy": 0.1982123538851738,
      "num_tokens": 16162994.0,
      "step": 8790
    },
    {
      "entropy": 5.27227931022644,
      "epoch": 0.7346308052121617,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004952314618393919,
      "loss": 5.1113,
      "mean_token_accuracy": 0.19368191957473754,
      "num_tokens": 16171286.0,
      "step": 8795
    },
    {
      "entropy": 5.29538369178772,
      "epoch": 0.7350484463748747,
      "grad_norm": 1.375,
      "learning_rate": 0.000495225363398154,
      "loss": 5.1919,
      "mean_token_accuracy": 0.1902650013566017,
      "num_tokens": 16180459.0,
      "step": 8800
    },
    {
      "entropy": 5.280352401733398,
      "epoch": 0.7354660875375877,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004952192611016061,
      "loss": 5.2056,
      "mean_token_accuracy": 0.1910347819328308,
      "num_tokens": 16190103.0,
      "step": 8805
    },
    {
      "entropy": 5.285508012771606,
      "epoch": 0.7358837287003007,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000495213154949855,
      "loss": 5.1054,
      "mean_token_accuracy": 0.19194310158491135,
      "num_tokens": 16199499.0,
      "step": 8810
    },
    {
      "entropy": 5.222455358505249,
      "epoch": 0.7363013698630136,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004952070449430075,
      "loss": 5.1914,
      "mean_token_accuracy": 0.1873357117176056,
      "num_tokens": 16209589.0,
      "step": 8815
    },
    {
      "entropy": 5.171773052215576,
      "epoch": 0.7367190110257267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004952009310811706,
      "loss": 5.0333,
      "mean_token_accuracy": 0.19819573312997818,
      "num_tokens": 16218499.0,
      "step": 8820
    },
    {
      "entropy": 5.20752968788147,
      "epoch": 0.7371366521884397,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004951948133644514,
      "loss": 5.0135,
      "mean_token_accuracy": 0.1979145109653473,
      "num_tokens": 16227183.0,
      "step": 8825
    },
    {
      "entropy": 5.244198894500732,
      "epoch": 0.7375542933511526,
      "grad_norm": 1.15625,
      "learning_rate": 0.000495188691792957,
      "loss": 5.1393,
      "mean_token_accuracy": 0.19220415204763414,
      "num_tokens": 16236573.0,
      "step": 8830
    },
    {
      "entropy": 5.285165500640869,
      "epoch": 0.7379719345138657,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004951825663667943,
      "loss": 5.1162,
      "mean_token_accuracy": 0.19057226926088333,
      "num_tokens": 16246399.0,
      "step": 8835
    },
    {
      "entropy": 5.216931533813477,
      "epoch": 0.7383895756765787,
      "grad_norm": 1.171875,
      "learning_rate": 0.000495176437086071,
      "loss": 5.2118,
      "mean_token_accuracy": 0.18463050872087478,
      "num_tokens": 16254964.0,
      "step": 8840
    },
    {
      "entropy": 5.303437376022339,
      "epoch": 0.7388072168392917,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000495170303950894,
      "loss": 5.2175,
      "mean_token_accuracy": 0.19016121625900267,
      "num_tokens": 16263579.0,
      "step": 8845
    },
    {
      "entropy": 5.350651025772095,
      "epoch": 0.7392248580020047,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004951641669613708,
      "loss": 5.2481,
      "mean_token_accuracy": 0.1829417511820793,
      "num_tokens": 16273000.0,
      "step": 8850
    },
    {
      "entropy": 5.324179792404175,
      "epoch": 0.7396424991647177,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004951580261176088,
      "loss": 5.1392,
      "mean_token_accuracy": 0.19244228303432465,
      "num_tokens": 16282582.0,
      "step": 8855
    },
    {
      "entropy": 5.201522397994995,
      "epoch": 0.7400601403274307,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004951518814197156,
      "loss": 5.091,
      "mean_token_accuracy": 0.19175723642110826,
      "num_tokens": 16291153.0,
      "step": 8860
    },
    {
      "entropy": 5.203221988677979,
      "epoch": 0.7404777814901436,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004951457328677989,
      "loss": 5.1401,
      "mean_token_accuracy": 0.19804461747407914,
      "num_tokens": 16300527.0,
      "step": 8865
    },
    {
      "entropy": 5.2294846534729,
      "epoch": 0.7408954226528567,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004951395804619659,
      "loss": 5.0197,
      "mean_token_accuracy": 0.2092771887779236,
      "num_tokens": 16309111.0,
      "step": 8870
    },
    {
      "entropy": 5.326550579071045,
      "epoch": 0.7413130638155697,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004951334242023247,
      "loss": 5.1978,
      "mean_token_accuracy": 0.18758321106433867,
      "num_tokens": 16318409.0,
      "step": 8875
    },
    {
      "entropy": 5.227953815460205,
      "epoch": 0.7417307049782826,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004951272640889828,
      "loss": 5.1295,
      "mean_token_accuracy": 0.19959003776311873,
      "num_tokens": 16327950.0,
      "step": 8880
    },
    {
      "entropy": 5.255214786529541,
      "epoch": 0.7421483461409957,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004951211001220484,
      "loss": 5.197,
      "mean_token_accuracy": 0.1865760087966919,
      "num_tokens": 16336776.0,
      "step": 8885
    },
    {
      "entropy": 5.3541590690612795,
      "epoch": 0.7425659873037087,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000495114932301629,
      "loss": 5.2186,
      "mean_token_accuracy": 0.18943388164043426,
      "num_tokens": 16346206.0,
      "step": 8890
    },
    {
      "entropy": 5.288184404373169,
      "epoch": 0.7429836284664216,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004951087606278328,
      "loss": 5.0831,
      "mean_token_accuracy": 0.194120354950428,
      "num_tokens": 16355163.0,
      "step": 8895
    },
    {
      "entropy": 5.276929664611816,
      "epoch": 0.7434012696291347,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004951025851007678,
      "loss": 5.1663,
      "mean_token_accuracy": 0.18412055671215058,
      "num_tokens": 16364028.0,
      "step": 8900
    },
    {
      "entropy": 5.255345296859741,
      "epoch": 0.7438189107918477,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004950964057205421,
      "loss": 5.1282,
      "mean_token_accuracy": 0.19027332961559296,
      "num_tokens": 16373669.0,
      "step": 8905
    },
    {
      "entropy": 5.321865844726562,
      "epoch": 0.7442365519545606,
      "grad_norm": 1.125,
      "learning_rate": 0.000495090222487264,
      "loss": 5.1548,
      "mean_token_accuracy": 0.1864067181944847,
      "num_tokens": 16383246.0,
      "step": 8910
    },
    {
      "entropy": 5.319370985031128,
      "epoch": 0.7446541931172737,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004950840354010415,
      "loss": 5.23,
      "mean_token_accuracy": 0.18542566746473313,
      "num_tokens": 16393589.0,
      "step": 8915
    },
    {
      "entropy": 5.220500659942627,
      "epoch": 0.7450718342799867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004950778444619831,
      "loss": 5.1178,
      "mean_token_accuracy": 0.18315428346395493,
      "num_tokens": 16402734.0,
      "step": 8920
    },
    {
      "entropy": 5.3359418392181395,
      "epoch": 0.7454894754426996,
      "grad_norm": 1.1875,
      "learning_rate": 0.000495071649670197,
      "loss": 5.2575,
      "mean_token_accuracy": 0.1865530490875244,
      "num_tokens": 16412317.0,
      "step": 8925
    },
    {
      "entropy": 5.311035108566284,
      "epoch": 0.7459071166054126,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000495065451025792,
      "loss": 5.1744,
      "mean_token_accuracy": 0.1900661453604698,
      "num_tokens": 16421472.0,
      "step": 8930
    },
    {
      "entropy": 5.20952353477478,
      "epoch": 0.7463247577681257,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004950592485288761,
      "loss": 5.0832,
      "mean_token_accuracy": 0.18919856995344161,
      "num_tokens": 16430694.0,
      "step": 8935
    },
    {
      "entropy": 5.282927417755127,
      "epoch": 0.7467423989308386,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004950530421795584,
      "loss": 5.2792,
      "mean_token_accuracy": 0.181670643389225,
      "num_tokens": 16440250.0,
      "step": 8940
    },
    {
      "entropy": 5.255068063735962,
      "epoch": 0.7471600400935516,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004950468319779472,
      "loss": 5.0908,
      "mean_token_accuracy": 0.19686186760663987,
      "num_tokens": 16449294.0,
      "step": 8945
    },
    {
      "entropy": 5.463743400573731,
      "epoch": 0.7475776812562647,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004950406179241515,
      "loss": 5.3551,
      "mean_token_accuracy": 0.17092504799365998,
      "num_tokens": 16459614.0,
      "step": 8950
    },
    {
      "entropy": 5.351200723648072,
      "epoch": 0.7479953224189776,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004950344000182798,
      "loss": 5.2253,
      "mean_token_accuracy": 0.1823495149612427,
      "num_tokens": 16469175.0,
      "step": 8955
    },
    {
      "entropy": 5.213696718215942,
      "epoch": 0.7484129635816906,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004950281782604411,
      "loss": 5.0605,
      "mean_token_accuracy": 0.19510286897420884,
      "num_tokens": 16477818.0,
      "step": 8960
    },
    {
      "entropy": 5.314765119552613,
      "epoch": 0.7488306047444037,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004950219526507442,
      "loss": 5.2133,
      "mean_token_accuracy": 0.1890265390276909,
      "num_tokens": 16487807.0,
      "step": 8965
    },
    {
      "entropy": 5.189211702346801,
      "epoch": 0.7492482459071166,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004950157231892983,
      "loss": 5.1089,
      "mean_token_accuracy": 0.2005615085363388,
      "num_tokens": 16497924.0,
      "step": 8970
    },
    {
      "entropy": 5.30037260055542,
      "epoch": 0.7496658870698296,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004950094898762123,
      "loss": 5.1715,
      "mean_token_accuracy": 0.1895117536187172,
      "num_tokens": 16508069.0,
      "step": 8975
    },
    {
      "entropy": 5.356766653060913,
      "epoch": 0.7500835282325427,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004950032527115954,
      "loss": 5.1945,
      "mean_token_accuracy": 0.19483387917280198,
      "num_tokens": 16517226.0,
      "step": 8980
    },
    {
      "entropy": 5.25833420753479,
      "epoch": 0.7505011693952556,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004949970116955566,
      "loss": 5.1704,
      "mean_token_accuracy": 0.17950759083032608,
      "num_tokens": 16526997.0,
      "step": 8985
    },
    {
      "entropy": 5.2498777389526365,
      "epoch": 0.7509188105579686,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004949907668282054,
      "loss": 5.1067,
      "mean_token_accuracy": 0.19318631738424302,
      "num_tokens": 16536772.0,
      "step": 8990
    },
    {
      "entropy": 5.23926477432251,
      "epoch": 0.7513364517206816,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000494984518109651,
      "loss": 5.0273,
      "mean_token_accuracy": 0.19715453833341598,
      "num_tokens": 16544901.0,
      "step": 8995
    },
    {
      "entropy": 5.190441751480103,
      "epoch": 0.7517540928833946,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004949782655400028,
      "loss": 5.106,
      "mean_token_accuracy": 0.19497038722038268,
      "num_tokens": 16553628.0,
      "step": 9000
    },
    {
      "epoch": 0.7517540928833946,
      "eval_entropy": 5.04891709866248,
      "eval_loss": 5.1309638023376465,
      "eval_mean_token_accuracy": 0.20150274327594933,
      "eval_num_tokens": 16553628.0,
      "eval_runtime": 27.7068,
      "eval_samples_per_second": 1362.948,
      "eval_steps_per_second": 170.391,
      "step": 9000
    },
    {
      "entropy": 5.284650564193726,
      "epoch": 0.7521717340461076,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004949720091193704,
      "loss": 5.1901,
      "mean_token_accuracy": 0.19129439890384675,
      "num_tokens": 16561619.0,
      "step": 9005
    },
    {
      "entropy": 5.236381864547729,
      "epoch": 0.7525893752088206,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000494965748847863,
      "loss": 5.1756,
      "mean_token_accuracy": 0.1949136033654213,
      "num_tokens": 16571316.0,
      "step": 9010
    },
    {
      "entropy": 5.305360794067383,
      "epoch": 0.7530070163715336,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004949594847255905,
      "loss": 5.2397,
      "mean_token_accuracy": 0.1870893120765686,
      "num_tokens": 16582014.0,
      "step": 9015
    },
    {
      "entropy": 5.26864538192749,
      "epoch": 0.7534246575342466,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004949532167526625,
      "loss": 5.1495,
      "mean_token_accuracy": 0.1951904073357582,
      "num_tokens": 16590179.0,
      "step": 9020
    },
    {
      "entropy": 5.282690906524659,
      "epoch": 0.7538422986969596,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004949469449291886,
      "loss": 5.0531,
      "mean_token_accuracy": 0.2060410797595978,
      "num_tokens": 16598732.0,
      "step": 9025
    },
    {
      "entropy": 5.315931081771851,
      "epoch": 0.7542599398596725,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004949406692552787,
      "loss": 5.2842,
      "mean_token_accuracy": 0.18190824687480928,
      "num_tokens": 16608844.0,
      "step": 9030
    },
    {
      "entropy": 5.31307282447815,
      "epoch": 0.7546775810223856,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004949343897310426,
      "loss": 5.1897,
      "mean_token_accuracy": 0.19130521267652512,
      "num_tokens": 16618969.0,
      "step": 9035
    },
    {
      "entropy": 5.309460687637329,
      "epoch": 0.7550952221850986,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004949281063565904,
      "loss": 5.2181,
      "mean_token_accuracy": 0.18992851078510284,
      "num_tokens": 16629081.0,
      "step": 9040
    },
    {
      "entropy": 5.191410732269287,
      "epoch": 0.7555128633478115,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004949218191320319,
      "loss": 5.0629,
      "mean_token_accuracy": 0.20415098816156388,
      "num_tokens": 16637588.0,
      "step": 9045
    },
    {
      "entropy": 5.301031255722046,
      "epoch": 0.7559305045105246,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004949155280574772,
      "loss": 5.1779,
      "mean_token_accuracy": 0.18461380451917647,
      "num_tokens": 16647763.0,
      "step": 9050
    },
    {
      "entropy": 5.196663284301758,
      "epoch": 0.7563481456732376,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004949092331330365,
      "loss": 5.0732,
      "mean_token_accuracy": 0.19086116552352905,
      "num_tokens": 16657208.0,
      "step": 9055
    },
    {
      "entropy": 5.267312288284302,
      "epoch": 0.7567657868359505,
      "grad_norm": 1.359375,
      "learning_rate": 0.00049490293435882,
      "loss": 5.0479,
      "mean_token_accuracy": 0.19404009878635406,
      "num_tokens": 16666375.0,
      "step": 9060
    },
    {
      "entropy": 5.252836084365844,
      "epoch": 0.7571834279986636,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004948966317349379,
      "loss": 5.0577,
      "mean_token_accuracy": 0.20120500922203063,
      "num_tokens": 16674847.0,
      "step": 9065
    },
    {
      "entropy": 5.229509449005127,
      "epoch": 0.7576010691613766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004948903252615006,
      "loss": 5.2056,
      "mean_token_accuracy": 0.18377767950296403,
      "num_tokens": 16684828.0,
      "step": 9070
    },
    {
      "entropy": 5.327714920043945,
      "epoch": 0.7580187103240895,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004948840149386184,
      "loss": 5.2967,
      "mean_token_accuracy": 0.18056150674819946,
      "num_tokens": 16694151.0,
      "step": 9075
    },
    {
      "entropy": 5.368087100982666,
      "epoch": 0.7584363514868026,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000494877700766402,
      "loss": 5.183,
      "mean_token_accuracy": 0.18970449417829513,
      "num_tokens": 16703584.0,
      "step": 9080
    },
    {
      "entropy": 5.225856018066406,
      "epoch": 0.7588539926495156,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004948713827449616,
      "loss": 5.0945,
      "mean_token_accuracy": 0.19073991924524308,
      "num_tokens": 16712593.0,
      "step": 9085
    },
    {
      "entropy": 5.2489885807037355,
      "epoch": 0.7592716338122285,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004948650608744081,
      "loss": 5.074,
      "mean_token_accuracy": 0.19166267067193984,
      "num_tokens": 16721848.0,
      "step": 9090
    },
    {
      "entropy": 5.303568649291992,
      "epoch": 0.7596892749749415,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000494858735154852,
      "loss": 5.1778,
      "mean_token_accuracy": 0.19576231390237808,
      "num_tokens": 16730465.0,
      "step": 9095
    },
    {
      "entropy": 5.268704795837403,
      "epoch": 0.7601069161376546,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004948524055864041,
      "loss": 5.1619,
      "mean_token_accuracy": 0.19145216047763824,
      "num_tokens": 16739518.0,
      "step": 9100
    },
    {
      "entropy": 5.322190666198731,
      "epoch": 0.7605245573003675,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004948460721691752,
      "loss": 5.2309,
      "mean_token_accuracy": 0.18203989714384078,
      "num_tokens": 16749332.0,
      "step": 9105
    },
    {
      "entropy": 5.2628867626190186,
      "epoch": 0.7609421984630805,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004948397349032762,
      "loss": 5.0373,
      "mean_token_accuracy": 0.20595274567604066,
      "num_tokens": 16757621.0,
      "step": 9110
    },
    {
      "entropy": 5.218843126296997,
      "epoch": 0.7613598396257936,
      "grad_norm": 1.140625,
      "learning_rate": 0.000494833393788818,
      "loss": 5.0836,
      "mean_token_accuracy": 0.19500521719455718,
      "num_tokens": 16766382.0,
      "step": 9115
    },
    {
      "entropy": 5.193441390991211,
      "epoch": 0.7617774807885065,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004948270488259117,
      "loss": 5.1198,
      "mean_token_accuracy": 0.19029657989740373,
      "num_tokens": 16774962.0,
      "step": 9120
    },
    {
      "entropy": 5.184526538848877,
      "epoch": 0.7621951219512195,
      "grad_norm": 1.125,
      "learning_rate": 0.0004948207000146682,
      "loss": 5.0885,
      "mean_token_accuracy": 0.19253478348255157,
      "num_tokens": 16784246.0,
      "step": 9125
    },
    {
      "entropy": 5.260955905914306,
      "epoch": 0.7626127631139326,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004948143473551989,
      "loss": 5.173,
      "mean_token_accuracy": 0.1836049288511276,
      "num_tokens": 16793396.0,
      "step": 9130
    },
    {
      "entropy": 5.30218358039856,
      "epoch": 0.7630304042766455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004948079908476147,
      "loss": 5.1486,
      "mean_token_accuracy": 0.192588971555233,
      "num_tokens": 16801837.0,
      "step": 9135
    },
    {
      "entropy": 5.331125211715698,
      "epoch": 0.7634480454393585,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004948016304920271,
      "loss": 5.183,
      "mean_token_accuracy": 0.19275147914886476,
      "num_tokens": 16810332.0,
      "step": 9140
    },
    {
      "entropy": 5.260197448730469,
      "epoch": 0.7638656866020715,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004947952662885473,
      "loss": 5.1359,
      "mean_token_accuracy": 0.19549433141946793,
      "num_tokens": 16820278.0,
      "step": 9145
    },
    {
      "entropy": 5.162685585021973,
      "epoch": 0.7642833277647845,
      "grad_norm": 1.234375,
      "learning_rate": 0.000494788898237287,
      "loss": 5.0417,
      "mean_token_accuracy": 0.19627458304166795,
      "num_tokens": 16828907.0,
      "step": 9150
    },
    {
      "entropy": 5.194856643676758,
      "epoch": 0.7647009689274975,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004947825263383573,
      "loss": 5.1112,
      "mean_token_accuracy": 0.18765236139297486,
      "num_tokens": 16838579.0,
      "step": 9155
    },
    {
      "entropy": 5.221455717086792,
      "epoch": 0.7651186100902105,
      "grad_norm": 1.125,
      "learning_rate": 0.00049477615059187,
      "loss": 5.1213,
      "mean_token_accuracy": 0.19863450527191162,
      "num_tokens": 16847143.0,
      "step": 9160
    },
    {
      "entropy": 5.3156860828399655,
      "epoch": 0.7655362512529235,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004947697709979367,
      "loss": 5.0926,
      "mean_token_accuracy": 0.19299326092004776,
      "num_tokens": 16855989.0,
      "step": 9165
    },
    {
      "entropy": 5.3172670841217045,
      "epoch": 0.7659538924156365,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004947633875566688,
      "loss": 5.1351,
      "mean_token_accuracy": 0.19252799600362777,
      "num_tokens": 16864701.0,
      "step": 9170
    },
    {
      "entropy": 5.337530899047851,
      "epoch": 0.7663715335783495,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004947570002681785,
      "loss": 5.2253,
      "mean_token_accuracy": 0.1833501622080803,
      "num_tokens": 16875032.0,
      "step": 9175
    },
    {
      "entropy": 5.299377059936523,
      "epoch": 0.7667891747410625,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004947506091325772,
      "loss": 5.2558,
      "mean_token_accuracy": 0.187994647026062,
      "num_tokens": 16884590.0,
      "step": 9180
    },
    {
      "entropy": 5.307699918746948,
      "epoch": 0.7672068159037755,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004947442141499771,
      "loss": 5.0864,
      "mean_token_accuracy": 0.18981925994157792,
      "num_tokens": 16894257.0,
      "step": 9185
    },
    {
      "entropy": 5.297932529449463,
      "epoch": 0.7676244570664885,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004947378153204899,
      "loss": 5.2037,
      "mean_token_accuracy": 0.19340216666460036,
      "num_tokens": 16904049.0,
      "step": 9190
    },
    {
      "entropy": 5.213099145889283,
      "epoch": 0.7680420982292014,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004947314126442279,
      "loss": 5.0921,
      "mean_token_accuracy": 0.192905293405056,
      "num_tokens": 16913147.0,
      "step": 9195
    },
    {
      "entropy": 5.2790021896362305,
      "epoch": 0.7684597393919145,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000494725006121303,
      "loss": 5.1342,
      "mean_token_accuracy": 0.18875229060649873,
      "num_tokens": 16922142.0,
      "step": 9200
    },
    {
      "entropy": 5.269748210906982,
      "epoch": 0.7688773805546275,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004947185957518273,
      "loss": 5.1538,
      "mean_token_accuracy": 0.19588406383991241,
      "num_tokens": 16931069.0,
      "step": 9205
    },
    {
      "entropy": 5.246332454681396,
      "epoch": 0.7692950217173404,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004947121815359133,
      "loss": 5.2461,
      "mean_token_accuracy": 0.19074745178222657,
      "num_tokens": 16940172.0,
      "step": 9210
    },
    {
      "entropy": 5.385737133026123,
      "epoch": 0.7697126628800535,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004947057634736729,
      "loss": 5.2617,
      "mean_token_accuracy": 0.18613625764846803,
      "num_tokens": 16951097.0,
      "step": 9215
    },
    {
      "entropy": 5.290096187591553,
      "epoch": 0.7701303040427665,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004946993415652186,
      "loss": 5.1571,
      "mean_token_accuracy": 0.19270050078630446,
      "num_tokens": 16960764.0,
      "step": 9220
    },
    {
      "entropy": 5.209099912643433,
      "epoch": 0.7705479452054794,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000494692915810663,
      "loss": 5.2061,
      "mean_token_accuracy": 0.1923610597848892,
      "num_tokens": 16969842.0,
      "step": 9225
    },
    {
      "entropy": 5.340341472625733,
      "epoch": 0.7709655863681925,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004946864862101185,
      "loss": 5.1012,
      "mean_token_accuracy": 0.19927529096603394,
      "num_tokens": 16979070.0,
      "step": 9230
    },
    {
      "entropy": 5.120919322967529,
      "epoch": 0.7713832275309055,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004946800527636975,
      "loss": 5.0648,
      "mean_token_accuracy": 0.19595357328653334,
      "num_tokens": 16988126.0,
      "step": 9235
    },
    {
      "entropy": 5.140116453170776,
      "epoch": 0.7718008686936184,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004946736154715129,
      "loss": 5.0797,
      "mean_token_accuracy": 0.19695620387792587,
      "num_tokens": 16997085.0,
      "step": 9240
    },
    {
      "entropy": 5.2858179092407225,
      "epoch": 0.7722185098563314,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004946671743336771,
      "loss": 5.1579,
      "mean_token_accuracy": 0.1884255200624466,
      "num_tokens": 17005678.0,
      "step": 9245
    },
    {
      "entropy": 5.364119005203247,
      "epoch": 0.7726361510190445,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004946607293503031,
      "loss": 5.0636,
      "mean_token_accuracy": 0.19261190444231033,
      "num_tokens": 17015081.0,
      "step": 9250
    },
    {
      "entropy": 5.245525598526001,
      "epoch": 0.7730537921817574,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004946542805215036,
      "loss": 5.1416,
      "mean_token_accuracy": 0.19777034372091293,
      "num_tokens": 17024036.0,
      "step": 9255
    },
    {
      "entropy": 5.278521966934204,
      "epoch": 0.7734714333444704,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004946478278473914,
      "loss": 5.1032,
      "mean_token_accuracy": 0.19373207092285155,
      "num_tokens": 17032120.0,
      "step": 9260
    },
    {
      "entropy": 5.275355339050293,
      "epoch": 0.7738890745071835,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004946413713280797,
      "loss": 5.108,
      "mean_token_accuracy": 0.1938307449221611,
      "num_tokens": 17041487.0,
      "step": 9265
    },
    {
      "entropy": 5.2276403427124025,
      "epoch": 0.7743067156698964,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004946349109636814,
      "loss": 5.1555,
      "mean_token_accuracy": 0.18773230016231537,
      "num_tokens": 17050793.0,
      "step": 9270
    },
    {
      "entropy": 5.275458908081054,
      "epoch": 0.7747243568326094,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004946284467543096,
      "loss": 5.1225,
      "mean_token_accuracy": 0.20014280527830125,
      "num_tokens": 17059536.0,
      "step": 9275
    },
    {
      "entropy": 5.2580150127410885,
      "epoch": 0.7751419979953225,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004946219787000776,
      "loss": 5.0973,
      "mean_token_accuracy": 0.19805282205343247,
      "num_tokens": 17069075.0,
      "step": 9280
    },
    {
      "entropy": 5.234522724151612,
      "epoch": 0.7755596391580354,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004946155068010984,
      "loss": 5.1531,
      "mean_token_accuracy": 0.19203932732343673,
      "num_tokens": 17079020.0,
      "step": 9285
    },
    {
      "entropy": 5.281824588775635,
      "epoch": 0.7759772803207484,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004946090310574854,
      "loss": 5.1468,
      "mean_token_accuracy": 0.1919183537364006,
      "num_tokens": 17087905.0,
      "step": 9290
    },
    {
      "entropy": 5.250629615783692,
      "epoch": 0.7763949214834615,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004946025514693519,
      "loss": 5.1538,
      "mean_token_accuracy": 0.19383215606212617,
      "num_tokens": 17097932.0,
      "step": 9295
    },
    {
      "entropy": 5.253558349609375,
      "epoch": 0.7768125626461744,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004945960680368115,
      "loss": 5.1104,
      "mean_token_accuracy": 0.19456967562437058,
      "num_tokens": 17107125.0,
      "step": 9300
    },
    {
      "entropy": 5.311372470855713,
      "epoch": 0.7772302038088874,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004945895807599775,
      "loss": 5.1628,
      "mean_token_accuracy": 0.18670535385608672,
      "num_tokens": 17116575.0,
      "step": 9305
    },
    {
      "entropy": 5.245980215072632,
      "epoch": 0.7776478449716004,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004945830896389636,
      "loss": 5.078,
      "mean_token_accuracy": 0.20212937295436859,
      "num_tokens": 17125422.0,
      "step": 9310
    },
    {
      "entropy": 5.1561438083648685,
      "epoch": 0.7780654861343134,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004945765946738835,
      "loss": 5.0469,
      "mean_token_accuracy": 0.19811525493860244,
      "num_tokens": 17134323.0,
      "step": 9315
    },
    {
      "entropy": 5.170308065414429,
      "epoch": 0.7784831272970264,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004945700958648507,
      "loss": 5.0361,
      "mean_token_accuracy": 0.20341728180646895,
      "num_tokens": 17143179.0,
      "step": 9320
    },
    {
      "entropy": 5.244274997711182,
      "epoch": 0.7789007684597394,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000494563593211979,
      "loss": 5.0887,
      "mean_token_accuracy": 0.19375640600919725,
      "num_tokens": 17152498.0,
      "step": 9325
    },
    {
      "entropy": 5.239305543899536,
      "epoch": 0.7793184096224524,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004945570867153826,
      "loss": 5.1398,
      "mean_token_accuracy": 0.18927657902240752,
      "num_tokens": 17161689.0,
      "step": 9330
    },
    {
      "entropy": 5.273555517196655,
      "epoch": 0.7797360507851654,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004945505763751749,
      "loss": 5.1441,
      "mean_token_accuracy": 0.1919786214828491,
      "num_tokens": 17170291.0,
      "step": 9335
    },
    {
      "entropy": 5.317656373977661,
      "epoch": 0.7801536919478784,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004945440621914702,
      "loss": 5.1671,
      "mean_token_accuracy": 0.19348419308662415,
      "num_tokens": 17179712.0,
      "step": 9340
    },
    {
      "entropy": 5.2609655380249025,
      "epoch": 0.7805713331105913,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004945375441643823,
      "loss": 5.174,
      "mean_token_accuracy": 0.192328542470932,
      "num_tokens": 17188224.0,
      "step": 9345
    },
    {
      "entropy": 5.299058151245117,
      "epoch": 0.7809889742733044,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004945310222940254,
      "loss": 5.1912,
      "mean_token_accuracy": 0.19506229758262633,
      "num_tokens": 17197868.0,
      "step": 9350
    },
    {
      "entropy": 5.241016960144043,
      "epoch": 0.7814066154360174,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004945244965805138,
      "loss": 5.1195,
      "mean_token_accuracy": 0.1913663700222969,
      "num_tokens": 17207576.0,
      "step": 9355
    },
    {
      "entropy": 5.218179702758789,
      "epoch": 0.7818242565987303,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004945179670239616,
      "loss": 5.1256,
      "mean_token_accuracy": 0.18980579525232316,
      "num_tokens": 17217508.0,
      "step": 9360
    },
    {
      "entropy": 5.370937633514404,
      "epoch": 0.7822418977614434,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004945114336244832,
      "loss": 5.2187,
      "mean_token_accuracy": 0.18452499210834503,
      "num_tokens": 17227890.0,
      "step": 9365
    },
    {
      "entropy": 5.333818960189819,
      "epoch": 0.7826595389241564,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000494504896382193,
      "loss": 5.1465,
      "mean_token_accuracy": 0.19608381241559983,
      "num_tokens": 17237038.0,
      "step": 9370
    },
    {
      "entropy": 5.267381048202514,
      "epoch": 0.7830771800868693,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004944983552972053,
      "loss": 5.2235,
      "mean_token_accuracy": 0.18515494018793105,
      "num_tokens": 17247653.0,
      "step": 9375
    },
    {
      "entropy": 5.293831014633179,
      "epoch": 0.7834948212495824,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004944918103696347,
      "loss": 5.2199,
      "mean_token_accuracy": 0.18410503566265107,
      "num_tokens": 17257183.0,
      "step": 9380
    },
    {
      "entropy": 5.312693881988525,
      "epoch": 0.7839124624122954,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004944852615995956,
      "loss": 5.1919,
      "mean_token_accuracy": 0.18909610509872438,
      "num_tokens": 17266315.0,
      "step": 9385
    },
    {
      "entropy": 5.318830728530884,
      "epoch": 0.7843301035750083,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000494478708987203,
      "loss": 5.1818,
      "mean_token_accuracy": 0.1851031720638275,
      "num_tokens": 17275829.0,
      "step": 9390
    },
    {
      "entropy": 5.4169337272644045,
      "epoch": 0.7847477447377214,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004944721525325713,
      "loss": 5.3293,
      "mean_token_accuracy": 0.17963435500860214,
      "num_tokens": 17285986.0,
      "step": 9395
    },
    {
      "entropy": 5.2347198009490965,
      "epoch": 0.7851653859004344,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004944655922358154,
      "loss": 4.9719,
      "mean_token_accuracy": 0.1993911534547806,
      "num_tokens": 17294314.0,
      "step": 9400
    },
    {
      "entropy": 5.12729434967041,
      "epoch": 0.7855830270631473,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004944590280970503,
      "loss": 5.0822,
      "mean_token_accuracy": 0.1968993991613388,
      "num_tokens": 17303437.0,
      "step": 9405
    },
    {
      "entropy": 5.2858038425445555,
      "epoch": 0.7860006682258603,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004944524601163905,
      "loss": 5.1801,
      "mean_token_accuracy": 0.18448719531297683,
      "num_tokens": 17313230.0,
      "step": 9410
    },
    {
      "entropy": 5.278919792175293,
      "epoch": 0.7864183093885734,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004944458882939513,
      "loss": 5.1374,
      "mean_token_accuracy": 0.18486365526914597,
      "num_tokens": 17323468.0,
      "step": 9415
    },
    {
      "entropy": 5.130036211013794,
      "epoch": 0.7868359505512863,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004944393126298478,
      "loss": 5.0167,
      "mean_token_accuracy": 0.20025767236948014,
      "num_tokens": 17332481.0,
      "step": 9420
    },
    {
      "entropy": 5.308260250091553,
      "epoch": 0.7872535917139993,
      "grad_norm": 1.234375,
      "learning_rate": 0.000494432733124195,
      "loss": 5.1428,
      "mean_token_accuracy": 0.18895395398139953,
      "num_tokens": 17341550.0,
      "step": 9425
    },
    {
      "entropy": 5.273244428634643,
      "epoch": 0.7876712328767124,
      "grad_norm": 1.140625,
      "learning_rate": 0.000494426149777108,
      "loss": 5.0677,
      "mean_token_accuracy": 0.1983897492289543,
      "num_tokens": 17350404.0,
      "step": 9430
    },
    {
      "entropy": 5.2244361400604244,
      "epoch": 0.7880888740394253,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000494419562588702,
      "loss": 5.0872,
      "mean_token_accuracy": 0.19000243693590163,
      "num_tokens": 17360198.0,
      "step": 9435
    },
    {
      "entropy": 5.231658792495727,
      "epoch": 0.7885065152021383,
      "grad_norm": 1.125,
      "learning_rate": 0.0004944129715590927,
      "loss": 5.1578,
      "mean_token_accuracy": 0.19901785999536514,
      "num_tokens": 17369681.0,
      "step": 9440
    },
    {
      "entropy": 5.186979579925537,
      "epoch": 0.7889241563648514,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000494406376688395,
      "loss": 5.0852,
      "mean_token_accuracy": 0.19972660690546035,
      "num_tokens": 17378695.0,
      "step": 9445
    },
    {
      "entropy": 5.314598655700683,
      "epoch": 0.7893417975275643,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004943997779767248,
      "loss": 5.2149,
      "mean_token_accuracy": 0.18282943814992905,
      "num_tokens": 17388161.0,
      "step": 9450
    },
    {
      "entropy": 5.319035291671753,
      "epoch": 0.7897594386902773,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004943931754241972,
      "loss": 5.2193,
      "mean_token_accuracy": 0.18679031282663344,
      "num_tokens": 17397082.0,
      "step": 9455
    },
    {
      "entropy": 5.203807306289673,
      "epoch": 0.7901770798529903,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000494386569030928,
      "loss": 5.0615,
      "mean_token_accuracy": 0.19658589959144593,
      "num_tokens": 17405842.0,
      "step": 9460
    },
    {
      "entropy": 5.252545881271362,
      "epoch": 0.7905947210157033,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004943799587970328,
      "loss": 5.1661,
      "mean_token_accuracy": 0.1925847500562668,
      "num_tokens": 17414639.0,
      "step": 9465
    },
    {
      "entropy": 5.266719675064087,
      "epoch": 0.7910123621784163,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004943733447226275,
      "loss": 5.1766,
      "mean_token_accuracy": 0.19542211592197417,
      "num_tokens": 17423923.0,
      "step": 9470
    },
    {
      "entropy": 5.138341331481934,
      "epoch": 0.7914300033411293,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004943667268078277,
      "loss": 5.0135,
      "mean_token_accuracy": 0.20514626950025558,
      "num_tokens": 17432309.0,
      "step": 9475
    },
    {
      "entropy": 5.2984966278076175,
      "epoch": 0.7918476445038423,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004943601050527493,
      "loss": 5.2009,
      "mean_token_accuracy": 0.1854607120156288,
      "num_tokens": 17441728.0,
      "step": 9480
    },
    {
      "entropy": 5.343739032745361,
      "epoch": 0.7922652856665553,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004943534794575081,
      "loss": 5.2138,
      "mean_token_accuracy": 0.18469537794589996,
      "num_tokens": 17451168.0,
      "step": 9485
    },
    {
      "entropy": 5.330528974533081,
      "epoch": 0.7926829268292683,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004943468500222203,
      "loss": 5.2315,
      "mean_token_accuracy": 0.19061953872442244,
      "num_tokens": 17461179.0,
      "step": 9490
    },
    {
      "entropy": 5.266789150238037,
      "epoch": 0.7931005679919813,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004943402167470019,
      "loss": 5.0782,
      "mean_token_accuracy": 0.1966783419251442,
      "num_tokens": 17469710.0,
      "step": 9495
    },
    {
      "entropy": 5.241172504425049,
      "epoch": 0.7935182091546943,
      "grad_norm": 1.203125,
      "learning_rate": 0.000494333579631969,
      "loss": 5.1666,
      "mean_token_accuracy": 0.19097824692726134,
      "num_tokens": 17479482.0,
      "step": 9500
    },
    {
      "entropy": 5.0443164825439455,
      "epoch": 0.7939358503174073,
      "grad_norm": 1.125,
      "learning_rate": 0.0004943269386772376,
      "loss": 4.9558,
      "mean_token_accuracy": 0.20915199518203736,
      "num_tokens": 17488458.0,
      "step": 9505
    },
    {
      "entropy": 5.306435489654541,
      "epoch": 0.7943534914801202,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004943202938829243,
      "loss": 5.1613,
      "mean_token_accuracy": 0.1944829910993576,
      "num_tokens": 17497817.0,
      "step": 9510
    },
    {
      "entropy": 5.214391088485717,
      "epoch": 0.7947711326428333,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004943136452491452,
      "loss": 5.0289,
      "mean_token_accuracy": 0.19421486407518387,
      "num_tokens": 17506020.0,
      "step": 9515
    },
    {
      "entropy": 5.237166023254394,
      "epoch": 0.7951887738055463,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004943069927760168,
      "loss": 5.1431,
      "mean_token_accuracy": 0.1870472624897957,
      "num_tokens": 17515113.0,
      "step": 9520
    },
    {
      "entropy": 5.283129596710205,
      "epoch": 0.7956064149682592,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004943003364636554,
      "loss": 5.1375,
      "mean_token_accuracy": 0.198609422147274,
      "num_tokens": 17523473.0,
      "step": 9525
    },
    {
      "entropy": 5.283037567138672,
      "epoch": 0.7960240561309723,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004942936763121777,
      "loss": 5.1631,
      "mean_token_accuracy": 0.18911215513944626,
      "num_tokens": 17532985.0,
      "step": 9530
    },
    {
      "entropy": 5.1913494110107425,
      "epoch": 0.7964416972936853,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004942870123217002,
      "loss": 5.1167,
      "mean_token_accuracy": 0.18947843164205552,
      "num_tokens": 17541859.0,
      "step": 9535
    },
    {
      "entropy": 5.215520143508911,
      "epoch": 0.7968593384563982,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004942803444923398,
      "loss": 5.0931,
      "mean_token_accuracy": 0.1978566288948059,
      "num_tokens": 17550263.0,
      "step": 9540
    },
    {
      "entropy": 5.3430342197418215,
      "epoch": 0.7972769796191113,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004942736728242127,
      "loss": 5.2382,
      "mean_token_accuracy": 0.19369956552982331,
      "num_tokens": 17560028.0,
      "step": 9545
    },
    {
      "entropy": 5.302721118927002,
      "epoch": 0.7976946207818243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004942669973174363,
      "loss": 5.1114,
      "mean_token_accuracy": 0.1931803658604622,
      "num_tokens": 17569082.0,
      "step": 9550
    },
    {
      "entropy": 5.2174601554870605,
      "epoch": 0.7981122619445372,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000494260317972127,
      "loss": 5.1208,
      "mean_token_accuracy": 0.19430044442415237,
      "num_tokens": 17577589.0,
      "step": 9555
    },
    {
      "entropy": 5.087086725234985,
      "epoch": 0.7985299031072502,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004942536347884019,
      "loss": 4.952,
      "mean_token_accuracy": 0.2063223659992218,
      "num_tokens": 17586066.0,
      "step": 9560
    },
    {
      "entropy": 5.1341081142425535,
      "epoch": 0.7989475442699633,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004942469477663782,
      "loss": 4.9554,
      "mean_token_accuracy": 0.20437541007995605,
      "num_tokens": 17594328.0,
      "step": 9565
    },
    {
      "entropy": 5.222323274612426,
      "epoch": 0.7993651854326762,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004942402569061726,
      "loss": 5.1827,
      "mean_token_accuracy": 0.19120361506938935,
      "num_tokens": 17603719.0,
      "step": 9570
    },
    {
      "entropy": 5.381272840499878,
      "epoch": 0.7997828265953892,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004942335622079025,
      "loss": 5.2307,
      "mean_token_accuracy": 0.19148974865674973,
      "num_tokens": 17613306.0,
      "step": 9575
    },
    {
      "entropy": 5.215620613098144,
      "epoch": 0.8002004677581023,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004942268636716848,
      "loss": 5.0946,
      "mean_token_accuracy": 0.19395049959421157,
      "num_tokens": 17622756.0,
      "step": 9580
    },
    {
      "entropy": 5.196308517456055,
      "epoch": 0.8006181089208152,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004942201612976372,
      "loss": 5.1196,
      "mean_token_accuracy": 0.20300968289375304,
      "num_tokens": 17632157.0,
      "step": 9585
    },
    {
      "entropy": 5.196787595748901,
      "epoch": 0.8010357500835282,
      "grad_norm": 1.125,
      "learning_rate": 0.0004942134550858767,
      "loss": 5.0451,
      "mean_token_accuracy": 0.2019111067056656,
      "num_tokens": 17641706.0,
      "step": 9590
    },
    {
      "entropy": 5.215884780883789,
      "epoch": 0.8014533912462413,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004942067450365208,
      "loss": 5.157,
      "mean_token_accuracy": 0.18573255836963654,
      "num_tokens": 17651572.0,
      "step": 9595
    },
    {
      "entropy": 5.258349895477295,
      "epoch": 0.8018710324089542,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004942000311496871,
      "loss": 5.1543,
      "mean_token_accuracy": 0.19262522161006929,
      "num_tokens": 17661354.0,
      "step": 9600
    },
    {
      "entropy": 5.297380208969116,
      "epoch": 0.8022886735716672,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004941933134254928,
      "loss": 5.1173,
      "mean_token_accuracy": 0.1944006398320198,
      "num_tokens": 17670482.0,
      "step": 9605
    },
    {
      "entropy": 5.255826997756958,
      "epoch": 0.8027063147343803,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004941865918640559,
      "loss": 5.1856,
      "mean_token_accuracy": 0.18901994377374648,
      "num_tokens": 17679429.0,
      "step": 9610
    },
    {
      "entropy": 5.297385931015015,
      "epoch": 0.8031239558970932,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004941798664654937,
      "loss": 5.231,
      "mean_token_accuracy": 0.18525495380163193,
      "num_tokens": 17688620.0,
      "step": 9615
    },
    {
      "entropy": 5.261661958694458,
      "epoch": 0.8035415970598062,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004941731372299241,
      "loss": 5.1269,
      "mean_token_accuracy": 0.18994914144277572,
      "num_tokens": 17698493.0,
      "step": 9620
    },
    {
      "entropy": 5.191858768463135,
      "epoch": 0.8039592382225192,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004941664041574651,
      "loss": 5.0444,
      "mean_token_accuracy": 0.19941396862268448,
      "num_tokens": 17707408.0,
      "step": 9625
    },
    {
      "entropy": 5.220510578155517,
      "epoch": 0.8043768793852322,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004941596672482343,
      "loss": 5.1254,
      "mean_token_accuracy": 0.19712237119674683,
      "num_tokens": 17716448.0,
      "step": 9630
    },
    {
      "entropy": 5.273178911209106,
      "epoch": 0.8047945205479452,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004941529265023498,
      "loss": 5.1904,
      "mean_token_accuracy": 0.19830915331840515,
      "num_tokens": 17725287.0,
      "step": 9635
    },
    {
      "entropy": 5.3732161045074465,
      "epoch": 0.8052121617106582,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004941461819199293,
      "loss": 5.1986,
      "mean_token_accuracy": 0.19137406647205352,
      "num_tokens": 17734674.0,
      "step": 9640
    },
    {
      "entropy": 5.3338597297668455,
      "epoch": 0.8056298028733712,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004941394335010912,
      "loss": 5.2277,
      "mean_token_accuracy": 0.1871379315853119,
      "num_tokens": 17743589.0,
      "step": 9645
    },
    {
      "entropy": 5.242267227172851,
      "epoch": 0.8060474440360842,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004941326812459535,
      "loss": 5.1747,
      "mean_token_accuracy": 0.19810663312673568,
      "num_tokens": 17753210.0,
      "step": 9650
    },
    {
      "entropy": 5.324686050415039,
      "epoch": 0.8064650851987972,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004941259251546346,
      "loss": 5.0993,
      "mean_token_accuracy": 0.19625033736228942,
      "num_tokens": 17762074.0,
      "step": 9655
    },
    {
      "entropy": 5.360092878341675,
      "epoch": 0.8068827263615101,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004941191652272526,
      "loss": 5.2374,
      "mean_token_accuracy": 0.18395915031433105,
      "num_tokens": 17771263.0,
      "step": 9660
    },
    {
      "entropy": 5.268192386627197,
      "epoch": 0.8073003675242232,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004941124014639258,
      "loss": 5.0822,
      "mean_token_accuracy": 0.19523436427116395,
      "num_tokens": 17781351.0,
      "step": 9665
    },
    {
      "entropy": 5.219300127029419,
      "epoch": 0.8077180086869362,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004941056338647726,
      "loss": 5.0486,
      "mean_token_accuracy": 0.2020439699292183,
      "num_tokens": 17790343.0,
      "step": 9670
    },
    {
      "entropy": 5.21505708694458,
      "epoch": 0.8081356498496491,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004940988624299116,
      "loss": 5.1281,
      "mean_token_accuracy": 0.1876239076256752,
      "num_tokens": 17799338.0,
      "step": 9675
    },
    {
      "entropy": 5.159978246688842,
      "epoch": 0.8085532910123622,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004940920871594613,
      "loss": 5.0965,
      "mean_token_accuracy": 0.19193944334983826,
      "num_tokens": 17808852.0,
      "step": 9680
    },
    {
      "entropy": 5.303667068481445,
      "epoch": 0.8089709321750752,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004940853080535403,
      "loss": 5.1213,
      "mean_token_accuracy": 0.1877689927816391,
      "num_tokens": 17817654.0,
      "step": 9685
    },
    {
      "entropy": 5.266427040100098,
      "epoch": 0.8093885733377881,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004940785251122673,
      "loss": 5.0823,
      "mean_token_accuracy": 0.1913834407925606,
      "num_tokens": 17826640.0,
      "step": 9690
    },
    {
      "entropy": 5.2331428050994875,
      "epoch": 0.8098062145005012,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004940717383357608,
      "loss": 5.1024,
      "mean_token_accuracy": 0.19399402737617494,
      "num_tokens": 17835493.0,
      "step": 9695
    },
    {
      "entropy": 5.15072922706604,
      "epoch": 0.8102238556632142,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004940649477241401,
      "loss": 5.1065,
      "mean_token_accuracy": 0.19196264296770096,
      "num_tokens": 17845375.0,
      "step": 9700
    },
    {
      "entropy": 5.286759853363037,
      "epoch": 0.8106414968259271,
      "grad_norm": 1.125,
      "learning_rate": 0.0004940581532775237,
      "loss": 5.0762,
      "mean_token_accuracy": 0.20250680446624755,
      "num_tokens": 17854473.0,
      "step": 9705
    },
    {
      "entropy": 5.361585426330566,
      "epoch": 0.8110591379886402,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004940513549960306,
      "loss": 5.1604,
      "mean_token_accuracy": 0.18932442665100097,
      "num_tokens": 17864242.0,
      "step": 9710
    },
    {
      "entropy": 5.264509153366089,
      "epoch": 0.8114767791513532,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004940445528797798,
      "loss": 5.1532,
      "mean_token_accuracy": 0.1892949566245079,
      "num_tokens": 17873292.0,
      "step": 9715
    },
    {
      "entropy": 5.189016437530517,
      "epoch": 0.8118944203140661,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004940377469288906,
      "loss": 5.0667,
      "mean_token_accuracy": 0.1957487627863884,
      "num_tokens": 17883601.0,
      "step": 9720
    },
    {
      "entropy": 5.241117000579834,
      "epoch": 0.8123120614767791,
      "grad_norm": 1.25,
      "learning_rate": 0.0004940309371434819,
      "loss": 5.1315,
      "mean_token_accuracy": 0.1988212063908577,
      "num_tokens": 17892241.0,
      "step": 9725
    },
    {
      "entropy": 5.30053129196167,
      "epoch": 0.8127297026394922,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004940241235236731,
      "loss": 5.1739,
      "mean_token_accuracy": 0.1967609107494354,
      "num_tokens": 17901236.0,
      "step": 9730
    },
    {
      "entropy": 5.297317600250244,
      "epoch": 0.8131473438022051,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004940173060695833,
      "loss": 5.1636,
      "mean_token_accuracy": 0.18704858422279358,
      "num_tokens": 17909871.0,
      "step": 9735
    },
    {
      "entropy": 5.109527587890625,
      "epoch": 0.8135649849649181,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004940104847813319,
      "loss": 4.9666,
      "mean_token_accuracy": 0.2048913300037384,
      "num_tokens": 17918158.0,
      "step": 9740
    },
    {
      "entropy": 5.180254507064819,
      "epoch": 0.8139826261276312,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004940036596590384,
      "loss": 5.0734,
      "mean_token_accuracy": 0.1968464434146881,
      "num_tokens": 17927778.0,
      "step": 9745
    },
    {
      "entropy": 5.2881190299987795,
      "epoch": 0.8144002672903441,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004939968307028221,
      "loss": 5.0988,
      "mean_token_accuracy": 0.19233490526676178,
      "num_tokens": 17936859.0,
      "step": 9750
    },
    {
      "entropy": 5.2676292896270756,
      "epoch": 0.8148179084530571,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004939899979128028,
      "loss": 5.1476,
      "mean_token_accuracy": 0.18997602611780168,
      "num_tokens": 17946477.0,
      "step": 9755
    },
    {
      "entropy": 5.239906692504883,
      "epoch": 0.8152355496157702,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004939831612890999,
      "loss": 5.0964,
      "mean_token_accuracy": 0.1961331143975258,
      "num_tokens": 17956362.0,
      "step": 9760
    },
    {
      "entropy": 5.228900671005249,
      "epoch": 0.8156531907784831,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004939763208318332,
      "loss": 5.1252,
      "mean_token_accuracy": 0.1941460058093071,
      "num_tokens": 17965837.0,
      "step": 9765
    },
    {
      "entropy": 5.312269496917724,
      "epoch": 0.8160708319411961,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004939694765411225,
      "loss": 5.1942,
      "mean_token_accuracy": 0.18751802444458007,
      "num_tokens": 17974750.0,
      "step": 9770
    },
    {
      "entropy": 5.2949858665466305,
      "epoch": 0.8164884731039092,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004939626284170874,
      "loss": 5.2531,
      "mean_token_accuracy": 0.18865209221839904,
      "num_tokens": 17985106.0,
      "step": 9775
    },
    {
      "entropy": 5.226165246963501,
      "epoch": 0.8169061142666221,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000493955776459848,
      "loss": 5.1893,
      "mean_token_accuracy": 0.19260022640228272,
      "num_tokens": 17995073.0,
      "step": 9780
    },
    {
      "entropy": 5.304164123535156,
      "epoch": 0.8173237554293351,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004939489206695241,
      "loss": 5.113,
      "mean_token_accuracy": 0.197147735953331,
      "num_tokens": 18004112.0,
      "step": 9785
    },
    {
      "entropy": 5.28415002822876,
      "epoch": 0.8177413965920481,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004939420610462359,
      "loss": 5.1212,
      "mean_token_accuracy": 0.1896842449903488,
      "num_tokens": 18013344.0,
      "step": 9790
    },
    {
      "entropy": 5.259722900390625,
      "epoch": 0.8181590377547611,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004939351975901034,
      "loss": 5.0822,
      "mean_token_accuracy": 0.19383978247642517,
      "num_tokens": 18022938.0,
      "step": 9795
    },
    {
      "entropy": 5.213902473449707,
      "epoch": 0.8185766789174741,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004939283303012467,
      "loss": 5.0598,
      "mean_token_accuracy": 0.19638063609600068,
      "num_tokens": 18031507.0,
      "step": 9800
    },
    {
      "entropy": 5.205690526962281,
      "epoch": 0.8189943200801871,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000493921459179786,
      "loss": 5.1122,
      "mean_token_accuracy": 0.1913120687007904,
      "num_tokens": 18039553.0,
      "step": 9805
    },
    {
      "entropy": 5.221726894378662,
      "epoch": 0.8194119612429,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004939145842258418,
      "loss": 5.0585,
      "mean_token_accuracy": 0.1963190972805023,
      "num_tokens": 18048550.0,
      "step": 9810
    },
    {
      "entropy": 5.296284818649292,
      "epoch": 0.8198296024056131,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004939077054395341,
      "loss": 5.0962,
      "mean_token_accuracy": 0.1979181319475174,
      "num_tokens": 18057298.0,
      "step": 9815
    },
    {
      "entropy": 5.162118244171142,
      "epoch": 0.8202472435683261,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004939008228209836,
      "loss": 5.0346,
      "mean_token_accuracy": 0.1951318070292473,
      "num_tokens": 18066740.0,
      "step": 9820
    },
    {
      "entropy": 5.175771713256836,
      "epoch": 0.820664884731039,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004938939363703106,
      "loss": 5.2077,
      "mean_token_accuracy": 0.19318842142820358,
      "num_tokens": 18076663.0,
      "step": 9825
    },
    {
      "entropy": 5.318018817901612,
      "epoch": 0.8210825258937521,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004938870460876359,
      "loss": 5.1897,
      "mean_token_accuracy": 0.18922351002693177,
      "num_tokens": 18085037.0,
      "step": 9830
    },
    {
      "entropy": 5.384563207626343,
      "epoch": 0.8215001670564651,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00049388015197308,
      "loss": 5.2135,
      "mean_token_accuracy": 0.18890699744224548,
      "num_tokens": 18094636.0,
      "step": 9835
    },
    {
      "entropy": 5.223188400268555,
      "epoch": 0.821917808219178,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004938732540267635,
      "loss": 5.1283,
      "mean_token_accuracy": 0.18941520452499389,
      "num_tokens": 18103083.0,
      "step": 9840
    },
    {
      "entropy": 5.216627311706543,
      "epoch": 0.8223354493818911,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004938663522488072,
      "loss": 5.093,
      "mean_token_accuracy": 0.1926014930009842,
      "num_tokens": 18112108.0,
      "step": 9845
    },
    {
      "entropy": 5.386596632003784,
      "epoch": 0.8227530905446041,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004938594466393321,
      "loss": 5.178,
      "mean_token_accuracy": 0.19452123641967772,
      "num_tokens": 18121001.0,
      "step": 9850
    },
    {
      "entropy": 5.224433565139771,
      "epoch": 0.823170731707317,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004938525371984588,
      "loss": 5.0303,
      "mean_token_accuracy": 0.20452803373336792,
      "num_tokens": 18129309.0,
      "step": 9855
    },
    {
      "entropy": 5.246949291229248,
      "epoch": 0.8235883728700301,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004938456239263085,
      "loss": 5.2203,
      "mean_token_accuracy": 0.18910325765609742,
      "num_tokens": 18138373.0,
      "step": 9860
    },
    {
      "entropy": 5.308945941925049,
      "epoch": 0.8240060140327431,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004938387068230021,
      "loss": 5.1647,
      "mean_token_accuracy": 0.18817575573921203,
      "num_tokens": 18148038.0,
      "step": 9865
    },
    {
      "entropy": 5.27157506942749,
      "epoch": 0.824423655195456,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004938317858886606,
      "loss": 5.0829,
      "mean_token_accuracy": 0.19335709512233734,
      "num_tokens": 18158132.0,
      "step": 9870
    },
    {
      "entropy": 5.231343030929565,
      "epoch": 0.824841296358169,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004938248611234055,
      "loss": 5.0906,
      "mean_token_accuracy": 0.19111096411943435,
      "num_tokens": 18167276.0,
      "step": 9875
    },
    {
      "entropy": 5.206383562088012,
      "epoch": 0.8252589375208821,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004938179325273577,
      "loss": 5.1495,
      "mean_token_accuracy": 0.19772316813468932,
      "num_tokens": 18177538.0,
      "step": 9880
    },
    {
      "entropy": 5.3209789276123045,
      "epoch": 0.825676578683595,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004938110001006385,
      "loss": 5.1507,
      "mean_token_accuracy": 0.19052454978227615,
      "num_tokens": 18187067.0,
      "step": 9885
    },
    {
      "entropy": 5.2642628192901615,
      "epoch": 0.826094219846308,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004938040638433694,
      "loss": 5.1233,
      "mean_token_accuracy": 0.18694647550582885,
      "num_tokens": 18196710.0,
      "step": 9890
    },
    {
      "entropy": 5.212479782104492,
      "epoch": 0.8265118610090211,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004937971237556719,
      "loss": 5.1221,
      "mean_token_accuracy": 0.19053595811128615,
      "num_tokens": 18205789.0,
      "step": 9895
    },
    {
      "entropy": 5.1881612777709964,
      "epoch": 0.826929502171734,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004937901798376672,
      "loss": 4.9781,
      "mean_token_accuracy": 0.19853330105543138,
      "num_tokens": 18214633.0,
      "step": 9900
    },
    {
      "entropy": 5.198566198348999,
      "epoch": 0.827347143334447,
      "grad_norm": 1.25,
      "learning_rate": 0.0004937832320894772,
      "loss": 5.1407,
      "mean_token_accuracy": 0.19435544610023497,
      "num_tokens": 18225019.0,
      "step": 9905
    },
    {
      "entropy": 5.203691387176514,
      "epoch": 0.8277647844971601,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004937762805112233,
      "loss": 5.0842,
      "mean_token_accuracy": 0.20435549169778824,
      "num_tokens": 18234610.0,
      "step": 9910
    },
    {
      "entropy": 5.222125911712647,
      "epoch": 0.828182425659873,
      "grad_norm": 1.25,
      "learning_rate": 0.0004937693251030273,
      "loss": 5.1017,
      "mean_token_accuracy": 0.19584225118160248,
      "num_tokens": 18243733.0,
      "step": 9915
    },
    {
      "entropy": 5.170470666885376,
      "epoch": 0.828600066822586,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000493762365865011,
      "loss": 5.1064,
      "mean_token_accuracy": 0.19895717054605483,
      "num_tokens": 18253105.0,
      "step": 9920
    },
    {
      "entropy": 5.308280277252197,
      "epoch": 0.8290177079852991,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004937554027972961,
      "loss": 5.1961,
      "mean_token_accuracy": 0.18659522533416747,
      "num_tokens": 18261862.0,
      "step": 9925
    },
    {
      "entropy": 5.334354639053345,
      "epoch": 0.829435349148012,
      "grad_norm": 1.25,
      "learning_rate": 0.0004937484359000046,
      "loss": 5.2195,
      "mean_token_accuracy": 0.19028422683477403,
      "num_tokens": 18271941.0,
      "step": 9930
    },
    {
      "entropy": 5.2598161697387695,
      "epoch": 0.829852990310725,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004937414651732584,
      "loss": 5.0414,
      "mean_token_accuracy": 0.19467880874872207,
      "num_tokens": 18281048.0,
      "step": 9935
    },
    {
      "entropy": 5.323976039886475,
      "epoch": 0.830270631473438,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004937344906171795,
      "loss": 5.2334,
      "mean_token_accuracy": 0.18143781125545502,
      "num_tokens": 18290140.0,
      "step": 9940
    },
    {
      "entropy": 5.168869733810425,
      "epoch": 0.830688272636151,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004937275122318902,
      "loss": 5.0459,
      "mean_token_accuracy": 0.1929491087794304,
      "num_tokens": 18298784.0,
      "step": 9945
    },
    {
      "entropy": 5.275314807891846,
      "epoch": 0.831105913798864,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004937205300175123,
      "loss": 5.1328,
      "mean_token_accuracy": 0.19509119540452957,
      "num_tokens": 18307750.0,
      "step": 9950
    },
    {
      "entropy": 5.2995954036712645,
      "epoch": 0.831523554961577,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004937135439741682,
      "loss": 5.1003,
      "mean_token_accuracy": 0.19828321933746337,
      "num_tokens": 18317042.0,
      "step": 9955
    },
    {
      "entropy": 5.244739675521851,
      "epoch": 0.83194119612429,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004937065541019806,
      "loss": 5.0778,
      "mean_token_accuracy": 0.1906749948859215,
      "num_tokens": 18326733.0,
      "step": 9960
    },
    {
      "entropy": 5.3648172378540036,
      "epoch": 0.832358837287003,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004936995604010711,
      "loss": 5.2887,
      "mean_token_accuracy": 0.1812009334564209,
      "num_tokens": 18336366.0,
      "step": 9965
    },
    {
      "entropy": 5.255705642700195,
      "epoch": 0.832776478449716,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004936925628715628,
      "loss": 5.1471,
      "mean_token_accuracy": 0.1914728492498398,
      "num_tokens": 18345260.0,
      "step": 9970
    },
    {
      "entropy": 5.322878313064575,
      "epoch": 0.833194119612429,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004936855615135779,
      "loss": 5.1805,
      "mean_token_accuracy": 0.18769506365060806,
      "num_tokens": 18354916.0,
      "step": 9975
    },
    {
      "entropy": 5.315698289871216,
      "epoch": 0.833611760775142,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000493678556327239,
      "loss": 5.1483,
      "mean_token_accuracy": 0.19391870498657227,
      "num_tokens": 18364012.0,
      "step": 9980
    },
    {
      "entropy": 5.300132942199707,
      "epoch": 0.834029401937855,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004936715473126688,
      "loss": 5.1344,
      "mean_token_accuracy": 0.1849350556731224,
      "num_tokens": 18374685.0,
      "step": 9985
    },
    {
      "entropy": 5.096344947814941,
      "epoch": 0.8344470431005679,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00049366453446999,
      "loss": 4.9808,
      "mean_token_accuracy": 0.2049587234854698,
      "num_tokens": 18383716.0,
      "step": 9990
    },
    {
      "entropy": 5.142494821548462,
      "epoch": 0.834864684263281,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004936575177993251,
      "loss": 5.085,
      "mean_token_accuracy": 0.19872361719608306,
      "num_tokens": 18393071.0,
      "step": 9995
    },
    {
      "entropy": 5.424183464050293,
      "epoch": 0.835282325425994,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004936504973007973,
      "loss": 5.2259,
      "mean_token_accuracy": 0.18933799862861633,
      "num_tokens": 18403593.0,
      "step": 10000
    },
    {
      "entropy": 5.3685239315032955,
      "epoch": 0.8356999665887069,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004936434729745294,
      "loss": 5.2419,
      "mean_token_accuracy": 0.18411905616521834,
      "num_tokens": 18412016.0,
      "step": 10005
    },
    {
      "entropy": 5.2355184078216555,
      "epoch": 0.83611760775142,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004936364448206442,
      "loss": 5.0537,
      "mean_token_accuracy": 0.19321360886096955,
      "num_tokens": 18421098.0,
      "step": 10010
    },
    {
      "entropy": 5.227397155761719,
      "epoch": 0.836535248914133,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004936294128392649,
      "loss": 5.1577,
      "mean_token_accuracy": 0.18819456398487092,
      "num_tokens": 18430264.0,
      "step": 10015
    },
    {
      "entropy": 5.332653760910034,
      "epoch": 0.8369528900768459,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004936223770305146,
      "loss": 5.1961,
      "mean_token_accuracy": 0.1921686962246895,
      "num_tokens": 18439595.0,
      "step": 10020
    },
    {
      "entropy": 5.297098541259766,
      "epoch": 0.837370531239559,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004936153373945165,
      "loss": 5.1008,
      "mean_token_accuracy": 0.1947207123041153,
      "num_tokens": 18448945.0,
      "step": 10025
    },
    {
      "entropy": 5.288499164581299,
      "epoch": 0.837788172402272,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004936082939313936,
      "loss": 5.0773,
      "mean_token_accuracy": 0.19456617832183837,
      "num_tokens": 18458623.0,
      "step": 10030
    },
    {
      "entropy": 5.249464321136474,
      "epoch": 0.8382058135649849,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004936012466412695,
      "loss": 5.1232,
      "mean_token_accuracy": 0.19926332384347917,
      "num_tokens": 18466462.0,
      "step": 10035
    },
    {
      "entropy": 5.2157652378082275,
      "epoch": 0.838623454727698,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004935941955242674,
      "loss": 5.1528,
      "mean_token_accuracy": 0.18945154994726182,
      "num_tokens": 18475936.0,
      "step": 10040
    },
    {
      "entropy": 5.190078020095825,
      "epoch": 0.839041095890411,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004935871405805109,
      "loss": 5.064,
      "mean_token_accuracy": 0.19731518477201462,
      "num_tokens": 18485526.0,
      "step": 10045
    },
    {
      "entropy": 5.2232341289520265,
      "epoch": 0.8394587370531239,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004935800818101232,
      "loss": 5.0717,
      "mean_token_accuracy": 0.19846357703208922,
      "num_tokens": 18494202.0,
      "step": 10050
    },
    {
      "entropy": 5.347828674316406,
      "epoch": 0.8398763782158369,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004935730192132282,
      "loss": 5.2263,
      "mean_token_accuracy": 0.18958183526992797,
      "num_tokens": 18503502.0,
      "step": 10055
    },
    {
      "entropy": 5.350813388824463,
      "epoch": 0.84029401937855,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004935659527899493,
      "loss": 5.1681,
      "mean_token_accuracy": 0.1950361981987953,
      "num_tokens": 18512607.0,
      "step": 10060
    },
    {
      "entropy": 5.269353723526001,
      "epoch": 0.8407116605412629,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004935588825404104,
      "loss": 5.0813,
      "mean_token_accuracy": 0.19499771147966385,
      "num_tokens": 18521875.0,
      "step": 10065
    },
    {
      "entropy": 5.298912239074707,
      "epoch": 0.8411293017039759,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004935518084647351,
      "loss": 5.172,
      "mean_token_accuracy": 0.2010849565267563,
      "num_tokens": 18530554.0,
      "step": 10070
    },
    {
      "entropy": 5.277385997772217,
      "epoch": 0.841546942866689,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004935447305630474,
      "loss": 5.0567,
      "mean_token_accuracy": 0.20428809076547622,
      "num_tokens": 18539037.0,
      "step": 10075
    },
    {
      "entropy": 5.143201160430908,
      "epoch": 0.8419645840294019,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004935376488354711,
      "loss": 5.0359,
      "mean_token_accuracy": 0.1989608809351921,
      "num_tokens": 18548218.0,
      "step": 10080
    },
    {
      "entropy": 5.296752023696899,
      "epoch": 0.8423822251921149,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004935305632821303,
      "loss": 5.0992,
      "mean_token_accuracy": 0.1943753942847252,
      "num_tokens": 18557968.0,
      "step": 10085
    },
    {
      "entropy": 5.300920295715332,
      "epoch": 0.842799866354828,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004935234739031488,
      "loss": 5.216,
      "mean_token_accuracy": 0.18711352348327637,
      "num_tokens": 18566809.0,
      "step": 10090
    },
    {
      "entropy": 5.256332349777222,
      "epoch": 0.8432175075175409,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004935163806986508,
      "loss": 5.0809,
      "mean_token_accuracy": 0.19128889888525008,
      "num_tokens": 18574419.0,
      "step": 10095
    },
    {
      "entropy": 5.299392127990723,
      "epoch": 0.8436351486802539,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004935092836687607,
      "loss": 5.1346,
      "mean_token_accuracy": 0.19238422960042953,
      "num_tokens": 18584470.0,
      "step": 10100
    },
    {
      "entropy": 5.344189453125,
      "epoch": 0.8440527898429669,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004935021828136025,
      "loss": 5.1308,
      "mean_token_accuracy": 0.19648874551057816,
      "num_tokens": 18593551.0,
      "step": 10105
    },
    {
      "entropy": 5.211934900283813,
      "epoch": 0.8444704310056799,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004934950781333005,
      "loss": 5.1172,
      "mean_token_accuracy": 0.19153958112001418,
      "num_tokens": 18602253.0,
      "step": 10110
    },
    {
      "entropy": 5.244383239746094,
      "epoch": 0.8448880721683929,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004934879696279793,
      "loss": 5.1548,
      "mean_token_accuracy": 0.18769302368164062,
      "num_tokens": 18611261.0,
      "step": 10115
    },
    {
      "entropy": 5.376082801818848,
      "epoch": 0.8453057133311059,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004934808572977632,
      "loss": 5.2433,
      "mean_token_accuracy": 0.1848089024424553,
      "num_tokens": 18620508.0,
      "step": 10120
    },
    {
      "entropy": 5.307138156890869,
      "epoch": 0.8457233544938189,
      "grad_norm": 1.25,
      "learning_rate": 0.0004934737411427765,
      "loss": 5.1541,
      "mean_token_accuracy": 0.20037644803524018,
      "num_tokens": 18629140.0,
      "step": 10125
    },
    {
      "entropy": 5.1886794090271,
      "epoch": 0.8461409956565319,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004934666211631442,
      "loss": 5.1137,
      "mean_token_accuracy": 0.20019556134939193,
      "num_tokens": 18637924.0,
      "step": 10130
    },
    {
      "entropy": 5.25268669128418,
      "epoch": 0.8465586368192449,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004934594973589906,
      "loss": 5.0658,
      "mean_token_accuracy": 0.19955097436904906,
      "num_tokens": 18646193.0,
      "step": 10135
    },
    {
      "entropy": 5.368896007537842,
      "epoch": 0.8469762779819578,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004934523697304406,
      "loss": 5.1888,
      "mean_token_accuracy": 0.18375039398670195,
      "num_tokens": 18655327.0,
      "step": 10140
    },
    {
      "entropy": 5.293934345245361,
      "epoch": 0.8473939191446709,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004934452382776189,
      "loss": 5.1477,
      "mean_token_accuracy": 0.18675615638494492,
      "num_tokens": 18664633.0,
      "step": 10145
    },
    {
      "entropy": 5.240318775177002,
      "epoch": 0.8478115603073839,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004934381030006504,
      "loss": 5.1621,
      "mean_token_accuracy": 0.18676041811704636,
      "num_tokens": 18673956.0,
      "step": 10150
    },
    {
      "entropy": 5.238237142562866,
      "epoch": 0.8482292014700968,
      "grad_norm": 1.234375,
      "learning_rate": 0.00049343096389966,
      "loss": 5.0467,
      "mean_token_accuracy": 0.19457723647356034,
      "num_tokens": 18682257.0,
      "step": 10155
    },
    {
      "entropy": 5.193057298660278,
      "epoch": 0.8486468426328099,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004934238209747724,
      "loss": 5.0388,
      "mean_token_accuracy": 0.19794312119483948,
      "num_tokens": 18691493.0,
      "step": 10160
    },
    {
      "entropy": 5.244580316543579,
      "epoch": 0.8490644837955229,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004934166742261132,
      "loss": 5.0825,
      "mean_token_accuracy": 0.1974581390619278,
      "num_tokens": 18700309.0,
      "step": 10165
    },
    {
      "entropy": 5.204142093658447,
      "epoch": 0.8494821249582358,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000493409523653807,
      "loss": 5.0857,
      "mean_token_accuracy": 0.19576020985841752,
      "num_tokens": 18709909.0,
      "step": 10170
    },
    {
      "entropy": 5.348386955261231,
      "epoch": 0.8498997661209489,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004934023692579793,
      "loss": 5.2759,
      "mean_token_accuracy": 0.18125803768634796,
      "num_tokens": 18719270.0,
      "step": 10175
    },
    {
      "entropy": 5.2265712261199955,
      "epoch": 0.8503174072836619,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004933952110387553,
      "loss": 5.0514,
      "mean_token_accuracy": 0.19341590255498886,
      "num_tokens": 18728141.0,
      "step": 10180
    },
    {
      "entropy": 5.239239454269409,
      "epoch": 0.8507350484463748,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004933880489962602,
      "loss": 5.0952,
      "mean_token_accuracy": 0.19350501596927644,
      "num_tokens": 18737925.0,
      "step": 10185
    },
    {
      "entropy": 5.255416917800903,
      "epoch": 0.8511526896090879,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004933808831306194,
      "loss": 5.0339,
      "mean_token_accuracy": 0.19600435346364975,
      "num_tokens": 18746622.0,
      "step": 10190
    },
    {
      "entropy": 5.179063606262207,
      "epoch": 0.8515703307718009,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004933737134419583,
      "loss": 5.0613,
      "mean_token_accuracy": 0.19782729744911193,
      "num_tokens": 18755316.0,
      "step": 10195
    },
    {
      "entropy": 5.176108932495117,
      "epoch": 0.8519879719345138,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004933665399304025,
      "loss": 5.0462,
      "mean_token_accuracy": 0.19505841583013533,
      "num_tokens": 18765264.0,
      "step": 10200
    },
    {
      "entropy": 5.233957576751709,
      "epoch": 0.8524056130972268,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004933593625960777,
      "loss": 5.0625,
      "mean_token_accuracy": 0.198008631169796,
      "num_tokens": 18774275.0,
      "step": 10205
    },
    {
      "entropy": 5.200260925292969,
      "epoch": 0.8528232542599399,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004933521814391094,
      "loss": 5.0551,
      "mean_token_accuracy": 0.198702871799469,
      "num_tokens": 18782848.0,
      "step": 10210
    },
    {
      "entropy": 5.218249845504761,
      "epoch": 0.8532408954226529,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004933449964596232,
      "loss": 5.1089,
      "mean_token_accuracy": 0.19394540935754775,
      "num_tokens": 18792072.0,
      "step": 10215
    },
    {
      "entropy": 5.3451320171356205,
      "epoch": 0.8536585365853658,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004933378076577451,
      "loss": 5.0968,
      "mean_token_accuracy": 0.1999950423836708,
      "num_tokens": 18800886.0,
      "step": 10220
    },
    {
      "entropy": 5.236936950683594,
      "epoch": 0.8540761777480789,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004933306150336009,
      "loss": 5.0912,
      "mean_token_accuracy": 0.20164105892181397,
      "num_tokens": 18809673.0,
      "step": 10225
    },
    {
      "entropy": 5.250994777679443,
      "epoch": 0.8544938189107919,
      "grad_norm": 1.375,
      "learning_rate": 0.0004933234185873164,
      "loss": 5.0717,
      "mean_token_accuracy": 0.20258415937423707,
      "num_tokens": 18818230.0,
      "step": 10230
    },
    {
      "entropy": 5.211773109436035,
      "epoch": 0.8549114600735048,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004933162183190177,
      "loss": 5.1338,
      "mean_token_accuracy": 0.1945189744234085,
      "num_tokens": 18827600.0,
      "step": 10235
    },
    {
      "entropy": 5.37983283996582,
      "epoch": 0.8553291012362179,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004933090142288308,
      "loss": 5.2233,
      "mean_token_accuracy": 0.1906238839030266,
      "num_tokens": 18837237.0,
      "step": 10240
    },
    {
      "entropy": 5.3497881412506105,
      "epoch": 0.8557467423989309,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004933018063168819,
      "loss": 5.1747,
      "mean_token_accuracy": 0.19596707224845886,
      "num_tokens": 18846859.0,
      "step": 10245
    },
    {
      "entropy": 5.280359506607056,
      "epoch": 0.8561643835616438,
      "grad_norm": 1.265625,
      "learning_rate": 0.000493294594583297,
      "loss": 5.1059,
      "mean_token_accuracy": 0.1929030016064644,
      "num_tokens": 18856226.0,
      "step": 10250
    },
    {
      "entropy": 5.155808019638061,
      "epoch": 0.8565820247243568,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004932873790282026,
      "loss": 4.9789,
      "mean_token_accuracy": 0.200956392288208,
      "num_tokens": 18864679.0,
      "step": 10255
    },
    {
      "entropy": 5.241528844833374,
      "epoch": 0.8569996658870699,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004932801596517248,
      "loss": 5.1633,
      "mean_token_accuracy": 0.18533792495727539,
      "num_tokens": 18874837.0,
      "step": 10260
    },
    {
      "entropy": 5.243050003051758,
      "epoch": 0.8574173070497828,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004932729364539901,
      "loss": 5.0109,
      "mean_token_accuracy": 0.2024909198284149,
      "num_tokens": 18883669.0,
      "step": 10265
    },
    {
      "entropy": 5.1785496234893795,
      "epoch": 0.8578349482124958,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004932657094351248,
      "loss": 5.0997,
      "mean_token_accuracy": 0.19862864315509796,
      "num_tokens": 18892199.0,
      "step": 10270
    },
    {
      "entropy": 5.375548553466797,
      "epoch": 0.8582525893752089,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004932584785952555,
      "loss": 5.2244,
      "mean_token_accuracy": 0.19348810166120528,
      "num_tokens": 18902393.0,
      "step": 10275
    },
    {
      "entropy": 5.2853258609771725,
      "epoch": 0.8586702305379218,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000493251243934509,
      "loss": 5.1866,
      "mean_token_accuracy": 0.19090647101402283,
      "num_tokens": 18912417.0,
      "step": 10280
    },
    {
      "entropy": 5.3069662094116214,
      "epoch": 0.8590878717006348,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004932440054530116,
      "loss": 5.1741,
      "mean_token_accuracy": 0.19052484184503554,
      "num_tokens": 18921456.0,
      "step": 10285
    },
    {
      "entropy": 5.2105762481689455,
      "epoch": 0.8595055128633479,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004932367631508903,
      "loss": 5.1191,
      "mean_token_accuracy": 0.1960864007472992,
      "num_tokens": 18930865.0,
      "step": 10290
    },
    {
      "entropy": 5.230211162567139,
      "epoch": 0.8599231540260608,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004932295170282717,
      "loss": 5.0339,
      "mean_token_accuracy": 0.1914416015148163,
      "num_tokens": 18940043.0,
      "step": 10295
    },
    {
      "entropy": 5.209995603561401,
      "epoch": 0.8603407951887738,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004932222670852828,
      "loss": 4.9968,
      "mean_token_accuracy": 0.1943082705140114,
      "num_tokens": 18948803.0,
      "step": 10300
    },
    {
      "entropy": 5.2017354488372805,
      "epoch": 0.8607584363514869,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004932150133220504,
      "loss": 5.1025,
      "mean_token_accuracy": 0.19546860903501512,
      "num_tokens": 18958014.0,
      "step": 10305
    },
    {
      "entropy": 5.165151882171631,
      "epoch": 0.8611760775141998,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004932077557387015,
      "loss": 5.0795,
      "mean_token_accuracy": 0.19663773030042647,
      "num_tokens": 18966425.0,
      "step": 10310
    },
    {
      "entropy": 5.19660005569458,
      "epoch": 0.8615937186769128,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004932004943353633,
      "loss": 5.1452,
      "mean_token_accuracy": 0.1869480550289154,
      "num_tokens": 18975445.0,
      "step": 10315
    },
    {
      "entropy": 5.377605724334717,
      "epoch": 0.8620113598396258,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004931932291121627,
      "loss": 5.1598,
      "mean_token_accuracy": 0.19668253511190414,
      "num_tokens": 18984271.0,
      "step": 10320
    },
    {
      "entropy": 5.313975191116333,
      "epoch": 0.8624290010023388,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004931859600692269,
      "loss": 5.1151,
      "mean_token_accuracy": 0.19372533708810807,
      "num_tokens": 18994179.0,
      "step": 10325
    },
    {
      "entropy": 5.196731853485107,
      "epoch": 0.8628466421650518,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004931786872066833,
      "loss": 5.0158,
      "mean_token_accuracy": 0.19707428365945817,
      "num_tokens": 19003313.0,
      "step": 10330
    },
    {
      "entropy": 5.239146041870117,
      "epoch": 0.8632642833277648,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004931714105246592,
      "loss": 5.1651,
      "mean_token_accuracy": 0.19171369969844818,
      "num_tokens": 19013982.0,
      "step": 10335
    },
    {
      "entropy": 5.201680326461792,
      "epoch": 0.8636819244904778,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004931641300232819,
      "loss": 5.0668,
      "mean_token_accuracy": 0.19870493859052657,
      "num_tokens": 19022634.0,
      "step": 10340
    },
    {
      "entropy": 5.211349678039551,
      "epoch": 0.8640995656531908,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004931568457026789,
      "loss": 5.1635,
      "mean_token_accuracy": 0.19099931567907333,
      "num_tokens": 19031903.0,
      "step": 10345
    },
    {
      "entropy": 5.3132282257080075,
      "epoch": 0.8645172068159038,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004931495575629778,
      "loss": 5.1187,
      "mean_token_accuracy": 0.19376570582389832,
      "num_tokens": 19040367.0,
      "step": 10350
    },
    {
      "entropy": 5.296181869506836,
      "epoch": 0.8649348479786167,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000493142265604306,
      "loss": 5.0824,
      "mean_token_accuracy": 0.20076291561126708,
      "num_tokens": 19049572.0,
      "step": 10355
    },
    {
      "entropy": 5.29019718170166,
      "epoch": 0.8653524891413298,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004931349698267914,
      "loss": 5.2106,
      "mean_token_accuracy": 0.1847943753004074,
      "num_tokens": 19059313.0,
      "step": 10360
    },
    {
      "entropy": 5.228671073913574,
      "epoch": 0.8657701303040428,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004931276702305615,
      "loss": 5.1047,
      "mean_token_accuracy": 0.19198494255542756,
      "num_tokens": 19067941.0,
      "step": 10365
    },
    {
      "entropy": 5.249677133560181,
      "epoch": 0.8661877714667557,
      "grad_norm": 1.125,
      "learning_rate": 0.0004931203668157441,
      "loss": 5.0521,
      "mean_token_accuracy": 0.19453371912240983,
      "num_tokens": 19077462.0,
      "step": 10370
    },
    {
      "entropy": 5.262096405029297,
      "epoch": 0.8666054126294688,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004931130595824672,
      "loss": 5.1642,
      "mean_token_accuracy": 0.19705860912799836,
      "num_tokens": 19086669.0,
      "step": 10375
    },
    {
      "entropy": 5.229280614852906,
      "epoch": 0.8670230537921818,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004931057485308587,
      "loss": 5.1279,
      "mean_token_accuracy": 0.1925350546836853,
      "num_tokens": 19095781.0,
      "step": 10380
    },
    {
      "entropy": 5.262217998504639,
      "epoch": 0.8674406949548947,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004930984336610465,
      "loss": 5.1182,
      "mean_token_accuracy": 0.19263307601213456,
      "num_tokens": 19105106.0,
      "step": 10385
    },
    {
      "entropy": 5.345351076126098,
      "epoch": 0.8678583361176078,
      "grad_norm": 1.125,
      "learning_rate": 0.0004930911149731587,
      "loss": 5.1662,
      "mean_token_accuracy": 0.19208741188049316,
      "num_tokens": 19114710.0,
      "step": 10390
    },
    {
      "entropy": 5.273029756546021,
      "epoch": 0.8682759772803208,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004930837924673233,
      "loss": 5.1505,
      "mean_token_accuracy": 0.2012007027864456,
      "num_tokens": 19123495.0,
      "step": 10395
    },
    {
      "entropy": 5.267076253890991,
      "epoch": 0.8686936184430337,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004930764661436688,
      "loss": 5.1545,
      "mean_token_accuracy": 0.18852797001600266,
      "num_tokens": 19131894.0,
      "step": 10400
    },
    {
      "entropy": 5.334974956512451,
      "epoch": 0.8691112596057468,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004930691360023231,
      "loss": 5.2038,
      "mean_token_accuracy": 0.18459987342357637,
      "num_tokens": 19141213.0,
      "step": 10405
    },
    {
      "entropy": 5.28029842376709,
      "epoch": 0.8695289007684598,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004930618020434147,
      "loss": 5.0678,
      "mean_token_accuracy": 0.19768889099359513,
      "num_tokens": 19150843.0,
      "step": 10410
    },
    {
      "entropy": 5.266949319839478,
      "epoch": 0.8699465419311727,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004930544642670721,
      "loss": 5.0873,
      "mean_token_accuracy": 0.1998864606022835,
      "num_tokens": 19160198.0,
      "step": 10415
    },
    {
      "entropy": 5.1807776927948,
      "epoch": 0.8703641830938857,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004930471226734235,
      "loss": 5.0903,
      "mean_token_accuracy": 0.19046980291604995,
      "num_tokens": 19169043.0,
      "step": 10420
    },
    {
      "entropy": 5.231479549407959,
      "epoch": 0.8707818242565988,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004930397772625976,
      "loss": 5.1255,
      "mean_token_accuracy": 0.1895667389035225,
      "num_tokens": 19177976.0,
      "step": 10425
    },
    {
      "entropy": 5.278642797470093,
      "epoch": 0.8711994654193117,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000493032428034723,
      "loss": 5.2187,
      "mean_token_accuracy": 0.19002075344324112,
      "num_tokens": 19187909.0,
      "step": 10430
    },
    {
      "entropy": 5.2143903255462645,
      "epoch": 0.8716171065820247,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004930250749899282,
      "loss": 5.0329,
      "mean_token_accuracy": 0.20639198571443557,
      "num_tokens": 19196297.0,
      "step": 10435
    },
    {
      "entropy": 5.2500848293304445,
      "epoch": 0.8720347477447378,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004930177181283422,
      "loss": 5.1713,
      "mean_token_accuracy": 0.1945100963115692,
      "num_tokens": 19205280.0,
      "step": 10440
    },
    {
      "entropy": 5.204897069931031,
      "epoch": 0.8724523889074507,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004930103574500934,
      "loss": 5.0322,
      "mean_token_accuracy": 0.19966092705726624,
      "num_tokens": 19214789.0,
      "step": 10445
    },
    {
      "entropy": 5.2736598491668705,
      "epoch": 0.8728700300701637,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004930029929553109,
      "loss": 5.1109,
      "mean_token_accuracy": 0.19714061170816422,
      "num_tokens": 19224597.0,
      "step": 10450
    },
    {
      "entropy": 5.213609647750855,
      "epoch": 0.8732876712328768,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004929956246441237,
      "loss": 5.0719,
      "mean_token_accuracy": 0.18928576856851578,
      "num_tokens": 19233716.0,
      "step": 10455
    },
    {
      "entropy": 5.222967720031738,
      "epoch": 0.8737053123955897,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004929882525166608,
      "loss": 4.9686,
      "mean_token_accuracy": 0.20641841143369674,
      "num_tokens": 19242444.0,
      "step": 10460
    },
    {
      "entropy": 5.20804271697998,
      "epoch": 0.8741229535583027,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004929808765730509,
      "loss": 5.1277,
      "mean_token_accuracy": 0.19357253313064576,
      "num_tokens": 19252380.0,
      "step": 10465
    },
    {
      "entropy": 5.258271265029907,
      "epoch": 0.8745405947210158,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004929734968134235,
      "loss": 5.1903,
      "mean_token_accuracy": 0.18180986642837524,
      "num_tokens": 19262250.0,
      "step": 10470
    },
    {
      "entropy": 5.302599859237671,
      "epoch": 0.8749582358837287,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004929661132379076,
      "loss": 5.2161,
      "mean_token_accuracy": 0.18860046267509462,
      "num_tokens": 19272228.0,
      "step": 10475
    },
    {
      "entropy": 5.323856353759766,
      "epoch": 0.8753758770464417,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004929587258466326,
      "loss": 5.1954,
      "mean_token_accuracy": 0.18734628111124038,
      "num_tokens": 19281493.0,
      "step": 10480
    },
    {
      "entropy": 5.338038778305053,
      "epoch": 0.8757935182091547,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004929513346397278,
      "loss": 5.2048,
      "mean_token_accuracy": 0.18421531319618226,
      "num_tokens": 19292300.0,
      "step": 10485
    },
    {
      "entropy": 5.313906908035278,
      "epoch": 0.8762111593718677,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004929439396173223,
      "loss": 5.1702,
      "mean_token_accuracy": 0.18565773963928223,
      "num_tokens": 19301516.0,
      "step": 10490
    },
    {
      "entropy": 5.201851177215576,
      "epoch": 0.8766288005345807,
      "grad_norm": 1.171875,
      "learning_rate": 0.000492936540779546,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20913029164075853,
      "num_tokens": 19311075.0,
      "step": 10495
    },
    {
      "entropy": 5.241045951843262,
      "epoch": 0.8770464416972937,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004929291381265281,
      "loss": 5.1369,
      "mean_token_accuracy": 0.189761783182621,
      "num_tokens": 19320605.0,
      "step": 10500
    },
    {
      "entropy": 5.243987941741944,
      "epoch": 0.8774640828600067,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004929217316583985,
      "loss": 5.0133,
      "mean_token_accuracy": 0.19928644746541976,
      "num_tokens": 19330604.0,
      "step": 10505
    },
    {
      "entropy": 5.21263108253479,
      "epoch": 0.8778817240227197,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004929143213752865,
      "loss": 5.1633,
      "mean_token_accuracy": 0.19733937233686447,
      "num_tokens": 19340650.0,
      "step": 10510
    },
    {
      "entropy": 5.371497297286988,
      "epoch": 0.8782993651854327,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000492906907277322,
      "loss": 5.315,
      "mean_token_accuracy": 0.17760407775640488,
      "num_tokens": 19350431.0,
      "step": 10515
    },
    {
      "entropy": 5.318976926803589,
      "epoch": 0.8787170063481456,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000492899489364635,
      "loss": 5.179,
      "mean_token_accuracy": 0.18663914054632186,
      "num_tokens": 19360171.0,
      "step": 10520
    },
    {
      "entropy": 5.238416910171509,
      "epoch": 0.8791346475108587,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004928920676373549,
      "loss": 5.117,
      "mean_token_accuracy": 0.19180480539798736,
      "num_tokens": 19369748.0,
      "step": 10525
    },
    {
      "entropy": 5.2819455623626705,
      "epoch": 0.8795522886735717,
      "grad_norm": 1.25,
      "learning_rate": 0.000492884642095612,
      "loss": 5.1708,
      "mean_token_accuracy": 0.18991636335849763,
      "num_tokens": 19378470.0,
      "step": 10530
    },
    {
      "entropy": 5.253831243515014,
      "epoch": 0.8799699298362846,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004928772127395361,
      "loss": 5.0772,
      "mean_token_accuracy": 0.1983237236738205,
      "num_tokens": 19387869.0,
      "step": 10535
    },
    {
      "entropy": 5.340179443359375,
      "epoch": 0.8803875709989977,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004928697795692575,
      "loss": 5.2365,
      "mean_token_accuracy": 0.184271876513958,
      "num_tokens": 19396805.0,
      "step": 10540
    },
    {
      "entropy": 5.182686042785645,
      "epoch": 0.8808052121617107,
      "grad_norm": 1.265625,
      "learning_rate": 0.000492862342584906,
      "loss": 5.035,
      "mean_token_accuracy": 0.19348762929439545,
      "num_tokens": 19405392.0,
      "step": 10545
    },
    {
      "entropy": 5.325152778625489,
      "epoch": 0.8812228533244236,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000492854901786612,
      "loss": 5.2229,
      "mean_token_accuracy": 0.19296234846115112,
      "num_tokens": 19413896.0,
      "step": 10550
    },
    {
      "entropy": 5.292904424667358,
      "epoch": 0.8816404944871367,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004928474571745058,
      "loss": 5.0862,
      "mean_token_accuracy": 0.19237807542085647,
      "num_tokens": 19422136.0,
      "step": 10555
    },
    {
      "entropy": 5.14391016960144,
      "epoch": 0.8820581356498497,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004928400087487175,
      "loss": 4.9528,
      "mean_token_accuracy": 0.19728517383337021,
      "num_tokens": 19431250.0,
      "step": 10560
    },
    {
      "entropy": 5.147278881072998,
      "epoch": 0.8824757768125626,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004928325565093777,
      "loss": 5.0427,
      "mean_token_accuracy": 0.19500260949134826,
      "num_tokens": 19439155.0,
      "step": 10565
    },
    {
      "entropy": 5.265140867233276,
      "epoch": 0.8828934179752757,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004928251004566167,
      "loss": 5.1863,
      "mean_token_accuracy": 0.18049997389316558,
      "num_tokens": 19448637.0,
      "step": 10570
    },
    {
      "entropy": 5.3482654094696045,
      "epoch": 0.8833110591379887,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004928176405905653,
      "loss": 5.189,
      "mean_token_accuracy": 0.18764340579509736,
      "num_tokens": 19457573.0,
      "step": 10575
    },
    {
      "entropy": 5.251343584060669,
      "epoch": 0.8837287003007016,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004928101769113537,
      "loss": 5.0353,
      "mean_token_accuracy": 0.20340760797262192,
      "num_tokens": 19466085.0,
      "step": 10580
    },
    {
      "entropy": 5.180642223358154,
      "epoch": 0.8841463414634146,
      "grad_norm": 1.265625,
      "learning_rate": 0.000492802709419113,
      "loss": 5.0456,
      "mean_token_accuracy": 0.2037220850586891,
      "num_tokens": 19475185.0,
      "step": 10585
    },
    {
      "entropy": 5.26083345413208,
      "epoch": 0.8845639826261277,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004927952381139735,
      "loss": 5.1699,
      "mean_token_accuracy": 0.18603014796972275,
      "num_tokens": 19484387.0,
      "step": 10590
    },
    {
      "entropy": 5.200754451751709,
      "epoch": 0.8849816237888406,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004927877629960663,
      "loss": 5.043,
      "mean_token_accuracy": 0.19057920426130295,
      "num_tokens": 19493309.0,
      "step": 10595
    },
    {
      "entropy": 5.269811058044434,
      "epoch": 0.8853992649515536,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004927802840655221,
      "loss": 5.1287,
      "mean_token_accuracy": 0.19439708888530732,
      "num_tokens": 19502284.0,
      "step": 10600
    },
    {
      "entropy": 5.339306306838989,
      "epoch": 0.8858169061142667,
      "grad_norm": 1.15625,
      "learning_rate": 0.000492772801322472,
      "loss": 5.1376,
      "mean_token_accuracy": 0.19411446750164033,
      "num_tokens": 19512720.0,
      "step": 10605
    },
    {
      "entropy": 5.2461244583129885,
      "epoch": 0.8862345472769796,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004927653147670468,
      "loss": 5.0667,
      "mean_token_accuracy": 0.2013916179537773,
      "num_tokens": 19521395.0,
      "step": 10610
    },
    {
      "entropy": 5.16604700088501,
      "epoch": 0.8866521884396926,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004927578243993777,
      "loss": 5.0372,
      "mean_token_accuracy": 0.19396515935659409,
      "num_tokens": 19530855.0,
      "step": 10615
    },
    {
      "entropy": 5.240046072006225,
      "epoch": 0.8870698296024057,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004927503302195957,
      "loss": 5.0518,
      "mean_token_accuracy": 0.19882560074329375,
      "num_tokens": 19540330.0,
      "step": 10620
    },
    {
      "entropy": 5.215531301498413,
      "epoch": 0.8874874707651186,
      "grad_norm": 1.171875,
      "learning_rate": 0.000492742832227832,
      "loss": 5.0698,
      "mean_token_accuracy": 0.1946081131696701,
      "num_tokens": 19549639.0,
      "step": 10625
    },
    {
      "entropy": 5.244396591186524,
      "epoch": 0.8879051119278316,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000492735330424218,
      "loss": 5.1566,
      "mean_token_accuracy": 0.1996217265725136,
      "num_tokens": 19558485.0,
      "step": 10630
    },
    {
      "entropy": 5.306069946289062,
      "epoch": 0.8883227530905446,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000492727824808885,
      "loss": 5.0977,
      "mean_token_accuracy": 0.1942100167274475,
      "num_tokens": 19566924.0,
      "step": 10635
    },
    {
      "entropy": 5.364131164550781,
      "epoch": 0.8887403942532576,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004927203153819643,
      "loss": 5.1587,
      "mean_token_accuracy": 0.19472651183605194,
      "num_tokens": 19575901.0,
      "step": 10640
    },
    {
      "entropy": 5.25958743095398,
      "epoch": 0.8891580354159706,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004927128021435875,
      "loss": 5.1298,
      "mean_token_accuracy": 0.1901501625776291,
      "num_tokens": 19585470.0,
      "step": 10645
    },
    {
      "entropy": 5.289333343505859,
      "epoch": 0.8895756765786836,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000492705285093886,
      "loss": 5.1414,
      "mean_token_accuracy": 0.19389011412858964,
      "num_tokens": 19594775.0,
      "step": 10650
    },
    {
      "entropy": 5.304211235046386,
      "epoch": 0.8899933177413966,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004926977642329913,
      "loss": 5.1312,
      "mean_token_accuracy": 0.19043017327785491,
      "num_tokens": 19604061.0,
      "step": 10655
    },
    {
      "entropy": 5.206965112686158,
      "epoch": 0.8904109589041096,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004926902395610353,
      "loss": 5.0574,
      "mean_token_accuracy": 0.20077430158853532,
      "num_tokens": 19613130.0,
      "step": 10660
    },
    {
      "entropy": 5.124741697311402,
      "epoch": 0.8908286000668226,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004926827110781496,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20327755063772202,
      "num_tokens": 19621951.0,
      "step": 10665
    },
    {
      "entropy": 5.177820444107056,
      "epoch": 0.8912462412295356,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000492675178784466,
      "loss": 5.0578,
      "mean_token_accuracy": 0.19026484489440917,
      "num_tokens": 19631167.0,
      "step": 10670
    },
    {
      "entropy": 5.3053307056427,
      "epoch": 0.8916638823922486,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004926676426801164,
      "loss": 5.1783,
      "mean_token_accuracy": 0.19123457968235016,
      "num_tokens": 19640548.0,
      "step": 10675
    },
    {
      "entropy": 5.225992441177368,
      "epoch": 0.8920815235549616,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004926601027652328,
      "loss": 5.083,
      "mean_token_accuracy": 0.1970062106847763,
      "num_tokens": 19650607.0,
      "step": 10680
    },
    {
      "entropy": 5.165138292312622,
      "epoch": 0.8924991647176745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000492652559039947,
      "loss": 4.9966,
      "mean_token_accuracy": 0.2015234112739563,
      "num_tokens": 19659845.0,
      "step": 10685
    },
    {
      "entropy": 5.194818210601807,
      "epoch": 0.8929168058803876,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004926450115043912,
      "loss": 5.0789,
      "mean_token_accuracy": 0.19947440475225447,
      "num_tokens": 19668863.0,
      "step": 10690
    },
    {
      "entropy": 5.210253524780273,
      "epoch": 0.8933344470431006,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004926374601586975,
      "loss": 5.0844,
      "mean_token_accuracy": 0.19982438981533052,
      "num_tokens": 19678335.0,
      "step": 10695
    },
    {
      "entropy": 5.2233171463012695,
      "epoch": 0.8937520882058135,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004926299050029981,
      "loss": 5.0808,
      "mean_token_accuracy": 0.19397748708724977,
      "num_tokens": 19687491.0,
      "step": 10700
    },
    {
      "entropy": 5.124798345565796,
      "epoch": 0.8941697293685266,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004926223460374251,
      "loss": 4.9945,
      "mean_token_accuracy": 0.20394137799739837,
      "num_tokens": 19696632.0,
      "step": 10705
    },
    {
      "entropy": 5.170961332321167,
      "epoch": 0.8945873705312396,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004926147832621111,
      "loss": 5.1438,
      "mean_token_accuracy": 0.18890809565782546,
      "num_tokens": 19706275.0,
      "step": 10710
    },
    {
      "entropy": 5.350946235656738,
      "epoch": 0.8950050116939525,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004926072166771884,
      "loss": 5.1658,
      "mean_token_accuracy": 0.18675604164600373,
      "num_tokens": 19716016.0,
      "step": 10715
    },
    {
      "entropy": 5.347976732254028,
      "epoch": 0.8954226528566656,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004925996462827895,
      "loss": 5.1013,
      "mean_token_accuracy": 0.19367033690214158,
      "num_tokens": 19725555.0,
      "step": 10720
    },
    {
      "entropy": 5.2164818286895756,
      "epoch": 0.8958402940193786,
      "grad_norm": 1.125,
      "learning_rate": 0.0004925920720790466,
      "loss": 5.1449,
      "mean_token_accuracy": 0.19570198953151702,
      "num_tokens": 19735184.0,
      "step": 10725
    },
    {
      "entropy": 5.206806325912476,
      "epoch": 0.8962579351820915,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004925844940660928,
      "loss": 5.0496,
      "mean_token_accuracy": 0.20031583905220032,
      "num_tokens": 19744065.0,
      "step": 10730
    },
    {
      "entropy": 5.23223352432251,
      "epoch": 0.8966755763448045,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004925769122440606,
      "loss": 5.0645,
      "mean_token_accuracy": 0.19948068112134934,
      "num_tokens": 19753251.0,
      "step": 10735
    },
    {
      "entropy": 5.223589324951172,
      "epoch": 0.8970932175075176,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004925693266130824,
      "loss": 5.0183,
      "mean_token_accuracy": 0.20291998535394667,
      "num_tokens": 19761290.0,
      "step": 10740
    },
    {
      "entropy": 5.302721738815308,
      "epoch": 0.8975108586702305,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004925617371732914,
      "loss": 5.0997,
      "mean_token_accuracy": 0.19556908011436464,
      "num_tokens": 19769169.0,
      "step": 10745
    },
    {
      "entropy": 5.2339414119720455,
      "epoch": 0.8979284998329435,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004925541439248203,
      "loss": 5.1186,
      "mean_token_accuracy": 0.19987917244434356,
      "num_tokens": 19778118.0,
      "step": 10750
    },
    {
      "entropy": 5.202046823501587,
      "epoch": 0.8983461409956566,
      "grad_norm": 1.265625,
      "learning_rate": 0.000492546546867802,
      "loss": 5.1588,
      "mean_token_accuracy": 0.19133514016866685,
      "num_tokens": 19787281.0,
      "step": 10755
    },
    {
      "entropy": 5.22608323097229,
      "epoch": 0.8987637821583695,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004925389460023696,
      "loss": 5.0445,
      "mean_token_accuracy": 0.19600608348846435,
      "num_tokens": 19795797.0,
      "step": 10760
    },
    {
      "entropy": 5.289606332778931,
      "epoch": 0.8991814233210825,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004925313413286559,
      "loss": 5.1401,
      "mean_token_accuracy": 0.19121022373437882,
      "num_tokens": 19804788.0,
      "step": 10765
    },
    {
      "entropy": 5.311471271514892,
      "epoch": 0.8995990644837956,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004925237328467945,
      "loss": 5.1649,
      "mean_token_accuracy": 0.19272789508104324,
      "num_tokens": 19814696.0,
      "step": 10770
    },
    {
      "entropy": 5.316903829574585,
      "epoch": 0.9000167056465085,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004925161205569182,
      "loss": 5.0972,
      "mean_token_accuracy": 0.19051541090011598,
      "num_tokens": 19824216.0,
      "step": 10775
    },
    {
      "entropy": 5.206996488571167,
      "epoch": 0.9004343468092215,
      "grad_norm": 1.25,
      "learning_rate": 0.0004925085044591604,
      "loss": 5.0755,
      "mean_token_accuracy": 0.19773166626691818,
      "num_tokens": 19832272.0,
      "step": 10780
    },
    {
      "entropy": 5.252897596359253,
      "epoch": 0.9008519879719346,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004925008845536544,
      "loss": 5.1952,
      "mean_token_accuracy": 0.19213270545005798,
      "num_tokens": 19841655.0,
      "step": 10785
    },
    {
      "entropy": 5.209755802154541,
      "epoch": 0.9012696291346475,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004924932608405336,
      "loss": 5.0542,
      "mean_token_accuracy": 0.20090115517377855,
      "num_tokens": 19851288.0,
      "step": 10790
    },
    {
      "entropy": 5.266305446624756,
      "epoch": 0.9016872702973605,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004924856333199315,
      "loss": 5.1244,
      "mean_token_accuracy": 0.1913056939840317,
      "num_tokens": 19860537.0,
      "step": 10795
    },
    {
      "entropy": 5.304848575592041,
      "epoch": 0.9021049114600735,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004924780019919816,
      "loss": 5.162,
      "mean_token_accuracy": 0.19724492132663726,
      "num_tokens": 19869618.0,
      "step": 10800
    },
    {
      "entropy": 5.248780965805054,
      "epoch": 0.9025225526227865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004924703668568174,
      "loss": 5.0654,
      "mean_token_accuracy": 0.20323309004306794,
      "num_tokens": 19879236.0,
      "step": 10805
    },
    {
      "entropy": 5.204006671905518,
      "epoch": 0.9029401937854995,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004924627279145728,
      "loss": 5.0876,
      "mean_token_accuracy": 0.19963822960853578,
      "num_tokens": 19887740.0,
      "step": 10810
    },
    {
      "entropy": 5.258492755889892,
      "epoch": 0.9033578349482125,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004924550851653814,
      "loss": 5.0504,
      "mean_token_accuracy": 0.19848864674568176,
      "num_tokens": 19896024.0,
      "step": 10815
    },
    {
      "entropy": 5.256488466262818,
      "epoch": 0.9037754761109255,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004924474386093769,
      "loss": 5.0812,
      "mean_token_accuracy": 0.1942973256111145,
      "num_tokens": 19905533.0,
      "step": 10820
    },
    {
      "entropy": 5.1709747314453125,
      "epoch": 0.9041931172736385,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004924397882466933,
      "loss": 4.9772,
      "mean_token_accuracy": 0.20180195122957229,
      "num_tokens": 19913560.0,
      "step": 10825
    },
    {
      "entropy": 5.192577600479126,
      "epoch": 0.9046107584363515,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004924321340774644,
      "loss": 5.0461,
      "mean_token_accuracy": 0.19146844744682312,
      "num_tokens": 19923159.0,
      "step": 10830
    },
    {
      "entropy": 5.125152158737182,
      "epoch": 0.9050283995990644,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004924244761018243,
      "loss": 4.9105,
      "mean_token_accuracy": 0.2023927837610245,
      "num_tokens": 19931485.0,
      "step": 10835
    },
    {
      "entropy": 5.24166054725647,
      "epoch": 0.9054460407617775,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000492416814319907,
      "loss": 5.1842,
      "mean_token_accuracy": 0.18846316039562225,
      "num_tokens": 19940552.0,
      "step": 10840
    },
    {
      "entropy": 5.260291194915771,
      "epoch": 0.9058636819244905,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004924091487318468,
      "loss": 5.0982,
      "mean_token_accuracy": 0.1973618224263191,
      "num_tokens": 19949334.0,
      "step": 10845
    },
    {
      "entropy": 5.260881805419922,
      "epoch": 0.9062813230872034,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004924014793377776,
      "loss": 5.1362,
      "mean_token_accuracy": 0.19280770570039749,
      "num_tokens": 19958569.0,
      "step": 10850
    },
    {
      "entropy": 5.245477533340454,
      "epoch": 0.9066989642499165,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004923938061378339,
      "loss": 5.1054,
      "mean_token_accuracy": 0.20338567942380906,
      "num_tokens": 19967568.0,
      "step": 10855
    },
    {
      "entropy": 5.281686592102051,
      "epoch": 0.9071166054126295,
      "grad_norm": 1.25,
      "learning_rate": 0.0004923861291321499,
      "loss": 5.1323,
      "mean_token_accuracy": 0.202757728099823,
      "num_tokens": 19976466.0,
      "step": 10860
    },
    {
      "entropy": 5.248608493804932,
      "epoch": 0.9075342465753424,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004923784483208601,
      "loss": 5.0318,
      "mean_token_accuracy": 0.2010367915034294,
      "num_tokens": 19986145.0,
      "step": 10865
    },
    {
      "entropy": 5.146245908737183,
      "epoch": 0.9079518877380555,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004923707637040989,
      "loss": 5.0026,
      "mean_token_accuracy": 0.19794430881738662,
      "num_tokens": 19994611.0,
      "step": 10870
    },
    {
      "entropy": 5.211374568939209,
      "epoch": 0.9083695289007685,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004923630752820008,
      "loss": 5.1462,
      "mean_token_accuracy": 0.18835837841033937,
      "num_tokens": 20004578.0,
      "step": 10875
    },
    {
      "entropy": 5.313225364685058,
      "epoch": 0.9087871700634814,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004923553830547004,
      "loss": 5.1708,
      "mean_token_accuracy": 0.19138211458921434,
      "num_tokens": 20014375.0,
      "step": 10880
    },
    {
      "entropy": 5.207137632369995,
      "epoch": 0.9092048112261945,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004923476870223324,
      "loss": 5.0217,
      "mean_token_accuracy": 0.1972305342555046,
      "num_tokens": 20023869.0,
      "step": 10885
    },
    {
      "entropy": 5.198520612716675,
      "epoch": 0.9096224523889075,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004923399871850315,
      "loss": 5.0785,
      "mean_token_accuracy": 0.20006888806819917,
      "num_tokens": 20033465.0,
      "step": 10890
    },
    {
      "entropy": 5.257648563385009,
      "epoch": 0.9100400935516204,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004923322835429326,
      "loss": 5.1385,
      "mean_token_accuracy": 0.18882854729890824,
      "num_tokens": 20042851.0,
      "step": 10895
    },
    {
      "entropy": 5.264758491516114,
      "epoch": 0.9104577347143334,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004923245760961704,
      "loss": 5.0609,
      "mean_token_accuracy": 0.19790236204862593,
      "num_tokens": 20052765.0,
      "step": 10900
    },
    {
      "entropy": 5.180483007431031,
      "epoch": 0.9108753758770465,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004923168648448799,
      "loss": 5.0272,
      "mean_token_accuracy": 0.19198423326015474,
      "num_tokens": 20061368.0,
      "step": 10905
    },
    {
      "entropy": 5.206907510757446,
      "epoch": 0.9112930170397594,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004923091497891961,
      "loss": 4.9937,
      "mean_token_accuracy": 0.19722962975502015,
      "num_tokens": 20070408.0,
      "step": 10910
    },
    {
      "entropy": 5.224948120117188,
      "epoch": 0.9117106582024724,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004923014309292541,
      "loss": 5.1412,
      "mean_token_accuracy": 0.18967356532812119,
      "num_tokens": 20080305.0,
      "step": 10915
    },
    {
      "entropy": 5.313349914550781,
      "epoch": 0.9121282993651855,
      "grad_norm": 1.421875,
      "learning_rate": 0.000492293708265189,
      "loss": 5.2039,
      "mean_token_accuracy": 0.1885618358850479,
      "num_tokens": 20090715.0,
      "step": 10920
    },
    {
      "entropy": 5.312185764312744,
      "epoch": 0.9125459405278984,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004922859817971359,
      "loss": 5.1537,
      "mean_token_accuracy": 0.19247750341892242,
      "num_tokens": 20099538.0,
      "step": 10925
    },
    {
      "entropy": 5.237669563293457,
      "epoch": 0.9129635816906114,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004922782515252302,
      "loss": 5.0776,
      "mean_token_accuracy": 0.19177051037549972,
      "num_tokens": 20109177.0,
      "step": 10930
    },
    {
      "entropy": 5.240066337585449,
      "epoch": 0.9133812228533245,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004922705174496072,
      "loss": 5.1292,
      "mean_token_accuracy": 0.19534749090671538,
      "num_tokens": 20117200.0,
      "step": 10935
    },
    {
      "entropy": 5.186641263961792,
      "epoch": 0.9137988640160374,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004922627795704022,
      "loss": 5.0144,
      "mean_token_accuracy": 0.19756677746772766,
      "num_tokens": 20126069.0,
      "step": 10940
    },
    {
      "entropy": 5.255351400375366,
      "epoch": 0.9142165051787504,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004922550378877507,
      "loss": 5.1211,
      "mean_token_accuracy": 0.19087007343769075,
      "num_tokens": 20136763.0,
      "step": 10945
    },
    {
      "entropy": 5.337204837799073,
      "epoch": 0.9146341463414634,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004922472924017882,
      "loss": 5.1772,
      "mean_token_accuracy": 0.18871946930885314,
      "num_tokens": 20145765.0,
      "step": 10950
    },
    {
      "entropy": 5.2932531356811525,
      "epoch": 0.9150517875041764,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004922395431126506,
      "loss": 5.1471,
      "mean_token_accuracy": 0.1875919371843338,
      "num_tokens": 20154504.0,
      "step": 10955
    },
    {
      "entropy": 5.257823657989502,
      "epoch": 0.9154694286668894,
      "grad_norm": 1.25,
      "learning_rate": 0.0004922317900204732,
      "loss": 5.1537,
      "mean_token_accuracy": 0.19012669026851653,
      "num_tokens": 20163903.0,
      "step": 10960
    },
    {
      "entropy": 5.218070697784424,
      "epoch": 0.9158870698296024,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004922240331253916,
      "loss": 5.0373,
      "mean_token_accuracy": 0.19441152960062028,
      "num_tokens": 20172910.0,
      "step": 10965
    },
    {
      "entropy": 5.180325841903686,
      "epoch": 0.9163047109923154,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004922162724275421,
      "loss": 5.0167,
      "mean_token_accuracy": 0.2012465238571167,
      "num_tokens": 20182238.0,
      "step": 10970
    },
    {
      "entropy": 5.18865852355957,
      "epoch": 0.9167223521550284,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004922085079270602,
      "loss": 4.9963,
      "mean_token_accuracy": 0.2020824134349823,
      "num_tokens": 20190673.0,
      "step": 10975
    },
    {
      "entropy": 5.313888502120972,
      "epoch": 0.9171399933177414,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004922007396240819,
      "loss": 5.2306,
      "mean_token_accuracy": 0.18127480894327164,
      "num_tokens": 20201219.0,
      "step": 10980
    },
    {
      "entropy": 5.260984754562378,
      "epoch": 0.9175576344804544,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004921929675187433,
      "loss": 5.0639,
      "mean_token_accuracy": 0.19574062377214432,
      "num_tokens": 20210756.0,
      "step": 10985
    },
    {
      "entropy": 5.267330932617187,
      "epoch": 0.9179752756431674,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004921851916111802,
      "loss": 5.0725,
      "mean_token_accuracy": 0.19784794598817826,
      "num_tokens": 20219959.0,
      "step": 10990
    },
    {
      "entropy": 5.185554504394531,
      "epoch": 0.9183929168058804,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004921774119015289,
      "loss": 5.0245,
      "mean_token_accuracy": 0.19807119071483612,
      "num_tokens": 20229633.0,
      "step": 10995
    },
    {
      "entropy": 5.091259717941284,
      "epoch": 0.9188105579685933,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004921696283899257,
      "loss": 5.02,
      "mean_token_accuracy": 0.20769816786050796,
      "num_tokens": 20238355.0,
      "step": 11000
    },
    {
      "entropy": 5.251894426345825,
      "epoch": 0.9192281991313064,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004921618410765067,
      "loss": 5.164,
      "mean_token_accuracy": 0.18543900847434996,
      "num_tokens": 20247461.0,
      "step": 11005
    },
    {
      "entropy": 5.360412073135376,
      "epoch": 0.9196458402940194,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004921540499614083,
      "loss": 5.1871,
      "mean_token_accuracy": 0.19539083391427994,
      "num_tokens": 20256505.0,
      "step": 11010
    },
    {
      "entropy": 5.260100698471069,
      "epoch": 0.9200634814567323,
      "grad_norm": 1.125,
      "learning_rate": 0.0004921462550447669,
      "loss": 5.0785,
      "mean_token_accuracy": 0.19470291435718537,
      "num_tokens": 20266169.0,
      "step": 11015
    },
    {
      "entropy": 5.139707279205322,
      "epoch": 0.9204811226194454,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004921384563267189,
      "loss": 4.9799,
      "mean_token_accuracy": 0.20562166571617127,
      "num_tokens": 20274620.0,
      "step": 11020
    },
    {
      "entropy": 5.185931539535522,
      "epoch": 0.9208987637821584,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004921306538074008,
      "loss": 4.9971,
      "mean_token_accuracy": 0.20378299057483673,
      "num_tokens": 20282829.0,
      "step": 11025
    },
    {
      "entropy": 5.188468170166016,
      "epoch": 0.9213164049448713,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004921228474869493,
      "loss": 5.0756,
      "mean_token_accuracy": 0.19281343817710878,
      "num_tokens": 20292510.0,
      "step": 11030
    },
    {
      "entropy": 5.194281053543091,
      "epoch": 0.9217340461075844,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000492115037365501,
      "loss": 5.0508,
      "mean_token_accuracy": 0.1901983931660652,
      "num_tokens": 20302095.0,
      "step": 11035
    },
    {
      "entropy": 5.3502805709838865,
      "epoch": 0.9221516872702974,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004921072234431925,
      "loss": 5.1271,
      "mean_token_accuracy": 0.19120514988899232,
      "num_tokens": 20310505.0,
      "step": 11040
    },
    {
      "entropy": 5.109669542312622,
      "epoch": 0.9225693284330103,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000492099405720161,
      "loss": 4.9946,
      "mean_token_accuracy": 0.1998417094349861,
      "num_tokens": 20319653.0,
      "step": 11045
    },
    {
      "entropy": 5.2052449703216555,
      "epoch": 0.9229869695957233,
      "grad_norm": 1.125,
      "learning_rate": 0.0004920915841965429,
      "loss": 5.0782,
      "mean_token_accuracy": 0.19663057029247283,
      "num_tokens": 20329156.0,
      "step": 11050
    },
    {
      "entropy": 5.2692626953125,
      "epoch": 0.9234046107584364,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004920837588724753,
      "loss": 5.1411,
      "mean_token_accuracy": 0.1952005445957184,
      "num_tokens": 20338761.0,
      "step": 11055
    },
    {
      "entropy": 5.152078437805176,
      "epoch": 0.9238222519211493,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004920759297480953,
      "loss": 5.0135,
      "mean_token_accuracy": 0.2031247466802597,
      "num_tokens": 20347080.0,
      "step": 11060
    },
    {
      "entropy": 5.244210433959961,
      "epoch": 0.9242398930838623,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004920680968235397,
      "loss": 5.1112,
      "mean_token_accuracy": 0.18960548639297486,
      "num_tokens": 20355824.0,
      "step": 11065
    },
    {
      "entropy": 5.2096168994903564,
      "epoch": 0.9246575342465754,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004920602600989459,
      "loss": 5.1273,
      "mean_token_accuracy": 0.19506494253873824,
      "num_tokens": 20365406.0,
      "step": 11070
    },
    {
      "entropy": 5.327987337112427,
      "epoch": 0.9250751754092883,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004920524195744509,
      "loss": 5.1592,
      "mean_token_accuracy": 0.19438378661870956,
      "num_tokens": 20375153.0,
      "step": 11075
    },
    {
      "entropy": 5.283703231811524,
      "epoch": 0.9254928165720013,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004920445752501921,
      "loss": 5.1197,
      "mean_token_accuracy": 0.19402943551540375,
      "num_tokens": 20384402.0,
      "step": 11080
    },
    {
      "entropy": 5.235755729675293,
      "epoch": 0.9259104577347144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004920367271263066,
      "loss": 5.0926,
      "mean_token_accuracy": 0.20057592689990997,
      "num_tokens": 20393913.0,
      "step": 11085
    },
    {
      "entropy": 5.120265626907349,
      "epoch": 0.9263280988974273,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004920288752029321,
      "loss": 4.9425,
      "mean_token_accuracy": 0.2154456079006195,
      "num_tokens": 20402391.0,
      "step": 11090
    },
    {
      "entropy": 5.251606893539429,
      "epoch": 0.9267457400601403,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004920210194802059,
      "loss": 5.0901,
      "mean_token_accuracy": 0.1960636094212532,
      "num_tokens": 20411682.0,
      "step": 11095
    },
    {
      "entropy": 5.187037181854248,
      "epoch": 0.9271633812228534,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004920131599582654,
      "loss": 5.1175,
      "mean_token_accuracy": 0.19483170360326768,
      "num_tokens": 20421764.0,
      "step": 11100
    },
    {
      "entropy": 5.344834232330323,
      "epoch": 0.9275810223855663,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004920052966372485,
      "loss": 5.1883,
      "mean_token_accuracy": 0.18701663613319397,
      "num_tokens": 20430420.0,
      "step": 11105
    },
    {
      "entropy": 5.310252952575683,
      "epoch": 0.9279986635482793,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004919974295172926,
      "loss": 5.1533,
      "mean_token_accuracy": 0.1896035060286522,
      "num_tokens": 20440037.0,
      "step": 11110
    },
    {
      "entropy": 5.168285655975342,
      "epoch": 0.9284163047109923,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004919895585985353,
      "loss": 5.0028,
      "mean_token_accuracy": 0.1997867465019226,
      "num_tokens": 20448307.0,
      "step": 11115
    },
    {
      "entropy": 5.181406164169312,
      "epoch": 0.9288339458737053,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004919816838811149,
      "loss": 5.0114,
      "mean_token_accuracy": 0.20516011863946915,
      "num_tokens": 20456524.0,
      "step": 11120
    },
    {
      "entropy": 5.337266445159912,
      "epoch": 0.9292515870364183,
      "grad_norm": 1.5,
      "learning_rate": 0.0004919738053651687,
      "loss": 5.2183,
      "mean_token_accuracy": 0.19022679328918457,
      "num_tokens": 20465352.0,
      "step": 11125
    },
    {
      "entropy": 5.267495965957641,
      "epoch": 0.9296692281991313,
      "grad_norm": 1.234375,
      "learning_rate": 0.000491965923050835,
      "loss": 5.007,
      "mean_token_accuracy": 0.20434942990541458,
      "num_tokens": 20474634.0,
      "step": 11130
    },
    {
      "entropy": 5.174790716171264,
      "epoch": 0.9300868693618443,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004919580369382516,
      "loss": 5.0471,
      "mean_token_accuracy": 0.199737286567688,
      "num_tokens": 20483359.0,
      "step": 11135
    },
    {
      "entropy": 5.2164794445037845,
      "epoch": 0.9305045105245573,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004919501470275567,
      "loss": 5.1703,
      "mean_token_accuracy": 0.19179025888442994,
      "num_tokens": 20494251.0,
      "step": 11140
    },
    {
      "entropy": 5.214709758758545,
      "epoch": 0.9309221516872703,
      "grad_norm": 1.375,
      "learning_rate": 0.0004919422533188882,
      "loss": 5.0251,
      "mean_token_accuracy": 0.1997251734137535,
      "num_tokens": 20503277.0,
      "step": 11145
    },
    {
      "entropy": 5.188405466079712,
      "epoch": 0.9313397928499833,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004919343558123845,
      "loss": 5.0582,
      "mean_token_accuracy": 0.19759026914834976,
      "num_tokens": 20513162.0,
      "step": 11150
    },
    {
      "entropy": 5.282330513000488,
      "epoch": 0.9317574340126963,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004919264545081838,
      "loss": 5.0757,
      "mean_token_accuracy": 0.19210958778858184,
      "num_tokens": 20522438.0,
      "step": 11155
    },
    {
      "entropy": 5.285643529891968,
      "epoch": 0.9321750751754093,
      "grad_norm": 1.25,
      "learning_rate": 0.0004919185494064244,
      "loss": 5.1217,
      "mean_token_accuracy": 0.19264516830444336,
      "num_tokens": 20531666.0,
      "step": 11160
    },
    {
      "entropy": 5.178625535964966,
      "epoch": 0.9325927163381222,
      "grad_norm": 1.375,
      "learning_rate": 0.0004919106405072446,
      "loss": 5.0725,
      "mean_token_accuracy": 0.20066103637218474,
      "num_tokens": 20540593.0,
      "step": 11165
    },
    {
      "entropy": 5.260520696640015,
      "epoch": 0.9330103575008353,
      "grad_norm": 1.3125,
      "learning_rate": 0.000491902727810783,
      "loss": 5.1831,
      "mean_token_accuracy": 0.18744262754917146,
      "num_tokens": 20550850.0,
      "step": 11170
    },
    {
      "entropy": 5.23972635269165,
      "epoch": 0.9334279986635483,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004918948113171781,
      "loss": 5.0732,
      "mean_token_accuracy": 0.18880825340747834,
      "num_tokens": 20559954.0,
      "step": 11175
    },
    {
      "entropy": 5.1876668453216555,
      "epoch": 0.9338456398262612,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004918868910265684,
      "loss": 5.0723,
      "mean_token_accuracy": 0.19822158813476562,
      "num_tokens": 20568750.0,
      "step": 11180
    },
    {
      "entropy": 5.224745655059815,
      "epoch": 0.9342632809889743,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004918789669390927,
      "loss": 5.0322,
      "mean_token_accuracy": 0.19278217554092408,
      "num_tokens": 20578615.0,
      "step": 11185
    },
    {
      "entropy": 5.3025250911712645,
      "epoch": 0.9346809221516873,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004918710390548895,
      "loss": 5.1924,
      "mean_token_accuracy": 0.19320757538080216,
      "num_tokens": 20587390.0,
      "step": 11190
    },
    {
      "entropy": 5.2235513687133786,
      "epoch": 0.9350985633144002,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004918631073740977,
      "loss": 5.0791,
      "mean_token_accuracy": 0.1922682762145996,
      "num_tokens": 20597179.0,
      "step": 11195
    },
    {
      "entropy": 5.204394435882568,
      "epoch": 0.9355162044771133,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004918551718968562,
      "loss": 5.0264,
      "mean_token_accuracy": 0.19775153398513795,
      "num_tokens": 20606074.0,
      "step": 11200
    },
    {
      "entropy": 5.113510417938232,
      "epoch": 0.9359338456398263,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000491847232623304,
      "loss": 4.9402,
      "mean_token_accuracy": 0.20684831887483596,
      "num_tokens": 20614977.0,
      "step": 11205
    },
    {
      "entropy": 5.171783351898194,
      "epoch": 0.9363514868025392,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004918392895535798,
      "loss": 5.041,
      "mean_token_accuracy": 0.1996773898601532,
      "num_tokens": 20623486.0,
      "step": 11210
    },
    {
      "entropy": 5.256261730194092,
      "epoch": 0.9367691279652522,
      "grad_norm": 1.515625,
      "learning_rate": 0.000491831342687823,
      "loss": 5.1237,
      "mean_token_accuracy": 0.19405850768089294,
      "num_tokens": 20632780.0,
      "step": 11215
    },
    {
      "entropy": 5.299038457870483,
      "epoch": 0.9371867691279653,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004918233920261723,
      "loss": 5.0795,
      "mean_token_accuracy": 0.19564387202262878,
      "num_tokens": 20642662.0,
      "step": 11220
    },
    {
      "entropy": 5.134138584136963,
      "epoch": 0.9376044102906782,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004918154375687673,
      "loss": 5.0123,
      "mean_token_accuracy": 0.1907337039709091,
      "num_tokens": 20651612.0,
      "step": 11225
    },
    {
      "entropy": 5.2558274269104,
      "epoch": 0.9380220514533912,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004918074793157472,
      "loss": 5.0946,
      "mean_token_accuracy": 0.2052842766046524,
      "num_tokens": 20659940.0,
      "step": 11230
    },
    {
      "entropy": 5.320597124099732,
      "epoch": 0.9384396926161043,
      "grad_norm": 1.25,
      "learning_rate": 0.0004917995172672511,
      "loss": 5.0957,
      "mean_token_accuracy": 0.19740411937236785,
      "num_tokens": 20669004.0,
      "step": 11235
    },
    {
      "entropy": 5.151448583602905,
      "epoch": 0.9388573337788172,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004917915514234184,
      "loss": 5.048,
      "mean_token_accuracy": 0.1947203740477562,
      "num_tokens": 20678469.0,
      "step": 11240
    },
    {
      "entropy": 5.2300561428070065,
      "epoch": 0.9392749749415302,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004917835817843887,
      "loss": 5.0961,
      "mean_token_accuracy": 0.18938689827919006,
      "num_tokens": 20688368.0,
      "step": 11245
    },
    {
      "entropy": 5.278209781646728,
      "epoch": 0.9396926161042433,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004917756083503014,
      "loss": 5.1024,
      "mean_token_accuracy": 0.1948768049478531,
      "num_tokens": 20698002.0,
      "step": 11250
    },
    {
      "entropy": 5.170444822311401,
      "epoch": 0.9401102572669562,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004917676311212962,
      "loss": 4.9805,
      "mean_token_accuracy": 0.20433486253023148,
      "num_tokens": 20707798.0,
      "step": 11255
    },
    {
      "entropy": 5.191978836059571,
      "epoch": 0.9405278984296692,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004917596500975127,
      "loss": 5.1695,
      "mean_token_accuracy": 0.19091614335775375,
      "num_tokens": 20716328.0,
      "step": 11260
    },
    {
      "entropy": 5.318720483779908,
      "epoch": 0.9409455395923823,
      "grad_norm": 1.125,
      "learning_rate": 0.0004917516652790907,
      "loss": 5.129,
      "mean_token_accuracy": 0.18771746456623079,
      "num_tokens": 20726526.0,
      "step": 11265
    },
    {
      "entropy": 5.334336757659912,
      "epoch": 0.9413631807550952,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004917436766661697,
      "loss": 5.1234,
      "mean_token_accuracy": 0.1918213739991188,
      "num_tokens": 20735086.0,
      "step": 11270
    },
    {
      "entropy": 5.1155815601348875,
      "epoch": 0.9417808219178082,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00049173568425889,
      "loss": 5.013,
      "mean_token_accuracy": 0.19667697250843047,
      "num_tokens": 20743677.0,
      "step": 11275
    },
    {
      "entropy": 5.195327043533325,
      "epoch": 0.9421984630805212,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004917276880573911,
      "loss": 5.1371,
      "mean_token_accuracy": 0.19365270137786866,
      "num_tokens": 20753351.0,
      "step": 11280
    },
    {
      "entropy": 5.301815176010132,
      "epoch": 0.9426161042432342,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004917196880618132,
      "loss": 5.1748,
      "mean_token_accuracy": 0.1888377621769905,
      "num_tokens": 20763654.0,
      "step": 11285
    },
    {
      "entropy": 5.420972490310669,
      "epoch": 0.9430337454059472,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004917116842722962,
      "loss": 5.2047,
      "mean_token_accuracy": 0.18214827477931977,
      "num_tokens": 20774424.0,
      "step": 11290
    },
    {
      "entropy": 5.304692554473877,
      "epoch": 0.9434513865686602,
      "grad_norm": 1.25,
      "learning_rate": 0.0004917036766889804,
      "loss": 5.163,
      "mean_token_accuracy": 0.19471319168806075,
      "num_tokens": 20782975.0,
      "step": 11295
    },
    {
      "entropy": 5.205314254760742,
      "epoch": 0.9438690277313732,
      "grad_norm": 1.25,
      "learning_rate": 0.0004916956653120058,
      "loss": 4.9707,
      "mean_token_accuracy": 0.20890066921710967,
      "num_tokens": 20791943.0,
      "step": 11300
    },
    {
      "entropy": 5.257737159729004,
      "epoch": 0.9442866688940862,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004916876501415129,
      "loss": 5.1569,
      "mean_token_accuracy": 0.19427275508642197,
      "num_tokens": 20800920.0,
      "step": 11305
    },
    {
      "entropy": 5.201376724243164,
      "epoch": 0.9447043100567992,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004916796311776417,
      "loss": 5.0888,
      "mean_token_accuracy": 0.18870961517095566,
      "num_tokens": 20810284.0,
      "step": 11310
    },
    {
      "entropy": 5.288651418685913,
      "epoch": 0.9451219512195121,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004916716084205328,
      "loss": 5.1404,
      "mean_token_accuracy": 0.18935603648424149,
      "num_tokens": 20819399.0,
      "step": 11315
    },
    {
      "entropy": 5.224337482452393,
      "epoch": 0.9455395923822252,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004916635818703266,
      "loss": 4.967,
      "mean_token_accuracy": 0.2022601321339607,
      "num_tokens": 20827933.0,
      "step": 11320
    },
    {
      "entropy": 5.220342350006104,
      "epoch": 0.9459572335449382,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004916555515271635,
      "loss": 5.0988,
      "mean_token_accuracy": 0.1943625971674919,
      "num_tokens": 20836636.0,
      "step": 11325
    },
    {
      "entropy": 5.260814476013183,
      "epoch": 0.9463748747076511,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004916475173911844,
      "loss": 5.0374,
      "mean_token_accuracy": 0.20449433326721192,
      "num_tokens": 20845712.0,
      "step": 11330
    },
    {
      "entropy": 5.236303758621216,
      "epoch": 0.9467925158703642,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004916394794625296,
      "loss": 5.1193,
      "mean_token_accuracy": 0.18921856880187987,
      "num_tokens": 20856874.0,
      "step": 11335
    },
    {
      "entropy": 5.236445903778076,
      "epoch": 0.9472101570330772,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004916314377413399,
      "loss": 5.1536,
      "mean_token_accuracy": 0.18807098269462585,
      "num_tokens": 20866829.0,
      "step": 11340
    },
    {
      "entropy": 5.359128189086914,
      "epoch": 0.9476277981957901,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004916233922277562,
      "loss": 5.2128,
      "mean_token_accuracy": 0.19296897500753402,
      "num_tokens": 20876409.0,
      "step": 11345
    },
    {
      "entropy": 5.323694801330566,
      "epoch": 0.9480454393585032,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004916153429219192,
      "loss": 5.1534,
      "mean_token_accuracy": 0.18872769176959991,
      "num_tokens": 20886145.0,
      "step": 11350
    },
    {
      "entropy": 5.243799114227295,
      "epoch": 0.9484630805212162,
      "grad_norm": 1.25,
      "learning_rate": 0.0004916072898239699,
      "loss": 5.0641,
      "mean_token_accuracy": 0.19877122193574906,
      "num_tokens": 20895471.0,
      "step": 11355
    },
    {
      "entropy": 5.274029493331909,
      "epoch": 0.9488807216839291,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004915992329340493,
      "loss": 5.0624,
      "mean_token_accuracy": 0.19515497237443924,
      "num_tokens": 20904213.0,
      "step": 11360
    },
    {
      "entropy": 5.2413797855377195,
      "epoch": 0.9492983628466422,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004915911722522984,
      "loss": 5.1275,
      "mean_token_accuracy": 0.19965126365423203,
      "num_tokens": 20914028.0,
      "step": 11365
    },
    {
      "entropy": 5.354149436950683,
      "epoch": 0.9497160040093552,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004915831077788583,
      "loss": 5.1703,
      "mean_token_accuracy": 0.19592286944389342,
      "num_tokens": 20923514.0,
      "step": 11370
    },
    {
      "entropy": 5.278187227249146,
      "epoch": 0.9501336451720681,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004915750395138702,
      "loss": 5.0476,
      "mean_token_accuracy": 0.1942390576004982,
      "num_tokens": 20932072.0,
      "step": 11375
    },
    {
      "entropy": 5.138377380371094,
      "epoch": 0.9505512863347811,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004915669674574754,
      "loss": 5.0244,
      "mean_token_accuracy": 0.1999897137284279,
      "num_tokens": 20940438.0,
      "step": 11380
    },
    {
      "entropy": 5.256815433502197,
      "epoch": 0.9509689274974942,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004915588916098151,
      "loss": 5.0604,
      "mean_token_accuracy": 0.19109569638967513,
      "num_tokens": 20949217.0,
      "step": 11385
    },
    {
      "entropy": 5.260021352767945,
      "epoch": 0.9513865686602071,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004915508119710307,
      "loss": 5.0646,
      "mean_token_accuracy": 0.19801012873649598,
      "num_tokens": 20958123.0,
      "step": 11390
    },
    {
      "entropy": 5.1494049549102785,
      "epoch": 0.9518042098229201,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004915427285412636,
      "loss": 4.9657,
      "mean_token_accuracy": 0.19963421076536178,
      "num_tokens": 20966908.0,
      "step": 11395
    },
    {
      "entropy": 5.206936645507812,
      "epoch": 0.9522218509856332,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004915346413206555,
      "loss": 5.0882,
      "mean_token_accuracy": 0.20151890963315963,
      "num_tokens": 20975241.0,
      "step": 11400
    },
    {
      "entropy": 5.22903881072998,
      "epoch": 0.9526394921483461,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004915265503093478,
      "loss": 5.1149,
      "mean_token_accuracy": 0.19123898148536683,
      "num_tokens": 20984336.0,
      "step": 11405
    },
    {
      "entropy": 5.1966948986053465,
      "epoch": 0.9530571333110591,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004915184555074824,
      "loss": 5.0446,
      "mean_token_accuracy": 0.20042452961206436,
      "num_tokens": 20993703.0,
      "step": 11410
    },
    {
      "entropy": 5.252202653884888,
      "epoch": 0.9534747744737722,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004915103569152005,
      "loss": 5.0284,
      "mean_token_accuracy": 0.19929608404636384,
      "num_tokens": 21002993.0,
      "step": 11415
    },
    {
      "entropy": 5.320909643173218,
      "epoch": 0.9538924156364851,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004915022545326444,
      "loss": 5.1994,
      "mean_token_accuracy": 0.18557144105434417,
      "num_tokens": 21012380.0,
      "step": 11420
    },
    {
      "entropy": 5.1560979843139645,
      "epoch": 0.9543100567991981,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004914941483599558,
      "loss": 4.9668,
      "mean_token_accuracy": 0.20174902826547622,
      "num_tokens": 21021585.0,
      "step": 11425
    },
    {
      "entropy": 5.1530389308929445,
      "epoch": 0.9547276979619111,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004914860383972763,
      "loss": 5.063,
      "mean_token_accuracy": 0.19964078813791275,
      "num_tokens": 21030461.0,
      "step": 11430
    },
    {
      "entropy": 5.316731595993042,
      "epoch": 0.9551453391246241,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004914779246447483,
      "loss": 5.1916,
      "mean_token_accuracy": 0.19034045040607453,
      "num_tokens": 21039654.0,
      "step": 11435
    },
    {
      "entropy": 5.323159790039062,
      "epoch": 0.9555629802873371,
      "grad_norm": 1.25,
      "learning_rate": 0.0004914698071025136,
      "loss": 5.182,
      "mean_token_accuracy": 0.18595382273197175,
      "num_tokens": 21048564.0,
      "step": 11440
    },
    {
      "entropy": 5.300195598602295,
      "epoch": 0.9559806214500501,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004914616857707143,
      "loss": 5.1345,
      "mean_token_accuracy": 0.18961497247219086,
      "num_tokens": 21057378.0,
      "step": 11445
    },
    {
      "entropy": 5.224439382553101,
      "epoch": 0.9563982626127631,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004914535606494927,
      "loss": 4.9706,
      "mean_token_accuracy": 0.20796577781438827,
      "num_tokens": 21066203.0,
      "step": 11450
    },
    {
      "entropy": 5.214624881744385,
      "epoch": 0.9568159037754761,
      "grad_norm": 1.09375,
      "learning_rate": 0.000491445431738991,
      "loss": 5.0542,
      "mean_token_accuracy": 0.19385241568088532,
      "num_tokens": 21075432.0,
      "step": 11455
    },
    {
      "entropy": 5.261520004272461,
      "epoch": 0.9572335449381891,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004914372990393515,
      "loss": 5.1388,
      "mean_token_accuracy": 0.19727037847042084,
      "num_tokens": 21084165.0,
      "step": 11460
    },
    {
      "entropy": 5.152784252166748,
      "epoch": 0.957651186100902,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004914291625507164,
      "loss": 5.009,
      "mean_token_accuracy": 0.20346649885177612,
      "num_tokens": 21093075.0,
      "step": 11465
    },
    {
      "entropy": 5.218937397003174,
      "epoch": 0.9580688272636151,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004914210222732284,
      "loss": 5.0449,
      "mean_token_accuracy": 0.1923544779419899,
      "num_tokens": 21101320.0,
      "step": 11470
    },
    {
      "entropy": 5.279229402542114,
      "epoch": 0.9584864684263281,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004914128782070299,
      "loss": 5.1912,
      "mean_token_accuracy": 0.19238204509019852,
      "num_tokens": 21110584.0,
      "step": 11475
    },
    {
      "entropy": 5.211146116256714,
      "epoch": 0.958904109589041,
      "grad_norm": 1.125,
      "learning_rate": 0.0004914047303522636,
      "loss": 4.9649,
      "mean_token_accuracy": 0.2016417533159256,
      "num_tokens": 21119558.0,
      "step": 11480
    },
    {
      "entropy": 5.207512331008911,
      "epoch": 0.9593217507517541,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004913965787090718,
      "loss": 5.043,
      "mean_token_accuracy": 0.1980799302458763,
      "num_tokens": 21129143.0,
      "step": 11485
    },
    {
      "entropy": 5.2786637306213375,
      "epoch": 0.9597393919144671,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004913884232775974,
      "loss": 5.1433,
      "mean_token_accuracy": 0.19374701976776124,
      "num_tokens": 21139428.0,
      "step": 11490
    },
    {
      "entropy": 5.199481582641601,
      "epoch": 0.96015703307718,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004913802640579833,
      "loss": 5.0655,
      "mean_token_accuracy": 0.20119820535182953,
      "num_tokens": 21148489.0,
      "step": 11495
    },
    {
      "entropy": 5.1523597717285154,
      "epoch": 0.9605746742398931,
      "grad_norm": 1.25,
      "learning_rate": 0.0004913721010503722,
      "loss": 4.9694,
      "mean_token_accuracy": 0.19520052671432495,
      "num_tokens": 21157419.0,
      "step": 11500
    },
    {
      "entropy": 5.122921371459961,
      "epoch": 0.9609923154026061,
      "grad_norm": 1.28125,
      "learning_rate": 0.000491363934254907,
      "loss": 4.9978,
      "mean_token_accuracy": 0.2023424580693245,
      "num_tokens": 21166628.0,
      "step": 11505
    },
    {
      "entropy": 5.185730600357056,
      "epoch": 0.961409956565319,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004913557636717307,
      "loss": 5.0624,
      "mean_token_accuracy": 0.1961621105670929,
      "num_tokens": 21176335.0,
      "step": 11510
    },
    {
      "entropy": 5.271285629272461,
      "epoch": 0.9618275977280321,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004913475893009864,
      "loss": 5.0375,
      "mean_token_accuracy": 0.20356837809085845,
      "num_tokens": 21184485.0,
      "step": 11515
    },
    {
      "entropy": 5.305818939208985,
      "epoch": 0.9622452388907451,
      "grad_norm": 1.328125,
      "learning_rate": 0.000491339411142817,
      "loss": 5.1098,
      "mean_token_accuracy": 0.1950333833694458,
      "num_tokens": 21194216.0,
      "step": 11520
    },
    {
      "entropy": 5.18507866859436,
      "epoch": 0.962662880053458,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004913312291973659,
      "loss": 5.101,
      "mean_token_accuracy": 0.19253588169813157,
      "num_tokens": 21203902.0,
      "step": 11525
    },
    {
      "entropy": 5.20624737739563,
      "epoch": 0.963080521216171,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004913230434647762,
      "loss": 5.0272,
      "mean_token_accuracy": 0.2101624295115471,
      "num_tokens": 21212618.0,
      "step": 11530
    },
    {
      "entropy": 5.266144752502441,
      "epoch": 0.9634981623788841,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004913148539451912,
      "loss": 5.0594,
      "mean_token_accuracy": 0.19851963073015214,
      "num_tokens": 21221820.0,
      "step": 11535
    },
    {
      "entropy": 5.1773560523986815,
      "epoch": 0.963915803541597,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004913066606387544,
      "loss": 5.1469,
      "mean_token_accuracy": 0.18888263404369354,
      "num_tokens": 21231085.0,
      "step": 11540
    },
    {
      "entropy": 5.215986490249634,
      "epoch": 0.96433344470431,
      "grad_norm": 1.296875,
      "learning_rate": 0.000491298463545609,
      "loss": 5.0629,
      "mean_token_accuracy": 0.1963535025715828,
      "num_tokens": 21240184.0,
      "step": 11545
    },
    {
      "entropy": 5.223777341842651,
      "epoch": 0.9647510858670231,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004912902626658988,
      "loss": 5.085,
      "mean_token_accuracy": 0.20060379803180695,
      "num_tokens": 21249085.0,
      "step": 11550
    },
    {
      "entropy": 5.217039728164673,
      "epoch": 0.965168727029736,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004912820579997672,
      "loss": 5.0314,
      "mean_token_accuracy": 0.20666501224040984,
      "num_tokens": 21258188.0,
      "step": 11555
    },
    {
      "entropy": 5.201117610931396,
      "epoch": 0.965586368192449,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004912738495473577,
      "loss": 4.9753,
      "mean_token_accuracy": 0.2020363837480545,
      "num_tokens": 21267223.0,
      "step": 11560
    },
    {
      "entropy": 5.204726505279541,
      "epoch": 0.9660040093551621,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004912656373088142,
      "loss": 5.091,
      "mean_token_accuracy": 0.20087748914957046,
      "num_tokens": 21276503.0,
      "step": 11565
    },
    {
      "entropy": 5.248251008987427,
      "epoch": 0.9664216505178751,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004912574212842804,
      "loss": 5.1315,
      "mean_token_accuracy": 0.1960167795419693,
      "num_tokens": 21286775.0,
      "step": 11570
    },
    {
      "entropy": 5.250494861602784,
      "epoch": 0.966839291680588,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004912492014739001,
      "loss": 5.0255,
      "mean_token_accuracy": 0.20541923195123674,
      "num_tokens": 21295956.0,
      "step": 11575
    },
    {
      "entropy": 5.1827648162841795,
      "epoch": 0.967256932843301,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004912409778778173,
      "loss": 5.0667,
      "mean_token_accuracy": 0.19221061617136,
      "num_tokens": 21305050.0,
      "step": 11580
    },
    {
      "entropy": 5.0948080062866214,
      "epoch": 0.9676745740060141,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004912327504961758,
      "loss": 4.9924,
      "mean_token_accuracy": 0.196906778216362,
      "num_tokens": 21313732.0,
      "step": 11585
    },
    {
      "entropy": 5.2102367877960205,
      "epoch": 0.968092215168727,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004912245193291198,
      "loss": 5.0319,
      "mean_token_accuracy": 0.19872675240039825,
      "num_tokens": 21322526.0,
      "step": 11590
    },
    {
      "entropy": 5.330317831039428,
      "epoch": 0.96850985633144,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004912162843767932,
      "loss": 5.1604,
      "mean_token_accuracy": 0.18777193427085875,
      "num_tokens": 21332243.0,
      "step": 11595
    },
    {
      "entropy": 5.290076112747192,
      "epoch": 0.9689274974941531,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004912080456393404,
      "loss": 5.087,
      "mean_token_accuracy": 0.19114756137132644,
      "num_tokens": 21340827.0,
      "step": 11600
    },
    {
      "entropy": 5.238047361373901,
      "epoch": 0.969345138656866,
      "grad_norm": 1.125,
      "learning_rate": 0.0004911998031169056,
      "loss": 5.0509,
      "mean_token_accuracy": 0.20302327424287797,
      "num_tokens": 21350506.0,
      "step": 11605
    },
    {
      "entropy": 5.1079185009002686,
      "epoch": 0.969762779819579,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004911915568096328,
      "loss": 4.9981,
      "mean_token_accuracy": 0.20552604645490646,
      "num_tokens": 21359690.0,
      "step": 11610
    },
    {
      "entropy": 5.253306722640991,
      "epoch": 0.9701804209822921,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004911833067176666,
      "loss": 5.1319,
      "mean_token_accuracy": 0.18593323677778245,
      "num_tokens": 21368953.0,
      "step": 11615
    },
    {
      "entropy": 5.418764972686768,
      "epoch": 0.970598062145005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004911750528411515,
      "loss": 5.2745,
      "mean_token_accuracy": 0.1897052988409996,
      "num_tokens": 21379015.0,
      "step": 11620
    },
    {
      "entropy": 5.212120008468628,
      "epoch": 0.971015703307718,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004911667951802318,
      "loss": 4.9982,
      "mean_token_accuracy": 0.20389384478330613,
      "num_tokens": 21388555.0,
      "step": 11625
    },
    {
      "entropy": 5.146392488479615,
      "epoch": 0.9714333444704311,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004911585337350524,
      "loss": 5.1613,
      "mean_token_accuracy": 0.18404101133346557,
      "num_tokens": 21399011.0,
      "step": 11630
    },
    {
      "entropy": 5.22923526763916,
      "epoch": 0.971850985633144,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004911502685057573,
      "loss": 5.0736,
      "mean_token_accuracy": 0.19762255996465683,
      "num_tokens": 21408711.0,
      "step": 11635
    },
    {
      "entropy": 5.322312688827514,
      "epoch": 0.972268626795857,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004911419994924918,
      "loss": 5.0824,
      "mean_token_accuracy": 0.18901695311069489,
      "num_tokens": 21417995.0,
      "step": 11640
    },
    {
      "entropy": 5.203036832809448,
      "epoch": 0.97268626795857,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004911337266954004,
      "loss": 5.0998,
      "mean_token_accuracy": 0.19022010415792465,
      "num_tokens": 21426550.0,
      "step": 11645
    },
    {
      "entropy": 5.164367580413819,
      "epoch": 0.973103909121283,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004911254501146279,
      "loss": 4.9446,
      "mean_token_accuracy": 0.20207127183675766,
      "num_tokens": 21435396.0,
      "step": 11650
    },
    {
      "entropy": 5.0929131507873535,
      "epoch": 0.973521550283996,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004911171697503193,
      "loss": 4.9324,
      "mean_token_accuracy": 0.21031402200460433,
      "num_tokens": 21445156.0,
      "step": 11655
    },
    {
      "entropy": 5.159189653396607,
      "epoch": 0.973939191446709,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004911088856026194,
      "loss": 5.1429,
      "mean_token_accuracy": 0.19453548192977904,
      "num_tokens": 21454924.0,
      "step": 11660
    },
    {
      "entropy": 5.235403156280517,
      "epoch": 0.974356832609422,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004911005976716734,
      "loss": 5.0152,
      "mean_token_accuracy": 0.20147766321897506,
      "num_tokens": 21465062.0,
      "step": 11665
    },
    {
      "entropy": 5.131636285781861,
      "epoch": 0.974774473772135,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004910923059576264,
      "loss": 5.0454,
      "mean_token_accuracy": 0.18579394072294236,
      "num_tokens": 21473744.0,
      "step": 11670
    },
    {
      "entropy": 5.230635404586792,
      "epoch": 0.975192114934848,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004910840104606234,
      "loss": 5.062,
      "mean_token_accuracy": 0.1954369306564331,
      "num_tokens": 21482913.0,
      "step": 11675
    },
    {
      "entropy": 5.258983325958252,
      "epoch": 0.975609756097561,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004910757111808097,
      "loss": 5.1629,
      "mean_token_accuracy": 0.1843745231628418,
      "num_tokens": 21491551.0,
      "step": 11680
    },
    {
      "entropy": 5.22159514427185,
      "epoch": 0.976027397260274,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004910674081183305,
      "loss": 5.0346,
      "mean_token_accuracy": 0.19817342311143876,
      "num_tokens": 21500389.0,
      "step": 11685
    },
    {
      "entropy": 5.22039532661438,
      "epoch": 0.976445038422987,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004910591012733313,
      "loss": 5.111,
      "mean_token_accuracy": 0.1977502539753914,
      "num_tokens": 21509720.0,
      "step": 11690
    },
    {
      "entropy": 5.2274007320404055,
      "epoch": 0.9768626795857,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004910507906459575,
      "loss": 5.1083,
      "mean_token_accuracy": 0.19756276905536652,
      "num_tokens": 21519326.0,
      "step": 11695
    },
    {
      "entropy": 5.226836395263672,
      "epoch": 0.977280320748413,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004910424762363545,
      "loss": 5.0536,
      "mean_token_accuracy": 0.1976720377802849,
      "num_tokens": 21528326.0,
      "step": 11700
    },
    {
      "entropy": 5.215452671051025,
      "epoch": 0.977697961911126,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000491034158044668,
      "loss": 5.0548,
      "mean_token_accuracy": 0.199582277238369,
      "num_tokens": 21536662.0,
      "step": 11705
    },
    {
      "entropy": 5.172195053100586,
      "epoch": 0.9781156030738389,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004910258360710434,
      "loss": 5.0278,
      "mean_token_accuracy": 0.2038472041487694,
      "num_tokens": 21545222.0,
      "step": 11710
    },
    {
      "entropy": 5.24199070930481,
      "epoch": 0.978533244236552,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004910175103156266,
      "loss": 5.0443,
      "mean_token_accuracy": 0.19624391198158264,
      "num_tokens": 21554051.0,
      "step": 11715
    },
    {
      "entropy": 5.171717643737793,
      "epoch": 0.978950885399265,
      "grad_norm": 1.25,
      "learning_rate": 0.0004910091807785632,
      "loss": 5.0108,
      "mean_token_accuracy": 0.2069511115550995,
      "num_tokens": 21562031.0,
      "step": 11720
    },
    {
      "entropy": 5.1460693836212155,
      "epoch": 0.9793685265619779,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004910008474599991,
      "loss": 4.9524,
      "mean_token_accuracy": 0.21157848984003066,
      "num_tokens": 21570758.0,
      "step": 11725
    },
    {
      "entropy": 5.1400731086730955,
      "epoch": 0.979786167724691,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004909925103600801,
      "loss": 5.0598,
      "mean_token_accuracy": 0.20378078520298004,
      "num_tokens": 21579605.0,
      "step": 11730
    },
    {
      "entropy": 5.289452171325683,
      "epoch": 0.980203808887404,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004909841694789523,
      "loss": 5.1556,
      "mean_token_accuracy": 0.18420615345239638,
      "num_tokens": 21588719.0,
      "step": 11735
    },
    {
      "entropy": 5.3434230327606205,
      "epoch": 0.9806214500501169,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004909758248167616,
      "loss": 5.1989,
      "mean_token_accuracy": 0.18786609619855882,
      "num_tokens": 21598038.0,
      "step": 11740
    },
    {
      "entropy": 5.268276119232178,
      "epoch": 0.98103909121283,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004909674763736542,
      "loss": 5.1406,
      "mean_token_accuracy": 0.18748406916856766,
      "num_tokens": 21607378.0,
      "step": 11745
    },
    {
      "entropy": 5.2020691394805905,
      "epoch": 0.981456732375543,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004909591241497761,
      "loss": 5.0712,
      "mean_token_accuracy": 0.19744382947683334,
      "num_tokens": 21616470.0,
      "step": 11750
    },
    {
      "entropy": 5.272403907775879,
      "epoch": 0.9818743735382559,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004909507681452736,
      "loss": 5.1191,
      "mean_token_accuracy": 0.18930997252464293,
      "num_tokens": 21626005.0,
      "step": 11755
    },
    {
      "entropy": 5.246461296081543,
      "epoch": 0.9822920147009689,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004909424083602929,
      "loss": 4.9739,
      "mean_token_accuracy": 0.20637580752372742,
      "num_tokens": 21635620.0,
      "step": 11760
    },
    {
      "entropy": 5.084247064590454,
      "epoch": 0.982709655863682,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004909340447949805,
      "loss": 4.9965,
      "mean_token_accuracy": 0.2055913195014,
      "num_tokens": 21644846.0,
      "step": 11765
    },
    {
      "entropy": 5.129577493667602,
      "epoch": 0.9831272970263949,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004909256774494827,
      "loss": 4.9648,
      "mean_token_accuracy": 0.19818419516086577,
      "num_tokens": 21653275.0,
      "step": 11770
    },
    {
      "entropy": 5.234425020217896,
      "epoch": 0.9835449381891079,
      "grad_norm": 1.28125,
      "learning_rate": 0.000490917306323946,
      "loss": 5.0995,
      "mean_token_accuracy": 0.20093469470739364,
      "num_tokens": 21661709.0,
      "step": 11775
    },
    {
      "entropy": 5.252344799041748,
      "epoch": 0.983962579351821,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004909089314185169,
      "loss": 5.101,
      "mean_token_accuracy": 0.19146595895290375,
      "num_tokens": 21672409.0,
      "step": 11780
    },
    {
      "entropy": 5.257564687728882,
      "epoch": 0.9843802205145339,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004909005527333421,
      "loss": 4.9536,
      "mean_token_accuracy": 0.20542110055685042,
      "num_tokens": 21680818.0,
      "step": 11785
    },
    {
      "entropy": 5.236087417602539,
      "epoch": 0.9847978616772469,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004908921702685683,
      "loss": 5.1146,
      "mean_token_accuracy": 0.20375936329364777,
      "num_tokens": 21689572.0,
      "step": 11790
    },
    {
      "entropy": 5.142813396453858,
      "epoch": 0.98521550283996,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004908837840243421,
      "loss": 5.0916,
      "mean_token_accuracy": 0.19214267432689666,
      "num_tokens": 21698244.0,
      "step": 11795
    },
    {
      "entropy": 5.223724031448365,
      "epoch": 0.9856331440026729,
      "grad_norm": 1.25,
      "learning_rate": 0.0004908753940008105,
      "loss": 4.952,
      "mean_token_accuracy": 0.2127091482281685,
      "num_tokens": 21706532.0,
      "step": 11800
    },
    {
      "entropy": 5.159911632537842,
      "epoch": 0.9860507851653859,
      "grad_norm": 1.25,
      "learning_rate": 0.0004908670001981201,
      "loss": 5.0205,
      "mean_token_accuracy": 0.20355266332626343,
      "num_tokens": 21715386.0,
      "step": 11805
    },
    {
      "entropy": 5.180237865447998,
      "epoch": 0.986468426328099,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004908586026164182,
      "loss": 5.0228,
      "mean_token_accuracy": 0.19893961995840073,
      "num_tokens": 21724544.0,
      "step": 11810
    },
    {
      "entropy": 5.314064025878906,
      "epoch": 0.9868860674908119,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004908502012558516,
      "loss": 5.1538,
      "mean_token_accuracy": 0.1846147820353508,
      "num_tokens": 21734453.0,
      "step": 11815
    },
    {
      "entropy": 5.215813064575196,
      "epoch": 0.9873037086535249,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004908417961165674,
      "loss": 5.0405,
      "mean_token_accuracy": 0.1935063660144806,
      "num_tokens": 21744243.0,
      "step": 11820
    },
    {
      "entropy": 5.1229248046875,
      "epoch": 0.9877213498162379,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004908333871987126,
      "loss": 4.9939,
      "mean_token_accuracy": 0.1990435466170311,
      "num_tokens": 21751843.0,
      "step": 11825
    },
    {
      "entropy": 5.199074554443359,
      "epoch": 0.9881389909789509,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004908249745024347,
      "loss": 5.0078,
      "mean_token_accuracy": 0.1981030821800232,
      "num_tokens": 21760812.0,
      "step": 11830
    },
    {
      "entropy": 5.271065616607666,
      "epoch": 0.9885566321416639,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000490816558027881,
      "loss": 5.0028,
      "mean_token_accuracy": 0.20252378433942794,
      "num_tokens": 21769460.0,
      "step": 11835
    },
    {
      "entropy": 5.15740180015564,
      "epoch": 0.9889742733043769,
      "grad_norm": 1.25,
      "learning_rate": 0.0004908081377751985,
      "loss": 4.9841,
      "mean_token_accuracy": 0.2069212093949318,
      "num_tokens": 21778440.0,
      "step": 11840
    },
    {
      "entropy": 5.261011838912964,
      "epoch": 0.9893919144670899,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004907997137445347,
      "loss": 5.1681,
      "mean_token_accuracy": 0.19034019410610198,
      "num_tokens": 21787708.0,
      "step": 11845
    },
    {
      "entropy": 5.185519695281982,
      "epoch": 0.9898095556298029,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004907912859360373,
      "loss": 5.0166,
      "mean_token_accuracy": 0.1997847318649292,
      "num_tokens": 21796517.0,
      "step": 11850
    },
    {
      "entropy": 5.1391078472137455,
      "epoch": 0.9902271967925159,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004907828543498536,
      "loss": 5.0265,
      "mean_token_accuracy": 0.2024710774421692,
      "num_tokens": 21805833.0,
      "step": 11855
    },
    {
      "entropy": 5.2227089405059814,
      "epoch": 0.9906448379552288,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004907744189861313,
      "loss": 5.0708,
      "mean_token_accuracy": 0.19794497787952423,
      "num_tokens": 21815953.0,
      "step": 11860
    },
    {
      "entropy": 5.237931632995606,
      "epoch": 0.9910624791179419,
      "grad_norm": 1.59375,
      "learning_rate": 0.000490765979845018,
      "loss": 5.0517,
      "mean_token_accuracy": 0.19614858776330948,
      "num_tokens": 21824780.0,
      "step": 11865
    },
    {
      "entropy": 5.17510495185852,
      "epoch": 0.9914801202806549,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004907575369266617,
      "loss": 5.0499,
      "mean_token_accuracy": 0.19313737154006957,
      "num_tokens": 21834514.0,
      "step": 11870
    },
    {
      "entropy": 5.250283336639404,
      "epoch": 0.9918977614433678,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004907490902312099,
      "loss": 5.0631,
      "mean_token_accuracy": 0.20352939665317535,
      "num_tokens": 21844090.0,
      "step": 11875
    },
    {
      "entropy": 5.19174108505249,
      "epoch": 0.9923154026060809,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004907406397588106,
      "loss": 5.0493,
      "mean_token_accuracy": 0.2015935465693474,
      "num_tokens": 21852982.0,
      "step": 11880
    },
    {
      "entropy": 5.221615839004516,
      "epoch": 0.9927330437687939,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004907321855096117,
      "loss": 5.1369,
      "mean_token_accuracy": 0.18960328102111818,
      "num_tokens": 21862734.0,
      "step": 11885
    },
    {
      "entropy": 5.364752626419067,
      "epoch": 0.9931506849315068,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004907237274837613,
      "loss": 5.1308,
      "mean_token_accuracy": 0.19494061917066574,
      "num_tokens": 21871927.0,
      "step": 11890
    },
    {
      "entropy": 5.1199297428131105,
      "epoch": 0.9935683260942199,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004907152656814072,
      "loss": 4.9387,
      "mean_token_accuracy": 0.20188662260770798,
      "num_tokens": 21881409.0,
      "step": 11895
    },
    {
      "entropy": 5.221092414855957,
      "epoch": 0.9939859672569329,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004907068001026978,
      "loss": 5.0282,
      "mean_token_accuracy": 0.2049349531531334,
      "num_tokens": 21890324.0,
      "step": 11900
    },
    {
      "entropy": 5.215235471725464,
      "epoch": 0.9944036084196458,
      "grad_norm": 1.25,
      "learning_rate": 0.0004906983307477813,
      "loss": 4.9838,
      "mean_token_accuracy": 0.2050636351108551,
      "num_tokens": 21898352.0,
      "step": 11905
    },
    {
      "entropy": 5.168635559082031,
      "epoch": 0.9948212495823588,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004906898576168057,
      "loss": 5.0071,
      "mean_token_accuracy": 0.20151031911373138,
      "num_tokens": 21906972.0,
      "step": 11910
    },
    {
      "entropy": 5.253898000717163,
      "epoch": 0.9952388907450719,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004906813807099197,
      "loss": 5.1078,
      "mean_token_accuracy": 0.19511166960000992,
      "num_tokens": 21916304.0,
      "step": 11915
    },
    {
      "entropy": 5.229467058181763,
      "epoch": 0.9956565319077848,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004906729000272714,
      "loss": 5.1169,
      "mean_token_accuracy": 0.19260700941085815,
      "num_tokens": 21926393.0,
      "step": 11920
    },
    {
      "entropy": 5.276017665863037,
      "epoch": 0.9960741730704978,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004906644155690095,
      "loss": 5.043,
      "mean_token_accuracy": 0.19708795547485353,
      "num_tokens": 21935657.0,
      "step": 11925
    },
    {
      "entropy": 5.30380859375,
      "epoch": 0.9964918142332109,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004906559273352823,
      "loss": 5.1019,
      "mean_token_accuracy": 0.19625190198421477,
      "num_tokens": 21945170.0,
      "step": 11930
    },
    {
      "entropy": 5.218558120727539,
      "epoch": 0.9969094553959238,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004906474353262386,
      "loss": 5.1229,
      "mean_token_accuracy": 0.19573179632425308,
      "num_tokens": 21954043.0,
      "step": 11935
    },
    {
      "entropy": 5.339192914962768,
      "epoch": 0.9973270965586368,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004906389395420269,
      "loss": 5.2255,
      "mean_token_accuracy": 0.18686571717262268,
      "num_tokens": 21963716.0,
      "step": 11940
    },
    {
      "entropy": 5.413056135177612,
      "epoch": 0.9977447377213499,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000490630439982796,
      "loss": 5.2052,
      "mean_token_accuracy": 0.18740773797035218,
      "num_tokens": 21973059.0,
      "step": 11945
    },
    {
      "entropy": 5.217510223388672,
      "epoch": 0.9981623788840628,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004906219366486947,
      "loss": 5.089,
      "mean_token_accuracy": 0.18914537727832795,
      "num_tokens": 21983189.0,
      "step": 11950
    },
    {
      "entropy": 5.200915908813476,
      "epoch": 0.9985800200467758,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004906134295398719,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20520598143339158,
      "num_tokens": 21992124.0,
      "step": 11955
    },
    {
      "entropy": 5.149120664596557,
      "epoch": 0.9989976612094889,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004906049186564764,
      "loss": 4.9893,
      "mean_token_accuracy": 0.20018115490674973,
      "num_tokens": 22002075.0,
      "step": 11960
    },
    {
      "entropy": 5.185933589935303,
      "epoch": 0.9994153023722018,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004905964039986574,
      "loss": 5.1241,
      "mean_token_accuracy": 0.1803090453147888,
      "num_tokens": 22010831.0,
      "step": 11965
    },
    {
      "entropy": 5.278057432174682,
      "epoch": 0.9998329435349148,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004905878855665637,
      "loss": 5.0838,
      "mean_token_accuracy": 0.19286288917064667,
      "num_tokens": 22020270.0,
      "step": 11970
    },
    {
      "entropy": 5.303911781311035,
      "epoch": 1.0002505846976277,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004905793633603446,
      "loss": 5.0136,
      "mean_token_accuracy": 0.1979333996772766,
      "num_tokens": 22030413.0,
      "step": 11975
    },
    {
      "entropy": 5.16483359336853,
      "epoch": 1.0006682258603408,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004905708373801492,
      "loss": 5.0115,
      "mean_token_accuracy": 0.20012013018131256,
      "num_tokens": 22040420.0,
      "step": 11980
    },
    {
      "entropy": 5.207770442962646,
      "epoch": 1.0010858670230538,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004905623076261269,
      "loss": 5.0625,
      "mean_token_accuracy": 0.1999570295214653,
      "num_tokens": 22049901.0,
      "step": 11985
    },
    {
      "entropy": 5.2084156513214115,
      "epoch": 1.0015035081857668,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000490553774098427,
      "loss": 4.9804,
      "mean_token_accuracy": 0.19033000320196153,
      "num_tokens": 22058939.0,
      "step": 11990
    },
    {
      "entropy": 5.133357000350952,
      "epoch": 1.0019211493484799,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004905452367971986,
      "loss": 4.9003,
      "mean_token_accuracy": 0.21050990521907806,
      "num_tokens": 22067553.0,
      "step": 11995
    },
    {
      "entropy": 5.148767614364624,
      "epoch": 1.0023387905111927,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004905366957225916,
      "loss": 5.0081,
      "mean_token_accuracy": 0.19771587252616882,
      "num_tokens": 22076264.0,
      "step": 12000
    },
    {
      "epoch": 1.0023387905111927,
      "eval_entropy": 5.040127530539347,
      "eval_loss": 5.047482967376709,
      "eval_mean_token_accuracy": 0.20809941453667066,
      "eval_num_tokens": 22076264.0,
      "eval_runtime": 27.8658,
      "eval_samples_per_second": 1355.176,
      "eval_steps_per_second": 169.419,
      "step": 12000
    },
    {
      "entropy": 5.169580554962158,
      "epoch": 1.0027564316739057,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004905281508747552,
      "loss": 4.8675,
      "mean_token_accuracy": 0.21027906537055968,
      "num_tokens": 22085033.0,
      "step": 12005
    },
    {
      "entropy": 5.161483812332153,
      "epoch": 1.0031740728366187,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004905196022538391,
      "loss": 5.0155,
      "mean_token_accuracy": 0.19453056305646896,
      "num_tokens": 22094619.0,
      "step": 12010
    },
    {
      "entropy": 5.228329372406006,
      "epoch": 1.0035917139993318,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000490511049859993,
      "loss": 5.0366,
      "mean_token_accuracy": 0.20216394513845443,
      "num_tokens": 22103611.0,
      "step": 12015
    },
    {
      "entropy": 5.27105746269226,
      "epoch": 1.0040093551620448,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004905024936933664,
      "loss": 5.0824,
      "mean_token_accuracy": 0.1958231121301651,
      "num_tokens": 22112478.0,
      "step": 12020
    },
    {
      "entropy": 5.13277850151062,
      "epoch": 1.0044269963247578,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004904939337541093,
      "loss": 4.8915,
      "mean_token_accuracy": 0.20297117084264754,
      "num_tokens": 22121969.0,
      "step": 12025
    },
    {
      "entropy": 5.204805755615235,
      "epoch": 1.0048446374874707,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004904853700423716,
      "loss": 4.9783,
      "mean_token_accuracy": 0.20368643403053283,
      "num_tokens": 22131241.0,
      "step": 12030
    },
    {
      "entropy": 5.211204957962036,
      "epoch": 1.0052622786501837,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004904768025583031,
      "loss": 4.9736,
      "mean_token_accuracy": 0.20170041024684907,
      "num_tokens": 22140473.0,
      "step": 12035
    },
    {
      "entropy": 5.18708381652832,
      "epoch": 1.0056799198128967,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004904682313020537,
      "loss": 5.0563,
      "mean_token_accuracy": 0.19304264485836028,
      "num_tokens": 22150071.0,
      "step": 12040
    },
    {
      "entropy": 5.315757369995117,
      "epoch": 1.0060975609756098,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004904596562737737,
      "loss": 5.1198,
      "mean_token_accuracy": 0.19075191915035247,
      "num_tokens": 22159665.0,
      "step": 12045
    },
    {
      "entropy": 5.163830184936524,
      "epoch": 1.0065152021383228,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000490451077473613,
      "loss": 4.9797,
      "mean_token_accuracy": 0.20455554127693176,
      "num_tokens": 22168856.0,
      "step": 12050
    },
    {
      "entropy": 5.133738899230957,
      "epoch": 1.0069328433010358,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004904424949017218,
      "loss": 4.987,
      "mean_token_accuracy": 0.1988204911351204,
      "num_tokens": 22177550.0,
      "step": 12055
    },
    {
      "entropy": 5.186951541900635,
      "epoch": 1.0073504844637486,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004904339085582504,
      "loss": 4.911,
      "mean_token_accuracy": 0.21347827911376954,
      "num_tokens": 22186839.0,
      "step": 12060
    },
    {
      "entropy": 5.282240104675293,
      "epoch": 1.0077681256264617,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004904253184433491,
      "loss": 5.0658,
      "mean_token_accuracy": 0.19386889189481735,
      "num_tokens": 22196865.0,
      "step": 12065
    },
    {
      "entropy": 5.134208154678345,
      "epoch": 1.0081857667891747,
      "grad_norm": 1.25,
      "learning_rate": 0.0004904167245571684,
      "loss": 4.9216,
      "mean_token_accuracy": 0.20560819208621978,
      "num_tokens": 22205140.0,
      "step": 12070
    },
    {
      "entropy": 5.072881841659546,
      "epoch": 1.0086034079518877,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004904081268998586,
      "loss": 4.9266,
      "mean_token_accuracy": 0.20570905208587648,
      "num_tokens": 22214846.0,
      "step": 12075
    },
    {
      "entropy": 5.169860172271728,
      "epoch": 1.0090210491146008,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004903995254715702,
      "loss": 5.0409,
      "mean_token_accuracy": 0.19393994361162187,
      "num_tokens": 22222984.0,
      "step": 12080
    },
    {
      "entropy": 5.253694105148315,
      "epoch": 1.0094386902773138,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004903909202724539,
      "loss": 5.0041,
      "mean_token_accuracy": 0.1973405286669731,
      "num_tokens": 22231995.0,
      "step": 12085
    },
    {
      "entropy": 5.1990556716918945,
      "epoch": 1.0098563314400266,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004903823113026602,
      "loss": 4.9759,
      "mean_token_accuracy": 0.21040778309106828,
      "num_tokens": 22240354.0,
      "step": 12090
    },
    {
      "entropy": 5.2165862083435055,
      "epoch": 1.0102739726027397,
      "grad_norm": 1.390625,
      "learning_rate": 0.00049037369856234,
      "loss": 5.0362,
      "mean_token_accuracy": 0.19550489634275436,
      "num_tokens": 22249866.0,
      "step": 12095
    },
    {
      "entropy": 5.243975257873535,
      "epoch": 1.0106916137654527,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004903650820516438,
      "loss": 5.0297,
      "mean_token_accuracy": 0.19310591667890548,
      "num_tokens": 22258837.0,
      "step": 12100
    },
    {
      "entropy": 5.196428728103638,
      "epoch": 1.0111092549281657,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004903564617707227,
      "loss": 4.9168,
      "mean_token_accuracy": 0.20485285073518752,
      "num_tokens": 22268035.0,
      "step": 12105
    },
    {
      "entropy": 5.109891319274903,
      "epoch": 1.0115268960908788,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004903478377197275,
      "loss": 4.8709,
      "mean_token_accuracy": 0.20405521094799042,
      "num_tokens": 22277251.0,
      "step": 12110
    },
    {
      "entropy": 5.001167345046997,
      "epoch": 1.0119445372535918,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004903392098988092,
      "loss": 4.8788,
      "mean_token_accuracy": 0.20342124700546266,
      "num_tokens": 22286269.0,
      "step": 12115
    },
    {
      "entropy": 5.145901727676391,
      "epoch": 1.0123621784163046,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004903305783081188,
      "loss": 4.9023,
      "mean_token_accuracy": 0.20229766219854356,
      "num_tokens": 22294671.0,
      "step": 12120
    },
    {
      "entropy": 5.104437685012817,
      "epoch": 1.0127798195790176,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004903219429478075,
      "loss": 4.9051,
      "mean_token_accuracy": 0.20748473405838014,
      "num_tokens": 22304448.0,
      "step": 12125
    },
    {
      "entropy": 5.224649906158447,
      "epoch": 1.0131974607417307,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004903133038180263,
      "loss": 5.0932,
      "mean_token_accuracy": 0.19456953555345535,
      "num_tokens": 22314425.0,
      "step": 12130
    },
    {
      "entropy": 5.1249021053314205,
      "epoch": 1.0136151019044437,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004903046609189266,
      "loss": 4.8811,
      "mean_token_accuracy": 0.20105402171611786,
      "num_tokens": 22324390.0,
      "step": 12135
    },
    {
      "entropy": 5.1608538150787355,
      "epoch": 1.0140327430671567,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004902960142506597,
      "loss": 4.9242,
      "mean_token_accuracy": 0.2045635387301445,
      "num_tokens": 22333274.0,
      "step": 12140
    },
    {
      "entropy": 5.1786257266998295,
      "epoch": 1.0144503842298698,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004902873638133768,
      "loss": 4.9867,
      "mean_token_accuracy": 0.20626837015151978,
      "num_tokens": 22342480.0,
      "step": 12145
    },
    {
      "entropy": 5.132463169097901,
      "epoch": 1.0148680253925826,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004902787096072296,
      "loss": 4.9355,
      "mean_token_accuracy": 0.2083053007721901,
      "num_tokens": 22351105.0,
      "step": 12150
    },
    {
      "entropy": 5.072145700454712,
      "epoch": 1.0152856665552956,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004902700516323693,
      "loss": 4.9254,
      "mean_token_accuracy": 0.21179743558168412,
      "num_tokens": 22360567.0,
      "step": 12155
    },
    {
      "entropy": 5.177220869064331,
      "epoch": 1.0157033077180087,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004902613898889477,
      "loss": 5.0,
      "mean_token_accuracy": 0.19947971999645234,
      "num_tokens": 22370140.0,
      "step": 12160
    },
    {
      "entropy": 5.127175760269165,
      "epoch": 1.0161209488807217,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004902527243771164,
      "loss": 4.7604,
      "mean_token_accuracy": 0.21180636435747147,
      "num_tokens": 22378676.0,
      "step": 12165
    },
    {
      "entropy": 5.150779676437378,
      "epoch": 1.0165385900434347,
      "grad_norm": 1.328125,
      "learning_rate": 0.000490244055097027,
      "loss": 4.9636,
      "mean_token_accuracy": 0.19429712295532225,
      "num_tokens": 22387411.0,
      "step": 12170
    },
    {
      "entropy": 5.177250909805298,
      "epoch": 1.0169562312061478,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004902353820488314,
      "loss": 5.0324,
      "mean_token_accuracy": 0.19497199803590776,
      "num_tokens": 22397023.0,
      "step": 12175
    },
    {
      "entropy": 5.129194545745849,
      "epoch": 1.0173738723688606,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004902267052326812,
      "loss": 4.9387,
      "mean_token_accuracy": 0.2021426558494568,
      "num_tokens": 22406730.0,
      "step": 12180
    },
    {
      "entropy": 5.274524354934693,
      "epoch": 1.0177915135315736,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004902180246487285,
      "loss": 5.0642,
      "mean_token_accuracy": 0.19656931459903718,
      "num_tokens": 22416920.0,
      "step": 12185
    },
    {
      "entropy": 5.166518926620483,
      "epoch": 1.0182091546942866,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004902093402971254,
      "loss": 4.9531,
      "mean_token_accuracy": 0.19672542363405227,
      "num_tokens": 22426073.0,
      "step": 12190
    },
    {
      "entropy": 5.256448125839233,
      "epoch": 1.0186267958569997,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004902006521780236,
      "loss": 5.0334,
      "mean_token_accuracy": 0.2026686891913414,
      "num_tokens": 22436291.0,
      "step": 12195
    },
    {
      "entropy": 5.068593072891235,
      "epoch": 1.0190444370197127,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004901919602915754,
      "loss": 4.9601,
      "mean_token_accuracy": 0.20481967628002168,
      "num_tokens": 22446053.0,
      "step": 12200
    },
    {
      "entropy": 5.178639125823975,
      "epoch": 1.0194620781824257,
      "grad_norm": 1.25,
      "learning_rate": 0.0004901832646379328,
      "loss": 4.9274,
      "mean_token_accuracy": 0.20121289789676666,
      "num_tokens": 22455474.0,
      "step": 12205
    },
    {
      "entropy": 5.2100988864898685,
      "epoch": 1.0198797193451385,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004901745652172483,
      "loss": 4.9383,
      "mean_token_accuracy": 0.20680596679449081,
      "num_tokens": 22464953.0,
      "step": 12210
    },
    {
      "entropy": 5.172916412353516,
      "epoch": 1.0202973605078516,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000490165862029674,
      "loss": 4.8983,
      "mean_token_accuracy": 0.1944560796022415,
      "num_tokens": 22474076.0,
      "step": 12215
    },
    {
      "entropy": 5.130919408798218,
      "epoch": 1.0207150016705646,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004901571550753622,
      "loss": 4.9585,
      "mean_token_accuracy": 0.201105098426342,
      "num_tokens": 22483128.0,
      "step": 12220
    },
    {
      "entropy": 5.211022424697876,
      "epoch": 1.0211326428332776,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004901484443544655,
      "loss": 5.055,
      "mean_token_accuracy": 0.20136988162994385,
      "num_tokens": 22492239.0,
      "step": 12225
    },
    {
      "entropy": 5.2796038627624515,
      "epoch": 1.0215502839959907,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004901397298671362,
      "loss": 5.1259,
      "mean_token_accuracy": 0.18717017769813538,
      "num_tokens": 22500996.0,
      "step": 12230
    },
    {
      "entropy": 5.226589727401733,
      "epoch": 1.0219679251587037,
      "grad_norm": 1.125,
      "learning_rate": 0.0004901310116135271,
      "loss": 4.9405,
      "mean_token_accuracy": 0.19812116622924805,
      "num_tokens": 22510524.0,
      "step": 12235
    },
    {
      "entropy": 5.1934960842132565,
      "epoch": 1.0223855663214167,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004901222895937905,
      "loss": 4.9661,
      "mean_token_accuracy": 0.2000878408551216,
      "num_tokens": 22519499.0,
      "step": 12240
    },
    {
      "entropy": 5.102998924255371,
      "epoch": 1.0228032074841296,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004901135638080796,
      "loss": 4.8471,
      "mean_token_accuracy": 0.2099151149392128,
      "num_tokens": 22528380.0,
      "step": 12245
    },
    {
      "entropy": 5.2012715339660645,
      "epoch": 1.0232208486468426,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004901048342565465,
      "loss": 5.0741,
      "mean_token_accuracy": 0.19413138777017594,
      "num_tokens": 22537488.0,
      "step": 12250
    },
    {
      "entropy": 5.20526533126831,
      "epoch": 1.0236384898095556,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004900961009393445,
      "loss": 5.0075,
      "mean_token_accuracy": 0.20385019332170487,
      "num_tokens": 22547093.0,
      "step": 12255
    },
    {
      "entropy": 5.104848146438599,
      "epoch": 1.0240561309722687,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004900873638566263,
      "loss": 4.8992,
      "mean_token_accuracy": 0.2035337969660759,
      "num_tokens": 22555996.0,
      "step": 12260
    },
    {
      "entropy": 5.170571279525757,
      "epoch": 1.0244737721349817,
      "grad_norm": 1.1875,
      "learning_rate": 0.000490078623008545,
      "loss": 4.9973,
      "mean_token_accuracy": 0.1936105325818062,
      "num_tokens": 22565582.0,
      "step": 12265
    },
    {
      "entropy": 5.220041990280151,
      "epoch": 1.0248914132976945,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004900698783952533,
      "loss": 5.0201,
      "mean_token_accuracy": 0.19505415558815004,
      "num_tokens": 22575181.0,
      "step": 12270
    },
    {
      "entropy": 5.267757701873779,
      "epoch": 1.0253090544604075,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004900611300169048,
      "loss": 5.0239,
      "mean_token_accuracy": 0.19922152310609817,
      "num_tokens": 22584799.0,
      "step": 12275
    },
    {
      "entropy": 5.139433813095093,
      "epoch": 1.0257266956231206,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004900523778736521,
      "loss": 4.8303,
      "mean_token_accuracy": 0.2111017256975174,
      "num_tokens": 22593235.0,
      "step": 12280
    },
    {
      "entropy": 5.120855760574341,
      "epoch": 1.0261443367858336,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004900436219656488,
      "loss": 5.0531,
      "mean_token_accuracy": 0.1971993088722229,
      "num_tokens": 22603435.0,
      "step": 12285
    },
    {
      "entropy": 5.1906991481781,
      "epoch": 1.0265619779485466,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000490034862293048,
      "loss": 5.0194,
      "mean_token_accuracy": 0.19752491116523743,
      "num_tokens": 22612739.0,
      "step": 12290
    },
    {
      "entropy": 5.1834887027740475,
      "epoch": 1.0269796191112597,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004900260988560031,
      "loss": 5.0314,
      "mean_token_accuracy": 0.19557403326034545,
      "num_tokens": 22622633.0,
      "step": 12295
    },
    {
      "entropy": 5.148898315429688,
      "epoch": 1.0273972602739727,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004900173316546675,
      "loss": 4.8792,
      "mean_token_accuracy": 0.20551477670669555,
      "num_tokens": 22631428.0,
      "step": 12300
    },
    {
      "entropy": 5.249334478378296,
      "epoch": 1.0278149014366855,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004900085606891947,
      "loss": 5.0629,
      "mean_token_accuracy": 0.20044250041246414,
      "num_tokens": 22639557.0,
      "step": 12305
    },
    {
      "entropy": 5.207519769668579,
      "epoch": 1.0282325425993986,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004899997859597382,
      "loss": 4.9854,
      "mean_token_accuracy": 0.20194839239120482,
      "num_tokens": 22648134.0,
      "step": 12310
    },
    {
      "entropy": 5.122716426849365,
      "epoch": 1.0286501837621116,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004899910074664518,
      "loss": 4.9464,
      "mean_token_accuracy": 0.20825906693935395,
      "num_tokens": 22657083.0,
      "step": 12315
    },
    {
      "entropy": 5.2278913974761965,
      "epoch": 1.0290678249248246,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004899822252094889,
      "loss": 5.0035,
      "mean_token_accuracy": 0.20374002158641816,
      "num_tokens": 22666724.0,
      "step": 12320
    },
    {
      "entropy": 5.184357690811157,
      "epoch": 1.0294854660875377,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004899734391890034,
      "loss": 5.0133,
      "mean_token_accuracy": 0.19929776936769486,
      "num_tokens": 22675955.0,
      "step": 12325
    },
    {
      "entropy": 5.184584522247315,
      "epoch": 1.0299031072502507,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004899646494051491,
      "loss": 4.9883,
      "mean_token_accuracy": 0.19981490075588226,
      "num_tokens": 22686838.0,
      "step": 12330
    },
    {
      "entropy": 5.269491672515869,
      "epoch": 1.0303207484129635,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004899558558580798,
      "loss": 5.0867,
      "mean_token_accuracy": 0.19477163702249528,
      "num_tokens": 22696257.0,
      "step": 12335
    },
    {
      "entropy": 5.096351575851441,
      "epoch": 1.0307383895756765,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004899470585479495,
      "loss": 4.9176,
      "mean_token_accuracy": 0.2018014132976532,
      "num_tokens": 22705726.0,
      "step": 12340
    },
    {
      "entropy": 5.156038856506347,
      "epoch": 1.0311560307383896,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004899382574749124,
      "loss": 4.9902,
      "mean_token_accuracy": 0.19875162541866304,
      "num_tokens": 22714765.0,
      "step": 12345
    },
    {
      "entropy": 5.237843799591064,
      "epoch": 1.0315736719011026,
      "grad_norm": 1.25,
      "learning_rate": 0.0004899294526391221,
      "loss": 4.9999,
      "mean_token_accuracy": 0.19588534384965897,
      "num_tokens": 22723423.0,
      "step": 12350
    },
    {
      "entropy": 5.195680475234985,
      "epoch": 1.0319913130638156,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004899206440407332,
      "loss": 4.9662,
      "mean_token_accuracy": 0.19669269323348998,
      "num_tokens": 22733276.0,
      "step": 12355
    },
    {
      "entropy": 5.1338584423065186,
      "epoch": 1.0324089542265287,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004899118316798996,
      "loss": 4.9623,
      "mean_token_accuracy": 0.20780497193336486,
      "num_tokens": 22742422.0,
      "step": 12360
    },
    {
      "entropy": 5.145673084259033,
      "epoch": 1.0328265953892415,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004899030155567759,
      "loss": 4.9209,
      "mean_token_accuracy": 0.19978038668632508,
      "num_tokens": 22751649.0,
      "step": 12365
    },
    {
      "entropy": 5.14819655418396,
      "epoch": 1.0332442365519545,
      "grad_norm": 1.1875,
      "learning_rate": 0.000489894195671516,
      "loss": 5.0577,
      "mean_token_accuracy": 0.19188131839036943,
      "num_tokens": 22760411.0,
      "step": 12370
    },
    {
      "entropy": 5.252345848083496,
      "epoch": 1.0336618777146676,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004898853720242747,
      "loss": 5.044,
      "mean_token_accuracy": 0.19762900173664094,
      "num_tokens": 22769289.0,
      "step": 12375
    },
    {
      "entropy": 5.2134363651275635,
      "epoch": 1.0340795188773806,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004898765446152062,
      "loss": 4.9581,
      "mean_token_accuracy": 0.20036045014858245,
      "num_tokens": 22778988.0,
      "step": 12380
    },
    {
      "entropy": 5.176703548431396,
      "epoch": 1.0344971600400936,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004898677134444652,
      "loss": 5.07,
      "mean_token_accuracy": 0.19614764750003816,
      "num_tokens": 22788115.0,
      "step": 12385
    },
    {
      "entropy": 5.244096040725708,
      "epoch": 1.0349148012028067,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004898588785122063,
      "loss": 5.045,
      "mean_token_accuracy": 0.20332542657852173,
      "num_tokens": 22797840.0,
      "step": 12390
    },
    {
      "entropy": 5.1602925777435305,
      "epoch": 1.0353324423655195,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004898500398185841,
      "loss": 4.9173,
      "mean_token_accuracy": 0.20884474366903305,
      "num_tokens": 22805947.0,
      "step": 12395
    },
    {
      "entropy": 5.168059778213501,
      "epoch": 1.0357500835282325,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004898411973637532,
      "loss": 4.9775,
      "mean_token_accuracy": 0.20047039836645125,
      "num_tokens": 22816104.0,
      "step": 12400
    },
    {
      "entropy": 5.200162935256958,
      "epoch": 1.0361677246909455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004898323511478686,
      "loss": 4.9294,
      "mean_token_accuracy": 0.20200901329517365,
      "num_tokens": 22824994.0,
      "step": 12405
    },
    {
      "entropy": 5.186477613449097,
      "epoch": 1.0365853658536586,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004898235011710851,
      "loss": 5.0972,
      "mean_token_accuracy": 0.19385089129209518,
      "num_tokens": 22834649.0,
      "step": 12410
    },
    {
      "entropy": 5.095633172988892,
      "epoch": 1.0370030070163716,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004898146474335577,
      "loss": 4.8704,
      "mean_token_accuracy": 0.21230559051036835,
      "num_tokens": 22843060.0,
      "step": 12415
    },
    {
      "entropy": 5.148401212692261,
      "epoch": 1.0374206481790846,
      "grad_norm": 1.25,
      "learning_rate": 0.0004898057899354412,
      "loss": 4.9672,
      "mean_token_accuracy": 0.20681700855493546,
      "num_tokens": 22852429.0,
      "step": 12420
    },
    {
      "entropy": 5.120256090164185,
      "epoch": 1.0378382893417974,
      "grad_norm": 1.25,
      "learning_rate": 0.000489796928676891,
      "loss": 4.9187,
      "mean_token_accuracy": 0.2054611250758171,
      "num_tokens": 22861170.0,
      "step": 12425
    },
    {
      "entropy": 5.110870981216431,
      "epoch": 1.0382559305045105,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004897880636580618,
      "loss": 4.9188,
      "mean_token_accuracy": 0.20782249122858049,
      "num_tokens": 22870030.0,
      "step": 12430
    },
    {
      "entropy": 5.111605072021485,
      "epoch": 1.0386735716672235,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000489779194879109,
      "loss": 4.9351,
      "mean_token_accuracy": 0.20994281023740768,
      "num_tokens": 22878728.0,
      "step": 12435
    },
    {
      "entropy": 5.214935398101806,
      "epoch": 1.0390912128299366,
      "grad_norm": 1.25,
      "learning_rate": 0.000489770322340188,
      "loss": 5.0358,
      "mean_token_accuracy": 0.19354919493198394,
      "num_tokens": 22887625.0,
      "step": 12440
    },
    {
      "entropy": 5.062684822082519,
      "epoch": 1.0395088539926496,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004897614460414539,
      "loss": 4.805,
      "mean_token_accuracy": 0.2110498443245888,
      "num_tokens": 22896791.0,
      "step": 12445
    },
    {
      "entropy": 5.134312391281128,
      "epoch": 1.0399264951553626,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004897525659830623,
      "loss": 4.9628,
      "mean_token_accuracy": 0.19939010292291642,
      "num_tokens": 22907228.0,
      "step": 12450
    },
    {
      "entropy": 5.0330335140228275,
      "epoch": 1.0403441363180754,
      "grad_norm": 1.25,
      "learning_rate": 0.0004897436821651683,
      "loss": 4.8197,
      "mean_token_accuracy": 0.21877533346414565,
      "num_tokens": 22915108.0,
      "step": 12455
    },
    {
      "entropy": 5.030098867416382,
      "epoch": 1.0407617774807885,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004897347945879278,
      "loss": 4.8653,
      "mean_token_accuracy": 0.21331406235694886,
      "num_tokens": 22924092.0,
      "step": 12460
    },
    {
      "entropy": 5.0984215259552,
      "epoch": 1.0411794186435015,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004897259032514963,
      "loss": 4.9083,
      "mean_token_accuracy": 0.20271498560905457,
      "num_tokens": 22932285.0,
      "step": 12465
    },
    {
      "entropy": 5.2102437019348145,
      "epoch": 1.0415970598062145,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004897170081560292,
      "loss": 4.9525,
      "mean_token_accuracy": 0.20671599060297013,
      "num_tokens": 22940727.0,
      "step": 12470
    },
    {
      "entropy": 5.205721664428711,
      "epoch": 1.0420147009689276,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004897081093016827,
      "loss": 5.0261,
      "mean_token_accuracy": 0.19765979200601577,
      "num_tokens": 22949678.0,
      "step": 12475
    },
    {
      "entropy": 5.155996799468994,
      "epoch": 1.0424323421316406,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004896992066886121,
      "loss": 5.0139,
      "mean_token_accuracy": 0.1960259646177292,
      "num_tokens": 22959295.0,
      "step": 12480
    },
    {
      "entropy": 5.179384565353393,
      "epoch": 1.0428499832943534,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004896903003169735,
      "loss": 4.9457,
      "mean_token_accuracy": 0.20502911508083344,
      "num_tokens": 22968883.0,
      "step": 12485
    },
    {
      "entropy": 5.201585149765014,
      "epoch": 1.0432676244570664,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004896813901869229,
      "loss": 5.0005,
      "mean_token_accuracy": 0.19676454067230226,
      "num_tokens": 22978199.0,
      "step": 12490
    },
    {
      "entropy": 5.1900026321411135,
      "epoch": 1.0436852656197795,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004896724762986159,
      "loss": 4.9785,
      "mean_token_accuracy": 0.204557004570961,
      "num_tokens": 22987236.0,
      "step": 12495
    },
    {
      "entropy": 5.165955305099487,
      "epoch": 1.0441029067824925,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004896635586522089,
      "loss": 4.928,
      "mean_token_accuracy": 0.20499594360589982,
      "num_tokens": 22996293.0,
      "step": 12500
    },
    {
      "entropy": 5.14666337966919,
      "epoch": 1.0445205479452055,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004896546372478581,
      "loss": 5.0343,
      "mean_token_accuracy": 0.2016330197453499,
      "num_tokens": 23006415.0,
      "step": 12505
    },
    {
      "entropy": 5.1992104053497314,
      "epoch": 1.0449381891079186,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004896457120857192,
      "loss": 5.0383,
      "mean_token_accuracy": 0.19809141457080842,
      "num_tokens": 23015819.0,
      "step": 12510
    },
    {
      "entropy": 5.17161922454834,
      "epoch": 1.0453558302706314,
      "grad_norm": 1.28125,
      "learning_rate": 0.000489636783165949,
      "loss": 4.8908,
      "mean_token_accuracy": 0.2102263629436493,
      "num_tokens": 23024206.0,
      "step": 12515
    },
    {
      "entropy": 5.228085517883301,
      "epoch": 1.0457734714333444,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004896278504887033,
      "loss": 5.0947,
      "mean_token_accuracy": 0.1936309427022934,
      "num_tokens": 23034003.0,
      "step": 12520
    },
    {
      "entropy": 5.21875147819519,
      "epoch": 1.0461911125960575,
      "grad_norm": 1.203125,
      "learning_rate": 0.000489618914054139,
      "loss": 5.0276,
      "mean_token_accuracy": 0.19882651120424272,
      "num_tokens": 23043514.0,
      "step": 12525
    },
    {
      "entropy": 5.1769670963287355,
      "epoch": 1.0466087537587705,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000489609973862412,
      "loss": 4.9531,
      "mean_token_accuracy": 0.19608525633811952,
      "num_tokens": 23053029.0,
      "step": 12530
    },
    {
      "entropy": 5.20466980934143,
      "epoch": 1.0470263949214835,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004896010299136793,
      "loss": 5.0187,
      "mean_token_accuracy": 0.1988625407218933,
      "num_tokens": 23062465.0,
      "step": 12535
    },
    {
      "entropy": 5.14296326637268,
      "epoch": 1.0474440360841966,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004895920822080972,
      "loss": 4.9222,
      "mean_token_accuracy": 0.2063967302441597,
      "num_tokens": 23071448.0,
      "step": 12540
    },
    {
      "entropy": 5.2012077331542965,
      "epoch": 1.0478616772469094,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004895831307458224,
      "loss": 5.0329,
      "mean_token_accuracy": 0.1977296620607376,
      "num_tokens": 23081882.0,
      "step": 12545
    },
    {
      "entropy": 5.236787366867065,
      "epoch": 1.0482793184096224,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004895741755270115,
      "loss": 5.0087,
      "mean_token_accuracy": 0.20333647727966309,
      "num_tokens": 23090723.0,
      "step": 12550
    },
    {
      "entropy": 5.0840507507324215,
      "epoch": 1.0486969595723354,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004895652165518215,
      "loss": 4.8531,
      "mean_token_accuracy": 0.21415795981884003,
      "num_tokens": 23099246.0,
      "step": 12555
    },
    {
      "entropy": 5.086436223983765,
      "epoch": 1.0491146007350485,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004895562538204091,
      "loss": 4.9332,
      "mean_token_accuracy": 0.20905129760503768,
      "num_tokens": 23107538.0,
      "step": 12560
    },
    {
      "entropy": 5.183541536331177,
      "epoch": 1.0495322418977615,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004895472873329311,
      "loss": 5.0162,
      "mean_token_accuracy": 0.20389313548803328,
      "num_tokens": 23116996.0,
      "step": 12565
    },
    {
      "entropy": 5.117693519592285,
      "epoch": 1.0499498830604745,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004895383170895447,
      "loss": 4.8888,
      "mean_token_accuracy": 0.2070747897028923,
      "num_tokens": 23126778.0,
      "step": 12570
    },
    {
      "entropy": 5.198449373245239,
      "epoch": 1.0503675242231874,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004895293430904067,
      "loss": 4.9951,
      "mean_token_accuracy": 0.20156582742929458,
      "num_tokens": 23135929.0,
      "step": 12575
    },
    {
      "entropy": 5.150644683837891,
      "epoch": 1.0507851653859004,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004895203653356745,
      "loss": 4.9307,
      "mean_token_accuracy": 0.20446724593639373,
      "num_tokens": 23144369.0,
      "step": 12580
    },
    {
      "entropy": 5.1575178623199465,
      "epoch": 1.0512028065486134,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004895113838255049,
      "loss": 4.9208,
      "mean_token_accuracy": 0.20366470664739608,
      "num_tokens": 23152813.0,
      "step": 12585
    },
    {
      "entropy": 5.131381559371948,
      "epoch": 1.0516204477113265,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004895023985600555,
      "loss": 4.9579,
      "mean_token_accuracy": 0.19924381673336028,
      "num_tokens": 23161826.0,
      "step": 12590
    },
    {
      "entropy": 5.1752588748931885,
      "epoch": 1.0520380888740395,
      "grad_norm": 1.375,
      "learning_rate": 0.0004894934095394832,
      "loss": 4.9146,
      "mean_token_accuracy": 0.20176558792591096,
      "num_tokens": 23170757.0,
      "step": 12595
    },
    {
      "entropy": 5.201378583908081,
      "epoch": 1.0524557300367525,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004894844167639458,
      "loss": 4.9779,
      "mean_token_accuracy": 0.1987537831068039,
      "num_tokens": 23179348.0,
      "step": 12600
    },
    {
      "entropy": 5.161959028244018,
      "epoch": 1.0528733711994653,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004894754202336004,
      "loss": 5.0132,
      "mean_token_accuracy": 0.1949801579117775,
      "num_tokens": 23188643.0,
      "step": 12605
    },
    {
      "entropy": 5.033377742767334,
      "epoch": 1.0532910123621784,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004894664199486047,
      "loss": 4.8965,
      "mean_token_accuracy": 0.19819177240133284,
      "num_tokens": 23197595.0,
      "step": 12610
    },
    {
      "entropy": 5.175791883468628,
      "epoch": 1.0537086535248914,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004894574159091161,
      "loss": 4.898,
      "mean_token_accuracy": 0.20845069587230683,
      "num_tokens": 23206463.0,
      "step": 12615
    },
    {
      "entropy": 5.228186511993409,
      "epoch": 1.0541262946876044,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004894484081152922,
      "loss": 5.0376,
      "mean_token_accuracy": 0.19057111740112304,
      "num_tokens": 23215740.0,
      "step": 12620
    },
    {
      "entropy": 5.191231679916382,
      "epoch": 1.0545439358503175,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004894393965672909,
      "loss": 5.0008,
      "mean_token_accuracy": 0.19950238466262818,
      "num_tokens": 23226046.0,
      "step": 12625
    },
    {
      "entropy": 5.2268517971038815,
      "epoch": 1.0549615770130305,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004894303812652698,
      "loss": 4.9995,
      "mean_token_accuracy": 0.19355977326631546,
      "num_tokens": 23234905.0,
      "step": 12630
    },
    {
      "entropy": 5.0929797172546385,
      "epoch": 1.0553792181757433,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004894213622093867,
      "loss": 4.8724,
      "mean_token_accuracy": 0.20581526756286622,
      "num_tokens": 23243392.0,
      "step": 12635
    },
    {
      "entropy": 5.1611052513122555,
      "epoch": 1.0557968593384564,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004894123393997997,
      "loss": 4.927,
      "mean_token_accuracy": 0.20097920447587966,
      "num_tokens": 23253276.0,
      "step": 12640
    },
    {
      "entropy": 5.077817010879516,
      "epoch": 1.0562145005011694,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004894033128366665,
      "loss": 4.8916,
      "mean_token_accuracy": 0.20813636034727095,
      "num_tokens": 23262336.0,
      "step": 12645
    },
    {
      "entropy": 5.143207693099976,
      "epoch": 1.0566321416638824,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004893942825201453,
      "loss": 5.0021,
      "mean_token_accuracy": 0.199725379049778,
      "num_tokens": 23272103.0,
      "step": 12650
    },
    {
      "entropy": 5.245752000808716,
      "epoch": 1.0570497828265955,
      "grad_norm": 1.140625,
      "learning_rate": 0.000489385248450394,
      "loss": 5.0615,
      "mean_token_accuracy": 0.2002743437886238,
      "num_tokens": 23281253.0,
      "step": 12655
    },
    {
      "entropy": 5.2179036140441895,
      "epoch": 1.0574674239893085,
      "grad_norm": 1.25,
      "learning_rate": 0.000489376210627571,
      "loss": 4.972,
      "mean_token_accuracy": 0.20117961317300798,
      "num_tokens": 23289920.0,
      "step": 12660
    },
    {
      "entropy": 5.1518162250518795,
      "epoch": 1.0578850651520213,
      "grad_norm": 1.25,
      "learning_rate": 0.0004893671690518343,
      "loss": 5.0275,
      "mean_token_accuracy": 0.19979559779167175,
      "num_tokens": 23299052.0,
      "step": 12665
    },
    {
      "entropy": 5.176765060424804,
      "epoch": 1.0583027063147343,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004893581237233423,
      "loss": 5.0397,
      "mean_token_accuracy": 0.20392135232686998,
      "num_tokens": 23307782.0,
      "step": 12670
    },
    {
      "entropy": 5.243820905685425,
      "epoch": 1.0587203474774474,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004893490746422533,
      "loss": 5.0096,
      "mean_token_accuracy": 0.2029566064476967,
      "num_tokens": 23317302.0,
      "step": 12675
    },
    {
      "entropy": 5.147712993621826,
      "epoch": 1.0591379886401604,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004893400218087258,
      "loss": 4.8902,
      "mean_token_accuracy": 0.21001974046230315,
      "num_tokens": 23325899.0,
      "step": 12680
    },
    {
      "entropy": 5.19874677658081,
      "epoch": 1.0595556298028734,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004893309652229183,
      "loss": 4.9753,
      "mean_token_accuracy": 0.20388401448726653,
      "num_tokens": 23335121.0,
      "step": 12685
    },
    {
      "entropy": 5.180070447921753,
      "epoch": 1.0599732709655865,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004893219048849891,
      "loss": 5.0237,
      "mean_token_accuracy": 0.1953381896018982,
      "num_tokens": 23345377.0,
      "step": 12690
    },
    {
      "entropy": 5.182038450241089,
      "epoch": 1.0603909121282993,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004893128407950971,
      "loss": 5.0178,
      "mean_token_accuracy": 0.19701989293098449,
      "num_tokens": 23353963.0,
      "step": 12695
    },
    {
      "entropy": 5.135223150253296,
      "epoch": 1.0608085532910123,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004893037729534007,
      "loss": 4.9136,
      "mean_token_accuracy": 0.21136902570724486,
      "num_tokens": 23363744.0,
      "step": 12700
    },
    {
      "entropy": 5.162247896194458,
      "epoch": 1.0612261944537253,
      "grad_norm": 1.25,
      "learning_rate": 0.0004892947013600589,
      "loss": 4.9404,
      "mean_token_accuracy": 0.20126497596502305,
      "num_tokens": 23372241.0,
      "step": 12705
    },
    {
      "entropy": 5.126316261291504,
      "epoch": 1.0616438356164384,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004892856260152304,
      "loss": 4.9482,
      "mean_token_accuracy": 0.20426203459501266,
      "num_tokens": 23380733.0,
      "step": 12710
    },
    {
      "entropy": 5.222818422317505,
      "epoch": 1.0620614767791514,
      "grad_norm": 1.265625,
      "learning_rate": 0.000489276546919074,
      "loss": 5.0219,
      "mean_token_accuracy": 0.1943354457616806,
      "num_tokens": 23389916.0,
      "step": 12715
    },
    {
      "entropy": 5.143852233886719,
      "epoch": 1.0624791179418644,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004892674640717488,
      "loss": 4.9281,
      "mean_token_accuracy": 0.20567619502544404,
      "num_tokens": 23398394.0,
      "step": 12720
    },
    {
      "entropy": 5.2402184963226315,
      "epoch": 1.0628967591045773,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004892583774734138,
      "loss": 5.078,
      "mean_token_accuracy": 0.19064523577690123,
      "num_tokens": 23408294.0,
      "step": 12725
    },
    {
      "entropy": 5.154877853393555,
      "epoch": 1.0633144002672903,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004892492871242278,
      "loss": 5.0173,
      "mean_token_accuracy": 0.19271493703126907,
      "num_tokens": 23417301.0,
      "step": 12730
    },
    {
      "entropy": 5.185716199874878,
      "epoch": 1.0637320414300033,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004892401930243504,
      "loss": 4.9789,
      "mean_token_accuracy": 0.2042860820889473,
      "num_tokens": 23426992.0,
      "step": 12735
    },
    {
      "entropy": 5.145590209960938,
      "epoch": 1.0641496825927164,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004892310951739404,
      "loss": 4.9459,
      "mean_token_accuracy": 0.2076074466109276,
      "num_tokens": 23435480.0,
      "step": 12740
    },
    {
      "entropy": 5.196811962127685,
      "epoch": 1.0645673237554294,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004892219935731571,
      "loss": 5.0442,
      "mean_token_accuracy": 0.19965091794729234,
      "num_tokens": 23444436.0,
      "step": 12745
    },
    {
      "entropy": 5.114441728591919,
      "epoch": 1.0649849649181424,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048921288822216,
      "loss": 4.8846,
      "mean_token_accuracy": 0.20941781401634216,
      "num_tokens": 23452913.0,
      "step": 12750
    },
    {
      "entropy": 5.035728788375854,
      "epoch": 1.0654026060808552,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004892037791211086,
      "loss": 4.927,
      "mean_token_accuracy": 0.2055075079202652,
      "num_tokens": 23462078.0,
      "step": 12755
    },
    {
      "entropy": 5.134254360198975,
      "epoch": 1.0658202472435683,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004891946662701621,
      "loss": 4.911,
      "mean_token_accuracy": 0.206013286113739,
      "num_tokens": 23470224.0,
      "step": 12760
    },
    {
      "entropy": 5.126784563064575,
      "epoch": 1.0662378884062813,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004891855496694801,
      "loss": 4.9499,
      "mean_token_accuracy": 0.1985659956932068,
      "num_tokens": 23478820.0,
      "step": 12765
    },
    {
      "entropy": 5.206686544418335,
      "epoch": 1.0666555295689943,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004891764293192222,
      "loss": 5.0392,
      "mean_token_accuracy": 0.1961066633462906,
      "num_tokens": 23488396.0,
      "step": 12770
    },
    {
      "entropy": 5.253154039382935,
      "epoch": 1.0670731707317074,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004891673052195482,
      "loss": 5.0355,
      "mean_token_accuracy": 0.19774702787399293,
      "num_tokens": 23497778.0,
      "step": 12775
    },
    {
      "entropy": 5.296729707717896,
      "epoch": 1.0674908118944204,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004891581773706177,
      "loss": 5.1552,
      "mean_token_accuracy": 0.1900760069489479,
      "num_tokens": 23506637.0,
      "step": 12780
    },
    {
      "entropy": 5.218043994903565,
      "epoch": 1.0679084530571332,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004891490457725906,
      "loss": 5.0379,
      "mean_token_accuracy": 0.19942990094423294,
      "num_tokens": 23515830.0,
      "step": 12785
    },
    {
      "entropy": 5.1818756580352785,
      "epoch": 1.0683260942198463,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004891399104256266,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20411134958267213,
      "num_tokens": 23525646.0,
      "step": 12790
    },
    {
      "entropy": 5.249075269699096,
      "epoch": 1.0687437353825593,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004891307713298857,
      "loss": 5.0111,
      "mean_token_accuracy": 0.19879214763641356,
      "num_tokens": 23535319.0,
      "step": 12795
    },
    {
      "entropy": 5.150713062286377,
      "epoch": 1.0691613765452723,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000489121628485528,
      "loss": 5.0117,
      "mean_token_accuracy": 0.1978086844086647,
      "num_tokens": 23544275.0,
      "step": 12800
    },
    {
      "entropy": 5.141674470901489,
      "epoch": 1.0695790177079854,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004891124818927133,
      "loss": 4.9421,
      "mean_token_accuracy": 0.20023730844259263,
      "num_tokens": 23553782.0,
      "step": 12805
    },
    {
      "entropy": 5.100135612487793,
      "epoch": 1.0699966588706984,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004891033315516019,
      "loss": 4.8026,
      "mean_token_accuracy": 0.21267443299293518,
      "num_tokens": 23562718.0,
      "step": 12810
    },
    {
      "entropy": 5.17625880241394,
      "epoch": 1.0704143000334112,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004890941774623541,
      "loss": 4.9943,
      "mean_token_accuracy": 0.1976195678114891,
      "num_tokens": 23572374.0,
      "step": 12815
    },
    {
      "entropy": 5.121626281738282,
      "epoch": 1.0708319411961242,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004890850196251299,
      "loss": 4.8892,
      "mean_token_accuracy": 0.20726636946201324,
      "num_tokens": 23581227.0,
      "step": 12820
    },
    {
      "entropy": 5.109693765640259,
      "epoch": 1.0712495823588373,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004890758580400898,
      "loss": 4.9082,
      "mean_token_accuracy": 0.2020815595984459,
      "num_tokens": 23591120.0,
      "step": 12825
    },
    {
      "entropy": 5.123488092422486,
      "epoch": 1.0716672235215503,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004890666927073941,
      "loss": 4.9118,
      "mean_token_accuracy": 0.200987546145916,
      "num_tokens": 23599584.0,
      "step": 12830
    },
    {
      "entropy": 5.171791982650757,
      "epoch": 1.0720848646842633,
      "grad_norm": 1.25,
      "learning_rate": 0.0004890575236272032,
      "loss": 4.9857,
      "mean_token_accuracy": 0.1995101511478424,
      "num_tokens": 23609479.0,
      "step": 12835
    },
    {
      "entropy": 5.181425142288208,
      "epoch": 1.0725025058469764,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004890483507996778,
      "loss": 4.9531,
      "mean_token_accuracy": 0.2066161513328552,
      "num_tokens": 23618662.0,
      "step": 12840
    },
    {
      "entropy": 5.2633469581604,
      "epoch": 1.0729201470096892,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004890391742249783,
      "loss": 5.0683,
      "mean_token_accuracy": 0.19162315130233765,
      "num_tokens": 23628268.0,
      "step": 12845
    },
    {
      "entropy": 5.1647422313690186,
      "epoch": 1.0733377881724022,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004890299939032654,
      "loss": 4.9685,
      "mean_token_accuracy": 0.20322999358177185,
      "num_tokens": 23637538.0,
      "step": 12850
    },
    {
      "entropy": 5.1583905696868895,
      "epoch": 1.0737554293351153,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004890208098346999,
      "loss": 5.093,
      "mean_token_accuracy": 0.19112809300422667,
      "num_tokens": 23647425.0,
      "step": 12855
    },
    {
      "entropy": 5.262768936157227,
      "epoch": 1.0741730704978283,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004890116220194425,
      "loss": 5.1012,
      "mean_token_accuracy": 0.19417205154895784,
      "num_tokens": 23656310.0,
      "step": 12860
    },
    {
      "entropy": 5.192828512191772,
      "epoch": 1.0745907116605413,
      "grad_norm": 1.15625,
      "learning_rate": 0.000489002430457654,
      "loss": 4.9892,
      "mean_token_accuracy": 0.20072500109672547,
      "num_tokens": 23664658.0,
      "step": 12865
    },
    {
      "entropy": 5.248310613632202,
      "epoch": 1.0750083528232544,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004889932351494954,
      "loss": 5.0072,
      "mean_token_accuracy": 0.2046247497200966,
      "num_tokens": 23674396.0,
      "step": 12870
    },
    {
      "entropy": 5.221344089508056,
      "epoch": 1.0754259939859672,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004889840360951277,
      "loss": 4.9238,
      "mean_token_accuracy": 0.20872651189565658,
      "num_tokens": 23682549.0,
      "step": 12875
    },
    {
      "entropy": 5.135237121582032,
      "epoch": 1.0758436351486802,
      "grad_norm": 1.375,
      "learning_rate": 0.0004889748332947118,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20462993681430816,
      "num_tokens": 23692334.0,
      "step": 12880
    },
    {
      "entropy": 5.170808744430542,
      "epoch": 1.0762612763113932,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004889656267484089,
      "loss": 4.9728,
      "mean_token_accuracy": 0.2003317207098007,
      "num_tokens": 23702081.0,
      "step": 12885
    },
    {
      "entropy": 5.2288188457489015,
      "epoch": 1.0766789174741063,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048895641645638,
      "loss": 5.0699,
      "mean_token_accuracy": 0.19660116881132125,
      "num_tokens": 23711019.0,
      "step": 12890
    },
    {
      "entropy": 5.180233001708984,
      "epoch": 1.0770965586368193,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004889472024187867,
      "loss": 4.945,
      "mean_token_accuracy": 0.20637753158807753,
      "num_tokens": 23719660.0,
      "step": 12895
    },
    {
      "entropy": 5.1542112827301025,
      "epoch": 1.0775141997995323,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048893798463579,
      "loss": 5.0071,
      "mean_token_accuracy": 0.19735390841960906,
      "num_tokens": 23728937.0,
      "step": 12900
    },
    {
      "entropy": 5.174329710006714,
      "epoch": 1.0779318409622451,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004889287631075513,
      "loss": 4.974,
      "mean_token_accuracy": 0.20844412595033646,
      "num_tokens": 23738530.0,
      "step": 12905
    },
    {
      "entropy": 5.195611524581909,
      "epoch": 1.0783494821249582,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004889195378342322,
      "loss": 5.0016,
      "mean_token_accuracy": 0.2000334620475769,
      "num_tokens": 23748285.0,
      "step": 12910
    },
    {
      "entropy": 5.125119829177857,
      "epoch": 1.0787671232876712,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004889103088159941,
      "loss": 4.9546,
      "mean_token_accuracy": 0.2008134201169014,
      "num_tokens": 23758312.0,
      "step": 12915
    },
    {
      "entropy": 5.254036998748779,
      "epoch": 1.0791847644503842,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004889010760529986,
      "loss": 5.0671,
      "mean_token_accuracy": 0.1980193793773651,
      "num_tokens": 23767092.0,
      "step": 12920
    },
    {
      "entropy": 5.326939868927002,
      "epoch": 1.0796024056130973,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004888918395454071,
      "loss": 5.0443,
      "mean_token_accuracy": 0.19739038795232772,
      "num_tokens": 23776257.0,
      "step": 12925
    },
    {
      "entropy": 5.249224376678467,
      "epoch": 1.0800200467758103,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004888825992933817,
      "loss": 5.0439,
      "mean_token_accuracy": 0.19262923896312714,
      "num_tokens": 23785311.0,
      "step": 12930
    },
    {
      "entropy": 5.11654953956604,
      "epoch": 1.0804376879385231,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004888733552970839,
      "loss": 4.9714,
      "mean_token_accuracy": 0.1992383360862732,
      "num_tokens": 23795050.0,
      "step": 12935
    },
    {
      "entropy": 5.214238834381104,
      "epoch": 1.0808553291012362,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004888641075566755,
      "loss": 5.0611,
      "mean_token_accuracy": 0.1990690752863884,
      "num_tokens": 23803959.0,
      "step": 12940
    },
    {
      "entropy": 5.201442718505859,
      "epoch": 1.0812729702639492,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004888548560723184,
      "loss": 4.9135,
      "mean_token_accuracy": 0.20271057933568953,
      "num_tokens": 23813568.0,
      "step": 12945
    },
    {
      "entropy": 5.178594398498535,
      "epoch": 1.0816906114266622,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004888456008441748,
      "loss": 5.0264,
      "mean_token_accuracy": 0.19284995049238204,
      "num_tokens": 23823076.0,
      "step": 12950
    },
    {
      "entropy": 5.07160267829895,
      "epoch": 1.0821082525893753,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004888363418724064,
      "loss": 4.8474,
      "mean_token_accuracy": 0.21805896162986754,
      "num_tokens": 23831624.0,
      "step": 12955
    },
    {
      "entropy": 5.186234712600708,
      "epoch": 1.0825258937520883,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004888270791571756,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20242290943861008,
      "num_tokens": 23841054.0,
      "step": 12960
    },
    {
      "entropy": 5.163810729980469,
      "epoch": 1.082943534914801,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004888178126986444,
      "loss": 5.0139,
      "mean_token_accuracy": 0.1979166552424431,
      "num_tokens": 23851190.0,
      "step": 12965
    },
    {
      "entropy": 5.181267261505127,
      "epoch": 1.0833611760775141,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004888085424969748,
      "loss": 5.0125,
      "mean_token_accuracy": 0.21049170941114426,
      "num_tokens": 23859560.0,
      "step": 12970
    },
    {
      "entropy": 5.210985898971558,
      "epoch": 1.0837788172402272,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004887992685523295,
      "loss": 4.9815,
      "mean_token_accuracy": 0.20190583020448685,
      "num_tokens": 23869103.0,
      "step": 12975
    },
    {
      "entropy": 5.162423515319825,
      "epoch": 1.0841964584029402,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004887899908648706,
      "loss": 4.9309,
      "mean_token_accuracy": 0.20793181359767915,
      "num_tokens": 23877641.0,
      "step": 12980
    },
    {
      "entropy": 5.164359426498413,
      "epoch": 1.0846140995656532,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004887807094347606,
      "loss": 5.0196,
      "mean_token_accuracy": 0.20215266793966294,
      "num_tokens": 23886671.0,
      "step": 12985
    },
    {
      "entropy": 5.119530200958252,
      "epoch": 1.0850317407283663,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004887714242621619,
      "loss": 4.9298,
      "mean_token_accuracy": 0.2109586700797081,
      "num_tokens": 23895781.0,
      "step": 12990
    },
    {
      "entropy": 5.2624927997589115,
      "epoch": 1.085449381891079,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004887621353472371,
      "loss": 5.0697,
      "mean_token_accuracy": 0.1970836952328682,
      "num_tokens": 23904478.0,
      "step": 12995
    },
    {
      "entropy": 5.242197561264038,
      "epoch": 1.0858670230537921,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004887528426901488,
      "loss": 4.9739,
      "mean_token_accuracy": 0.19853319078683854,
      "num_tokens": 23913899.0,
      "step": 13000
    },
    {
      "entropy": 5.141314172744751,
      "epoch": 1.0862846642165052,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004887435462910597,
      "loss": 5.0512,
      "mean_token_accuracy": 0.20020175278186797,
      "num_tokens": 23923583.0,
      "step": 13005
    },
    {
      "entropy": 5.197400522232056,
      "epoch": 1.0867023053792182,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004887342461501326,
      "loss": 5.0075,
      "mean_token_accuracy": 0.1925540715456009,
      "num_tokens": 23932392.0,
      "step": 13010
    },
    {
      "entropy": 5.326419448852539,
      "epoch": 1.0871199465419312,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004887249422675301,
      "loss": 5.1,
      "mean_token_accuracy": 0.1848290041089058,
      "num_tokens": 23941480.0,
      "step": 13015
    },
    {
      "entropy": 5.172903347015381,
      "epoch": 1.0875375877046443,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004887156346434154,
      "loss": 5.0,
      "mean_token_accuracy": 0.20673845261335372,
      "num_tokens": 23950720.0,
      "step": 13020
    },
    {
      "entropy": 5.141679382324218,
      "epoch": 1.087955228867357,
      "grad_norm": 1.25,
      "learning_rate": 0.0004887063232779513,
      "loss": 4.9031,
      "mean_token_accuracy": 0.20663689821958542,
      "num_tokens": 23959441.0,
      "step": 13025
    },
    {
      "entropy": 5.206944608688355,
      "epoch": 1.08837287003007,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004886970081713006,
      "loss": 4.9439,
      "mean_token_accuracy": 0.20231774151325227,
      "num_tokens": 23968638.0,
      "step": 13030
    },
    {
      "entropy": 5.169567012786866,
      "epoch": 1.0887905111927831,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004886876893236266,
      "loss": 4.9977,
      "mean_token_accuracy": 0.19726155549287797,
      "num_tokens": 23977820.0,
      "step": 13035
    },
    {
      "entropy": 5.104059791564941,
      "epoch": 1.0892081523554962,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004886783667350925,
      "loss": 4.8743,
      "mean_token_accuracy": 0.21075446605682374,
      "num_tokens": 23986674.0,
      "step": 13040
    },
    {
      "entropy": 5.178409242630005,
      "epoch": 1.0896257935182092,
      "grad_norm": 1.125,
      "learning_rate": 0.0004886690404058613,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20530558824539186,
      "num_tokens": 23995647.0,
      "step": 13045
    },
    {
      "entropy": 5.219105625152588,
      "epoch": 1.0900434346809222,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004886597103360963,
      "loss": 4.9407,
      "mean_token_accuracy": 0.20715427100658418,
      "num_tokens": 24004440.0,
      "step": 13050
    },
    {
      "entropy": 5.24730863571167,
      "epoch": 1.090461075843635,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004886503765259609,
      "loss": 5.0722,
      "mean_token_accuracy": 0.19461475312709808,
      "num_tokens": 24014632.0,
      "step": 13055
    },
    {
      "entropy": 5.158810997009278,
      "epoch": 1.090878717006348,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004886410389756186,
      "loss": 4.9049,
      "mean_token_accuracy": 0.20974108576774597,
      "num_tokens": 24024096.0,
      "step": 13060
    },
    {
      "entropy": 5.034177207946778,
      "epoch": 1.0912963581690611,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004886316976852326,
      "loss": 4.943,
      "mean_token_accuracy": 0.20606043934822083,
      "num_tokens": 24033410.0,
      "step": 13065
    },
    {
      "entropy": 5.175591087341308,
      "epoch": 1.0917139993317742,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004886223526549666,
      "loss": 4.9942,
      "mean_token_accuracy": 0.20546575784683227,
      "num_tokens": 24043110.0,
      "step": 13070
    },
    {
      "entropy": 5.19419264793396,
      "epoch": 1.0921316404944872,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004886130038849843,
      "loss": 4.9846,
      "mean_token_accuracy": 0.20609234273433685,
      "num_tokens": 24052393.0,
      "step": 13075
    },
    {
      "entropy": 5.232043170928955,
      "epoch": 1.0925492816572002,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004886036513754491,
      "loss": 5.0362,
      "mean_token_accuracy": 0.1979564294219017,
      "num_tokens": 24061086.0,
      "step": 13080
    },
    {
      "entropy": 5.184473705291748,
      "epoch": 1.092966922819913,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004885942951265251,
      "loss": 5.0194,
      "mean_token_accuracy": 0.2031309112906456,
      "num_tokens": 24070494.0,
      "step": 13085
    },
    {
      "entropy": 5.156550788879395,
      "epoch": 1.093384563982626,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004885849351383756,
      "loss": 4.8616,
      "mean_token_accuracy": 0.2136886090040207,
      "num_tokens": 24080740.0,
      "step": 13090
    },
    {
      "entropy": 5.12438645362854,
      "epoch": 1.093802205145339,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004885755714111649,
      "loss": 4.9262,
      "mean_token_accuracy": 0.19929886609315872,
      "num_tokens": 24090372.0,
      "step": 13095
    },
    {
      "entropy": 5.1872323513031,
      "epoch": 1.0942198463080521,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004885662039450565,
      "loss": 4.9925,
      "mean_token_accuracy": 0.19762236922979354,
      "num_tokens": 24099294.0,
      "step": 13100
    },
    {
      "entropy": 5.19448676109314,
      "epoch": 1.0946374874707652,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004885568327402149,
      "loss": 4.9533,
      "mean_token_accuracy": 0.20549339205026626,
      "num_tokens": 24108066.0,
      "step": 13105
    },
    {
      "entropy": 5.197881031036377,
      "epoch": 1.0950551286334782,
      "grad_norm": 1.125,
      "learning_rate": 0.0004885474577968038,
      "loss": 5.0869,
      "mean_token_accuracy": 0.19202090948820114,
      "num_tokens": 24118070.0,
      "step": 13110
    },
    {
      "entropy": 5.273807048797607,
      "epoch": 1.095472769796191,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004885380791149874,
      "loss": 5.0325,
      "mean_token_accuracy": 0.20345836281776428,
      "num_tokens": 24127314.0,
      "step": 13115
    },
    {
      "entropy": 5.201112556457519,
      "epoch": 1.095890410958904,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004885286966949298,
      "loss": 5.0319,
      "mean_token_accuracy": 0.20222124308347703,
      "num_tokens": 24136440.0,
      "step": 13120
    },
    {
      "entropy": 5.143569278717041,
      "epoch": 1.096308052121617,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004885193105367955,
      "loss": 5.0539,
      "mean_token_accuracy": 0.19882403761148454,
      "num_tokens": 24146268.0,
      "step": 13125
    },
    {
      "entropy": 5.1855854988098145,
      "epoch": 1.0967256932843301,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004885099206407486,
      "loss": 4.9441,
      "mean_token_accuracy": 0.20687180757522583,
      "num_tokens": 24156034.0,
      "step": 13130
    },
    {
      "entropy": 5.233466005325317,
      "epoch": 1.0971433344470432,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004885005270069535,
      "loss": 4.9679,
      "mean_token_accuracy": 0.2024045169353485,
      "num_tokens": 24165001.0,
      "step": 13135
    },
    {
      "entropy": 5.130460453033447,
      "epoch": 1.0975609756097562,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004884911296355746,
      "loss": 4.9053,
      "mean_token_accuracy": 0.2046682208776474,
      "num_tokens": 24174490.0,
      "step": 13140
    },
    {
      "entropy": 5.190751647949218,
      "epoch": 1.097978616772469,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004884817285267766,
      "loss": 5.0345,
      "mean_token_accuracy": 0.19710303395986556,
      "num_tokens": 24184471.0,
      "step": 13145
    },
    {
      "entropy": 5.20024003982544,
      "epoch": 1.098396257935182,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004884723236807239,
      "loss": 4.8934,
      "mean_token_accuracy": 0.20126610845327378,
      "num_tokens": 24193219.0,
      "step": 13150
    },
    {
      "entropy": 5.22159161567688,
      "epoch": 1.098813899097895,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004884629150975814,
      "loss": 5.0232,
      "mean_token_accuracy": 0.19876063764095306,
      "num_tokens": 24202138.0,
      "step": 13155
    },
    {
      "entropy": 5.132899808883667,
      "epoch": 1.099231540260608,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004884535027775135,
      "loss": 4.9457,
      "mean_token_accuracy": 0.1982187569141388,
      "num_tokens": 24211689.0,
      "step": 13160
    },
    {
      "entropy": 5.085301733016967,
      "epoch": 1.0996491814233211,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004884440867206851,
      "loss": 4.9129,
      "mean_token_accuracy": 0.2032475396990776,
      "num_tokens": 24221822.0,
      "step": 13165
    },
    {
      "entropy": 5.212867975234985,
      "epoch": 1.1000668225860342,
      "grad_norm": 1.203125,
      "learning_rate": 0.000488434666927261,
      "loss": 5.0589,
      "mean_token_accuracy": 0.19549386501312255,
      "num_tokens": 24231504.0,
      "step": 13170
    },
    {
      "entropy": 5.1744537353515625,
      "epoch": 1.100484463748747,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004884252433974062,
      "loss": 4.9641,
      "mean_token_accuracy": 0.19761509597301483,
      "num_tokens": 24240855.0,
      "step": 13175
    },
    {
      "entropy": 5.13556022644043,
      "epoch": 1.10090210491146,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004884158161312856,
      "loss": 4.8875,
      "mean_token_accuracy": 0.21134347319602967,
      "num_tokens": 24249659.0,
      "step": 13180
    },
    {
      "entropy": 5.113099575042725,
      "epoch": 1.101319746074173,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004884063851290642,
      "loss": 4.9746,
      "mean_token_accuracy": 0.20077899396419524,
      "num_tokens": 24259277.0,
      "step": 13185
    },
    {
      "entropy": 5.164052152633667,
      "epoch": 1.101737387236886,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004883969503909073,
      "loss": 5.0051,
      "mean_token_accuracy": 0.2017353042960167,
      "num_tokens": 24269744.0,
      "step": 13190
    },
    {
      "entropy": 5.128753232955932,
      "epoch": 1.1021550283995991,
      "grad_norm": 1.25,
      "learning_rate": 0.0004883875119169798,
      "loss": 4.9181,
      "mean_token_accuracy": 0.20521418154239654,
      "num_tokens": 24278843.0,
      "step": 13195
    },
    {
      "entropy": 5.129987287521362,
      "epoch": 1.1025726695623121,
      "grad_norm": 1.1875,
      "learning_rate": 0.000488378069707447,
      "loss": 4.8459,
      "mean_token_accuracy": 0.21534471362829208,
      "num_tokens": 24287586.0,
      "step": 13200
    },
    {
      "entropy": 5.127574729919433,
      "epoch": 1.102990310725025,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004883686237624744,
      "loss": 4.9565,
      "mean_token_accuracy": 0.20345209687948226,
      "num_tokens": 24296318.0,
      "step": 13205
    },
    {
      "entropy": 5.142273044586181,
      "epoch": 1.103407951887738,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004883591740822271,
      "loss": 4.9469,
      "mean_token_accuracy": 0.20400548428297044,
      "num_tokens": 24305901.0,
      "step": 13210
    },
    {
      "entropy": 5.177602243423462,
      "epoch": 1.103825593050451,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004883497206668707,
      "loss": 4.9236,
      "mean_token_accuracy": 0.2033698320388794,
      "num_tokens": 24315075.0,
      "step": 13215
    },
    {
      "entropy": 5.13605694770813,
      "epoch": 1.104243234213164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004883402635165705,
      "loss": 4.9944,
      "mean_token_accuracy": 0.2014315515756607,
      "num_tokens": 24323771.0,
      "step": 13220
    },
    {
      "entropy": 5.190080261230468,
      "epoch": 1.104660875375877,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004883308026314923,
      "loss": 5.067,
      "mean_token_accuracy": 0.19955671280622483,
      "num_tokens": 24334175.0,
      "step": 13225
    },
    {
      "entropy": 5.219483137130737,
      "epoch": 1.1050785165385901,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004883213380118016,
      "loss": 4.9881,
      "mean_token_accuracy": 0.20053200870752336,
      "num_tokens": 24344076.0,
      "step": 13230
    },
    {
      "entropy": 5.172171211242675,
      "epoch": 1.105496157701303,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004883118696576642,
      "loss": 5.0512,
      "mean_token_accuracy": 0.18882194459438323,
      "num_tokens": 24353333.0,
      "step": 13235
    },
    {
      "entropy": 5.230933332443238,
      "epoch": 1.105913798864016,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004883023975692457,
      "loss": 5.0325,
      "mean_token_accuracy": 0.19122972190380097,
      "num_tokens": 24362000.0,
      "step": 13240
    },
    {
      "entropy": 5.180014657974243,
      "epoch": 1.106331440026729,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004882929217467118,
      "loss": 4.8783,
      "mean_token_accuracy": 0.20691960752010347,
      "num_tokens": 24370490.0,
      "step": 13245
    },
    {
      "entropy": 5.101082992553711,
      "epoch": 1.106749081189442,
      "grad_norm": 1.25,
      "learning_rate": 0.0004882834421902287,
      "loss": 4.8896,
      "mean_token_accuracy": 0.21209489703178405,
      "num_tokens": 24378597.0,
      "step": 13250
    },
    {
      "entropy": 5.068929004669189,
      "epoch": 1.107166722352155,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048827395889996226,
      "loss": 4.9167,
      "mean_token_accuracy": 0.2033285140991211,
      "num_tokens": 24387050.0,
      "step": 13255
    },
    {
      "entropy": 5.133096075057983,
      "epoch": 1.107584363514868,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004882644718760784,
      "loss": 4.8922,
      "mean_token_accuracy": 0.21195074766874314,
      "num_tokens": 24395283.0,
      "step": 13260
    },
    {
      "entropy": 5.004158878326416,
      "epoch": 1.108002004677581,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048825498111874325,
      "loss": 4.8525,
      "mean_token_accuracy": 0.20842288434505463,
      "num_tokens": 24404206.0,
      "step": 13265
    },
    {
      "entropy": 5.135065317153931,
      "epoch": 1.108419645840294,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048824548662812294,
      "loss": 4.9051,
      "mean_token_accuracy": 0.20477092564105986,
      "num_tokens": 24414080.0,
      "step": 13270
    },
    {
      "entropy": 5.162567758560181,
      "epoch": 1.108837287003007,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048823598840438384,
      "loss": 4.9882,
      "mean_token_accuracy": 0.20155803114175797,
      "num_tokens": 24422406.0,
      "step": 13275
    },
    {
      "entropy": 5.159024286270141,
      "epoch": 1.10925492816572,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048822648644769194,
      "loss": 4.9675,
      "mean_token_accuracy": 0.20203019976615905,
      "num_tokens": 24432159.0,
      "step": 13280
    },
    {
      "entropy": 5.1401063919067385,
      "epoch": 1.109672569328433,
      "grad_norm": 1.25,
      "learning_rate": 0.0004882169807582138,
      "loss": 4.9689,
      "mean_token_accuracy": 0.20956576466560364,
      "num_tokens": 24440442.0,
      "step": 13285
    },
    {
      "entropy": 5.159703493118286,
      "epoch": 1.110090210491146,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004882074713361157,
      "loss": 4.965,
      "mean_token_accuracy": 0.203843355178833,
      "num_tokens": 24449717.0,
      "step": 13290
    },
    {
      "entropy": 5.15858302116394,
      "epoch": 1.110507851653859,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004881979581815642,
      "loss": 4.8864,
      "mean_token_accuracy": 0.21253306269645691,
      "num_tokens": 24458279.0,
      "step": 13295
    },
    {
      "entropy": 5.085260105133057,
      "epoch": 1.110925492816572,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004881884412947259,
      "loss": 4.8685,
      "mean_token_accuracy": 0.21036750227212905,
      "num_tokens": 24466096.0,
      "step": 13300
    },
    {
      "entropy": 5.149126386642456,
      "epoch": 1.111343133979285,
      "grad_norm": 1.25,
      "learning_rate": 0.00048817892067576726,
      "loss": 5.0266,
      "mean_token_accuracy": 0.1937076061964035,
      "num_tokens": 24474432.0,
      "step": 13305
    },
    {
      "entropy": 5.240054988861084,
      "epoch": 1.111760775141998,
      "grad_norm": 1.265625,
      "learning_rate": 0.000488169396324855,
      "loss": 5.057,
      "mean_token_accuracy": 0.19685342460870742,
      "num_tokens": 24484627.0,
      "step": 13310
    },
    {
      "entropy": 5.1504744529724125,
      "epoch": 1.112178416304711,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048815986824215585,
      "loss": 4.944,
      "mean_token_accuracy": 0.20632336288690567,
      "num_tokens": 24494010.0,
      "step": 13315
    },
    {
      "entropy": 5.131055307388306,
      "epoch": 1.112596057467424,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004881503364278366,
      "loss": 4.8728,
      "mean_token_accuracy": 0.2072224423289299,
      "num_tokens": 24503286.0,
      "step": 13320
    },
    {
      "entropy": 5.076667547225952,
      "epoch": 1.1130136986301369,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004881408008820642,
      "loss": 4.8475,
      "mean_token_accuracy": 0.2096168354153633,
      "num_tokens": 24512595.0,
      "step": 13325
    },
    {
      "entropy": 5.115263652801514,
      "epoch": 1.11343133979285,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048813126160500555,
      "loss": 4.9166,
      "mean_token_accuracy": 0.20265370160341262,
      "num_tokens": 24521358.0,
      "step": 13330
    },
    {
      "entropy": 5.274457168579102,
      "epoch": 1.113848980955563,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004881217185968276,
      "loss": 5.0811,
      "mean_token_accuracy": 0.19628337770700455,
      "num_tokens": 24530233.0,
      "step": 13335
    },
    {
      "entropy": 5.310393667221069,
      "epoch": 1.114266622118276,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004881121718576974,
      "loss": 5.0862,
      "mean_token_accuracy": 0.18637516647577285,
      "num_tokens": 24539645.0,
      "step": 13340
    },
    {
      "entropy": 5.243384790420532,
      "epoch": 1.114684263280989,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048810262138778213,
      "loss": 5.035,
      "mean_token_accuracy": 0.19591438323259353,
      "num_tokens": 24549085.0,
      "step": 13345
    },
    {
      "entropy": 5.195956802368164,
      "epoch": 1.115101904443702,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048809306718724886,
      "loss": 5.0405,
      "mean_token_accuracy": 0.19637491852045058,
      "num_tokens": 24557750.0,
      "step": 13350
    },
    {
      "entropy": 5.188299465179443,
      "epoch": 1.115519545606415,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048808350925626514,
      "loss": 4.9934,
      "mean_token_accuracy": 0.2018349677324295,
      "num_tokens": 24567281.0,
      "step": 13355
    },
    {
      "entropy": 5.118918943405151,
      "epoch": 1.115937186769128,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048807394759499797,
      "loss": 4.9445,
      "mean_token_accuracy": 0.20531732887029647,
      "num_tokens": 24576548.0,
      "step": 13360
    },
    {
      "entropy": 5.043518447875977,
      "epoch": 1.116354827931841,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004880643822036149,
      "loss": 4.8711,
      "mean_token_accuracy": 0.21077648252248765,
      "num_tokens": 24585528.0,
      "step": 13365
    },
    {
      "entropy": 5.171388339996338,
      "epoch": 1.116772469094554,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048805481308228333,
      "loss": 4.9786,
      "mean_token_accuracy": 0.20152204632759094,
      "num_tokens": 24594179.0,
      "step": 13370
    },
    {
      "entropy": 5.13182806968689,
      "epoch": 1.117190110257267,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004880452402311709,
      "loss": 5.0136,
      "mean_token_accuracy": 0.20144240707159042,
      "num_tokens": 24603365.0,
      "step": 13375
    },
    {
      "entropy": 5.174013376235962,
      "epoch": 1.11760775141998,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048803566365044493,
      "loss": 5.0233,
      "mean_token_accuracy": 0.19114116430282593,
      "num_tokens": 24613472.0,
      "step": 13380
    },
    {
      "entropy": 5.226609516143799,
      "epoch": 1.1180253925826928,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004880260833402734,
      "loss": 4.9788,
      "mean_token_accuracy": 0.2001221925020218,
      "num_tokens": 24623492.0,
      "step": 13385
    },
    {
      "entropy": 5.218543767929077,
      "epoch": 1.1184430337454059,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048801649930082375,
      "loss": 5.0594,
      "mean_token_accuracy": 0.19790757894515992,
      "num_tokens": 24633245.0,
      "step": 13390
    },
    {
      "entropy": 5.182901668548584,
      "epoch": 1.118860674908119,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000488006911532264,
      "loss": 4.8934,
      "mean_token_accuracy": 0.2092584937810898,
      "num_tokens": 24642652.0,
      "step": 13395
    },
    {
      "entropy": 5.095616149902344,
      "epoch": 1.119278316070832,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004879973200347617,
      "loss": 5.009,
      "mean_token_accuracy": 0.2051973298192024,
      "num_tokens": 24651650.0,
      "step": 13400
    },
    {
      "entropy": 5.195728492736817,
      "epoch": 1.119695957233545,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048798772480848504,
      "loss": 5.0327,
      "mean_token_accuracy": 0.1953406438231468,
      "num_tokens": 24661356.0,
      "step": 13405
    },
    {
      "entropy": 5.179117012023926,
      "epoch": 1.120113598396258,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048797812585360187,
      "loss": 4.9379,
      "mean_token_accuracy": 0.2067585438489914,
      "num_tokens": 24670100.0,
      "step": 13410
    },
    {
      "entropy": 5.179187488555908,
      "epoch": 1.120531239558971,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048796852317028024,
      "loss": 5.003,
      "mean_token_accuracy": 0.19608547389507294,
      "num_tokens": 24679437.0,
      "step": 13415
    },
    {
      "entropy": 5.205917310714722,
      "epoch": 1.1209488807216839,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004879589167586882,
      "loss": 4.958,
      "mean_token_accuracy": 0.1991334691643715,
      "num_tokens": 24689818.0,
      "step": 13420
    },
    {
      "entropy": 5.210038661956787,
      "epoch": 1.121366521884397,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048794930661899414,
      "loss": 5.0387,
      "mean_token_accuracy": 0.20575864464044571,
      "num_tokens": 24698323.0,
      "step": 13425
    },
    {
      "entropy": 5.1990148544311525,
      "epoch": 1.12178416304711,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048793969275136593,
      "loss": 5.0078,
      "mean_token_accuracy": 0.20010074228048325,
      "num_tokens": 24707447.0,
      "step": 13430
    },
    {
      "entropy": 5.1694653034210205,
      "epoch": 1.122201804209823,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004879300751559722,
      "loss": 4.9073,
      "mean_token_accuracy": 0.21182999312877654,
      "num_tokens": 24716470.0,
      "step": 13435
    },
    {
      "entropy": 5.138053607940674,
      "epoch": 1.122619445372536,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004879204538329812,
      "loss": 4.9062,
      "mean_token_accuracy": 0.2077472612261772,
      "num_tokens": 24725051.0,
      "step": 13440
    },
    {
      "entropy": 5.245497798919677,
      "epoch": 1.1230370865352488,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004879108287825614,
      "loss": 5.1168,
      "mean_token_accuracy": 0.19481105506420135,
      "num_tokens": 24735303.0,
      "step": 13445
    },
    {
      "entropy": 5.265983581542969,
      "epoch": 1.1234547276979618,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048790120000488115,
      "loss": 4.9505,
      "mean_token_accuracy": 0.19625573754310607,
      "num_tokens": 24743637.0,
      "step": 13450
    },
    {
      "entropy": 5.158891105651856,
      "epoch": 1.1238723688606749,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004878915675001092,
      "loss": 5.0244,
      "mean_token_accuracy": 0.19919659197330475,
      "num_tokens": 24752424.0,
      "step": 13455
    },
    {
      "entropy": 5.212686491012573,
      "epoch": 1.124290010023388,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004878819312684141,
      "loss": 5.0159,
      "mean_token_accuracy": 0.1992620497941971,
      "num_tokens": 24762224.0,
      "step": 13460
    },
    {
      "entropy": 5.274323987960815,
      "epoch": 1.124707651186101,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004878722913099645,
      "loss": 4.9957,
      "mean_token_accuracy": 0.20171975791454316,
      "num_tokens": 24771118.0,
      "step": 13465
    },
    {
      "entropy": 5.201798629760742,
      "epoch": 1.125125292348814,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004878626476249293,
      "loss": 5.0442,
      "mean_token_accuracy": 0.19630365520715715,
      "num_tokens": 24779941.0,
      "step": 13470
    },
    {
      "entropy": 5.123790073394775,
      "epoch": 1.125542933511527,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004878530002134772,
      "loss": 4.9662,
      "mean_token_accuracy": 0.20277682244777678,
      "num_tokens": 24788650.0,
      "step": 13475
    },
    {
      "entropy": 5.246025085449219,
      "epoch": 1.1259605746742398,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004878433490757771,
      "loss": 5.107,
      "mean_token_accuracy": 0.1920781061053276,
      "num_tokens": 24797953.0,
      "step": 13480
    },
    {
      "entropy": 5.219470548629761,
      "epoch": 1.1263782158369529,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004878336942119979,
      "loss": 4.9667,
      "mean_token_accuracy": 0.2088983654975891,
      "num_tokens": 24807073.0,
      "step": 13485
    },
    {
      "entropy": 5.0711140632629395,
      "epoch": 1.126795856999666,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048782403562230874,
      "loss": 4.9107,
      "mean_token_accuracy": 0.20147655159235,
      "num_tokens": 24815839.0,
      "step": 13490
    },
    {
      "entropy": 5.189441967010498,
      "epoch": 1.127213498162379,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004878143733068787,
      "loss": 5.0925,
      "mean_token_accuracy": 0.19571423828601836,
      "num_tokens": 24825496.0,
      "step": 13495
    },
    {
      "entropy": 5.253018617630005,
      "epoch": 1.127631139325092,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004878047072658769,
      "loss": 4.9395,
      "mean_token_accuracy": 0.20792672634124756,
      "num_tokens": 24834527.0,
      "step": 13500
    },
    {
      "entropy": 5.09864068031311,
      "epoch": 1.1280487804878048,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004877950374994725,
      "loss": 4.9409,
      "mean_token_accuracy": 0.21470311284065247,
      "num_tokens": 24843136.0,
      "step": 13505
    },
    {
      "entropy": 5.228253555297852,
      "epoch": 1.1284664216505178,
      "grad_norm": 1.25,
      "learning_rate": 0.0004877853640078349,
      "loss": 5.1015,
      "mean_token_accuracy": 0.1951835945248604,
      "num_tokens": 24853304.0,
      "step": 13510
    },
    {
      "entropy": 5.164680814743042,
      "epoch": 1.1288840628132308,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004877756867911333,
      "loss": 4.9019,
      "mean_token_accuracy": 0.20373250246047975,
      "num_tokens": 24861793.0,
      "step": 13515
    },
    {
      "entropy": 5.159054517745972,
      "epoch": 1.1293017039759439,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004877660058495372,
      "loss": 4.9767,
      "mean_token_accuracy": 0.20625679492950438,
      "num_tokens": 24871340.0,
      "step": 13520
    },
    {
      "entropy": 5.252390623092651,
      "epoch": 1.129719345138657,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004877563211832161,
      "loss": 5.0525,
      "mean_token_accuracy": 0.19956205189228057,
      "num_tokens": 24880796.0,
      "step": 13525
    },
    {
      "entropy": 5.157743787765503,
      "epoch": 1.13013698630137,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048774663279233947,
      "loss": 4.975,
      "mean_token_accuracy": 0.21114081144332886,
      "num_tokens": 24889801.0,
      "step": 13530
    },
    {
      "entropy": 5.190232229232788,
      "epoch": 1.130554627464083,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048773694067707693,
      "loss": 4.9939,
      "mean_token_accuracy": 0.20347868651151657,
      "num_tokens": 24899032.0,
      "step": 13535
    },
    {
      "entropy": 5.214659404754639,
      "epoch": 1.1309722686267958,
      "grad_norm": 1.25,
      "learning_rate": 0.00048772724483759825,
      "loss": 5.0968,
      "mean_token_accuracy": 0.1981140211224556,
      "num_tokens": 24908792.0,
      "step": 13540
    },
    {
      "entropy": 5.19938097000122,
      "epoch": 1.1313899097895088,
      "grad_norm": 1.375,
      "learning_rate": 0.000487717545274073,
      "loss": 5.0152,
      "mean_token_accuracy": 0.20325371474027634,
      "num_tokens": 24918330.0,
      "step": 13545
    },
    {
      "entropy": 5.227181625366211,
      "epoch": 1.1318075509522219,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004877078419866711,
      "loss": 5.0013,
      "mean_token_accuracy": 0.20410169214010238,
      "num_tokens": 24928039.0,
      "step": 13550
    },
    {
      "entropy": 5.212827396392822,
      "epoch": 1.1322251921149349,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048769813497556236,
      "loss": 5.0783,
      "mean_token_accuracy": 0.19871995598077774,
      "num_tokens": 24937145.0,
      "step": 13555
    },
    {
      "entropy": 5.195865774154663,
      "epoch": 1.132642833277648,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004876884242409168,
      "loss": 4.9927,
      "mean_token_accuracy": 0.19791145473718644,
      "num_tokens": 24945879.0,
      "step": 13560
    },
    {
      "entropy": 5.100244140625,
      "epoch": 1.1330604744403607,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048767870978290434,
      "loss": 4.8822,
      "mean_token_accuracy": 0.21645372807979585,
      "num_tokens": 24954900.0,
      "step": 13565
    },
    {
      "entropy": 5.286684465408325,
      "epoch": 1.1334781156030738,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004876689916016951,
      "loss": 5.1428,
      "mean_token_accuracy": 0.19452081322669984,
      "num_tokens": 24965015.0,
      "step": 13570
    },
    {
      "entropy": 5.253557825088501,
      "epoch": 1.1338957567657868,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004876592696974591,
      "loss": 4.9446,
      "mean_token_accuracy": 0.20477031767368317,
      "num_tokens": 24975125.0,
      "step": 13575
    },
    {
      "entropy": 5.163647079467774,
      "epoch": 1.1343133979284998,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004876495440703667,
      "loss": 5.0411,
      "mean_token_accuracy": 0.19979415982961654,
      "num_tokens": 24983895.0,
      "step": 13580
    },
    {
      "entropy": 5.119611310958862,
      "epoch": 1.1347310390912129,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004876398147205879,
      "loss": 4.9492,
      "mean_token_accuracy": 0.2106444537639618,
      "num_tokens": 24993186.0,
      "step": 13585
    },
    {
      "entropy": 5.158004188537598,
      "epoch": 1.135148680253926,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004876300816482933,
      "loss": 4.9868,
      "mean_token_accuracy": 0.21135794520378112,
      "num_tokens": 25001785.0,
      "step": 13590
    },
    {
      "entropy": 5.243739128112793,
      "epoch": 1.135566321416639,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004876203448536532,
      "loss": 4.9836,
      "mean_token_accuracy": 0.20373328626155854,
      "num_tokens": 25010717.0,
      "step": 13595
    },
    {
      "entropy": 5.1449426174163815,
      "epoch": 1.1359839625793517,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004876106043368381,
      "loss": 4.8966,
      "mean_token_accuracy": 0.2063333049416542,
      "num_tokens": 25019438.0,
      "step": 13600
    },
    {
      "entropy": 5.12303524017334,
      "epoch": 1.1364016037420648,
      "grad_norm": 1.25,
      "learning_rate": 0.0004876008600980183,
      "loss": 4.9525,
      "mean_token_accuracy": 0.19983595311641694,
      "num_tokens": 25028601.0,
      "step": 13605
    },
    {
      "entropy": 5.147025442123413,
      "epoch": 1.1368192449047778,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048759111213736455,
      "loss": 4.9979,
      "mean_token_accuracy": 0.19687586426734924,
      "num_tokens": 25038046.0,
      "step": 13610
    },
    {
      "entropy": 5.138490438461304,
      "epoch": 1.1372368860674908,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004875813604550476,
      "loss": 4.9032,
      "mean_token_accuracy": 0.20973242968320846,
      "num_tokens": 25046712.0,
      "step": 13615
    },
    {
      "entropy": 5.081827640533447,
      "epoch": 1.1376545272302039,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004875716050512379,
      "loss": 4.885,
      "mean_token_accuracy": 0.20888490825891495,
      "num_tokens": 25054918.0,
      "step": 13620
    },
    {
      "entropy": 5.157643461227417,
      "epoch": 1.1380721683929167,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004875618459261065,
      "loss": 4.9421,
      "mean_token_accuracy": 0.20359669476747513,
      "num_tokens": 25064187.0,
      "step": 13625
    },
    {
      "entropy": 5.24801115989685,
      "epoch": 1.1384898095556297,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048755208307982414,
      "loss": 4.9668,
      "mean_token_accuracy": 0.200511372089386,
      "num_tokens": 25073700.0,
      "step": 13630
    },
    {
      "entropy": 5.2073126316070555,
      "epoch": 1.1389074507183428,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004875423165125616,
      "loss": 5.0504,
      "mean_token_accuracy": 0.19598623365163803,
      "num_tokens": 25082974.0,
      "step": 13635
    },
    {
      "entropy": 5.178184223175049,
      "epoch": 1.1393250918810558,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048753254622449006,
      "loss": 4.988,
      "mean_token_accuracy": 0.1999707415699959,
      "num_tokens": 25092450.0,
      "step": 13640
    },
    {
      "entropy": 5.119438695907593,
      "epoch": 1.1397427330437688,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004875227722157804,
      "loss": 4.8981,
      "mean_token_accuracy": 0.20395178496837615,
      "num_tokens": 25102731.0,
      "step": 13645
    },
    {
      "entropy": 5.076331281661988,
      "epoch": 1.1401603742064819,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004875129944866039,
      "loss": 4.8764,
      "mean_token_accuracy": 0.207842880487442,
      "num_tokens": 25111159.0,
      "step": 13650
    },
    {
      "entropy": 5.144111442565918,
      "epoch": 1.140578015369195,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004875032130371316,
      "loss": 4.9689,
      "mean_token_accuracy": 0.20648204386234284,
      "num_tokens": 25120151.0,
      "step": 13655
    },
    {
      "entropy": 5.142707967758179,
      "epoch": 1.1409956565319077,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004874934278675347,
      "loss": 4.9528,
      "mean_token_accuracy": 0.20534465461969376,
      "num_tokens": 25128788.0,
      "step": 13660
    },
    {
      "entropy": 5.107631874084473,
      "epoch": 1.1414132976946207,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004874836389779846,
      "loss": 4.9065,
      "mean_token_accuracy": 0.2037864461541176,
      "num_tokens": 25138149.0,
      "step": 13665
    },
    {
      "entropy": 5.172539710998535,
      "epoch": 1.1418309388573338,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048747384636865267,
      "loss": 4.9767,
      "mean_token_accuracy": 0.19343682676553725,
      "num_tokens": 25147798.0,
      "step": 13670
    },
    {
      "entropy": 5.138930892944336,
      "epoch": 1.1422485800200468,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004874640500397103,
      "loss": 4.8863,
      "mean_token_accuracy": 0.20320495814085007,
      "num_tokens": 25156626.0,
      "step": 13675
    },
    {
      "entropy": 5.175389337539673,
      "epoch": 1.1426662211827598,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004874542499913289,
      "loss": 4.9928,
      "mean_token_accuracy": 0.19642375260591508,
      "num_tokens": 25165846.0,
      "step": 13680
    },
    {
      "entropy": 5.12396273612976,
      "epoch": 1.1430838623454727,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048744444622368015,
      "loss": 4.9234,
      "mean_token_accuracy": 0.20784615576267243,
      "num_tokens": 25175218.0,
      "step": 13685
    },
    {
      "entropy": 5.2208428382873535,
      "epoch": 1.1435015035081857,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048743463873693576,
      "loss": 5.0311,
      "mean_token_accuracy": 0.19916563481092453,
      "num_tokens": 25185257.0,
      "step": 13690
    },
    {
      "entropy": 5.273836898803711,
      "epoch": 1.1439191446708987,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004874248275312673,
      "loss": 5.0836,
      "mean_token_accuracy": 0.1926201105117798,
      "num_tokens": 25195348.0,
      "step": 13695
    },
    {
      "entropy": 5.1572037696838375,
      "epoch": 1.1443367858336118,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048741501260684644,
      "loss": 4.989,
      "mean_token_accuracy": 0.2076288104057312,
      "num_tokens": 25204810.0,
      "step": 13700
    },
    {
      "entropy": 5.0716104984283445,
      "epoch": 1.1447544269963248,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048740519396384515,
      "loss": 4.8365,
      "mean_token_accuracy": 0.21286435276269913,
      "num_tokens": 25214490.0,
      "step": 13705
    },
    {
      "entropy": 5.07154655456543,
      "epoch": 1.1451720681590378,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048739537160243524,
      "loss": 4.8653,
      "mean_token_accuracy": 0.21740389466285706,
      "num_tokens": 25222840.0,
      "step": 13710
    },
    {
      "entropy": 5.124548149108887,
      "epoch": 1.1455897093217509,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048738554552278874,
      "loss": 4.8618,
      "mean_token_accuracy": 0.20677186101675032,
      "num_tokens": 25231056.0,
      "step": 13715
    },
    {
      "entropy": 5.1250114917755125,
      "epoch": 1.1460073504844637,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048737571572507753,
      "loss": 4.9489,
      "mean_token_accuracy": 0.19861506521701813,
      "num_tokens": 25240445.0,
      "step": 13720
    },
    {
      "entropy": 5.142857789993286,
      "epoch": 1.1464249916471767,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048736588220947384,
      "loss": 4.9519,
      "mean_token_accuracy": 0.20813391208648682,
      "num_tokens": 25249113.0,
      "step": 13725
    },
    {
      "entropy": 5.135421466827393,
      "epoch": 1.1468426328098897,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048735604497614974,
      "loss": 4.9378,
      "mean_token_accuracy": 0.20647278875112535,
      "num_tokens": 25258202.0,
      "step": 13730
    },
    {
      "entropy": 5.275015830993652,
      "epoch": 1.1472602739726028,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004873462040252775,
      "loss": 5.0426,
      "mean_token_accuracy": 0.1973186269402504,
      "num_tokens": 25268052.0,
      "step": 13735
    },
    {
      "entropy": 5.186558437347412,
      "epoch": 1.1476779151353158,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004873363593570293,
      "loss": 5.0298,
      "mean_token_accuracy": 0.19410761892795564,
      "num_tokens": 25276696.0,
      "step": 13740
    },
    {
      "entropy": 5.09901065826416,
      "epoch": 1.1480955562980286,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004873265109715776,
      "loss": 4.8988,
      "mean_token_accuracy": 0.207858669757843,
      "num_tokens": 25284920.0,
      "step": 13745
    },
    {
      "entropy": 5.084775638580322,
      "epoch": 1.1485131974607417,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004873166588690947,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2109785035252571,
      "num_tokens": 25293960.0,
      "step": 13750
    },
    {
      "entropy": 5.1927878856658936,
      "epoch": 1.1489308386234547,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004873068030497532,
      "loss": 5.0758,
      "mean_token_accuracy": 0.19657916575670242,
      "num_tokens": 25303948.0,
      "step": 13755
    },
    {
      "entropy": 5.298254442214966,
      "epoch": 1.1493484797861677,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048729694351372545,
      "loss": 5.0694,
      "mean_token_accuracy": 0.2034141942858696,
      "num_tokens": 25313343.0,
      "step": 13760
    },
    {
      "entropy": 5.1492541313171385,
      "epoch": 1.1497661209488808,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004872870802611842,
      "loss": 4.8618,
      "mean_token_accuracy": 0.2067318856716156,
      "num_tokens": 25321664.0,
      "step": 13765
    },
    {
      "entropy": 5.177978086471557,
      "epoch": 1.1501837621115938,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004872772132923021,
      "loss": 5.0244,
      "mean_token_accuracy": 0.2090502381324768,
      "num_tokens": 25331112.0,
      "step": 13770
    },
    {
      "entropy": 5.226603078842163,
      "epoch": 1.1506014032743068,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048726734260725186,
      "loss": 4.9726,
      "mean_token_accuracy": 0.2077540248632431,
      "num_tokens": 25340357.0,
      "step": 13775
    },
    {
      "entropy": 5.193742418289185,
      "epoch": 1.1510190444370196,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004872574682062063,
      "loss": 5.0279,
      "mean_token_accuracy": 0.20372698158025743,
      "num_tokens": 25349886.0,
      "step": 13780
    },
    {
      "entropy": 5.2048546314239506,
      "epoch": 1.1514366855997327,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004872475900893382,
      "loss": 5.0382,
      "mean_token_accuracy": 0.200131356716156,
      "num_tokens": 25358957.0,
      "step": 13785
    },
    {
      "entropy": 5.133332443237305,
      "epoch": 1.1518543267624457,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004872377082568206,
      "loss": 4.8686,
      "mean_token_accuracy": 0.21120152473449708,
      "num_tokens": 25368237.0,
      "step": 13790
    },
    {
      "entropy": 5.205290651321411,
      "epoch": 1.1522719679251587,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048722782270882643,
      "loss": 4.8979,
      "mean_token_accuracy": 0.21422120928764343,
      "num_tokens": 25376626.0,
      "step": 13795
    },
    {
      "entropy": 5.114291381835938,
      "epoch": 1.1526896090878718,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004872179334455288,
      "loss": 4.9181,
      "mean_token_accuracy": 0.20702963024377824,
      "num_tokens": 25385824.0,
      "step": 13800
    },
    {
      "entropy": 5.110055780410766,
      "epoch": 1.1531072502505846,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048720804046710074,
      "loss": 4.8924,
      "mean_token_accuracy": 0.2123393937945366,
      "num_tokens": 25394317.0,
      "step": 13805
    },
    {
      "entropy": 5.142131996154785,
      "epoch": 1.1535248914132976,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048719814377371553,
      "loss": 4.9778,
      "mean_token_accuracy": 0.19662246555089952,
      "num_tokens": 25402962.0,
      "step": 13810
    },
    {
      "entropy": 5.136072254180908,
      "epoch": 1.1539425325760106,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004871882433655464,
      "loss": 4.8757,
      "mean_token_accuracy": 0.2076478362083435,
      "num_tokens": 25411660.0,
      "step": 13815
    },
    {
      "entropy": 5.138869524002075,
      "epoch": 1.1543601737387237,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004871783392427666,
      "loss": 4.9553,
      "mean_token_accuracy": 0.2050904870033264,
      "num_tokens": 25420377.0,
      "step": 13820
    },
    {
      "entropy": 5.0928856372833256,
      "epoch": 1.1547778149014367,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004871684314055497,
      "loss": 4.9176,
      "mean_token_accuracy": 0.20411113649606705,
      "num_tokens": 25428811.0,
      "step": 13825
    },
    {
      "entropy": 5.16126503944397,
      "epoch": 1.1551954560641498,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004871585198540689,
      "loss": 4.9944,
      "mean_token_accuracy": 0.196958290040493,
      "num_tokens": 25438247.0,
      "step": 13830
    },
    {
      "entropy": 5.121724271774292,
      "epoch": 1.1556130972268628,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048714860458849794,
      "loss": 4.8795,
      "mean_token_accuracy": 0.21387530565261842,
      "num_tokens": 25447122.0,
      "step": 13835
    },
    {
      "entropy": 5.218702983856201,
      "epoch": 1.1560307383895756,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048713868560901013,
      "loss": 5.0711,
      "mean_token_accuracy": 0.19090910404920577,
      "num_tokens": 25455737.0,
      "step": 13840
    },
    {
      "entropy": 5.1803319454193115,
      "epoch": 1.1564483795522886,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004871287629157794,
      "loss": 4.9,
      "mean_token_accuracy": 0.20876794159412385,
      "num_tokens": 25465285.0,
      "step": 13845
    },
    {
      "entropy": 5.166810464859009,
      "epoch": 1.1568660207150017,
      "grad_norm": 1.25,
      "learning_rate": 0.0004871188365089793,
      "loss": 4.9546,
      "mean_token_accuracy": 0.20883765518665315,
      "num_tokens": 25474933.0,
      "step": 13850
    },
    {
      "entropy": 5.104287767410279,
      "epoch": 1.1572836618777147,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004871089063887837,
      "loss": 4.9502,
      "mean_token_accuracy": 0.20794036388397216,
      "num_tokens": 25483279.0,
      "step": 13855
    },
    {
      "entropy": 5.187629079818725,
      "epoch": 1.1577013030404277,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048709897255536626,
      "loss": 5.004,
      "mean_token_accuracy": 0.20151136815547943,
      "num_tokens": 25492295.0,
      "step": 13860
    },
    {
      "entropy": 5.2023850440979,
      "epoch": 1.1581189442031405,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004870890350089011,
      "loss": 5.0011,
      "mean_token_accuracy": 0.20169200897216796,
      "num_tokens": 25501397.0,
      "step": 13865
    },
    {
      "entropy": 5.062053966522217,
      "epoch": 1.1585365853658536,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004870790937495621,
      "loss": 4.9096,
      "mean_token_accuracy": 0.2062204048037529,
      "num_tokens": 25510588.0,
      "step": 13870
    },
    {
      "entropy": 5.235393095016479,
      "epoch": 1.1589542265285666,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048706914877752325,
      "loss": 4.9901,
      "mean_token_accuracy": 0.20219841599464417,
      "num_tokens": 25519531.0,
      "step": 13875
    },
    {
      "entropy": 5.154824829101562,
      "epoch": 1.1593718676912796,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048705920009295867,
      "loss": 4.8935,
      "mean_token_accuracy": 0.20541852563619614,
      "num_tokens": 25529241.0,
      "step": 13880
    },
    {
      "entropy": 5.186248350143432,
      "epoch": 1.1597895088539927,
      "grad_norm": 1.25,
      "learning_rate": 0.0004870492476960426,
      "loss": 5.016,
      "mean_token_accuracy": 0.19206317216157914,
      "num_tokens": 25538463.0,
      "step": 13885
    },
    {
      "entropy": 5.167735052108765,
      "epoch": 1.1602071500167057,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004870392915869491,
      "loss": 4.9974,
      "mean_token_accuracy": 0.20156778991222382,
      "num_tokens": 25547481.0,
      "step": 13890
    },
    {
      "entropy": 5.2040567874908445,
      "epoch": 1.1606247911794187,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004870293317658526,
      "loss": 4.9459,
      "mean_token_accuracy": 0.20362902134656907,
      "num_tokens": 25556083.0,
      "step": 13895
    },
    {
      "entropy": 5.152432250976562,
      "epoch": 1.1610424323421316,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004870193682329275,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20719525516033171,
      "num_tokens": 25564975.0,
      "step": 13900
    },
    {
      "entropy": 5.019022464752197,
      "epoch": 1.1614600735048446,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048700940098834805,
      "loss": 4.7809,
      "mean_token_accuracy": 0.2113082990050316,
      "num_tokens": 25574576.0,
      "step": 13905
    },
    {
      "entropy": 5.159143400192261,
      "epoch": 1.1618777146675576,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004869994300322889,
      "loss": 4.9487,
      "mean_token_accuracy": 0.20774728953838348,
      "num_tokens": 25583485.0,
      "step": 13910
    },
    {
      "entropy": 5.16772494316101,
      "epoch": 1.1622953558302707,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048698945536492454,
      "loss": 4.9571,
      "mean_token_accuracy": 0.20200437903404236,
      "num_tokens": 25592736.0,
      "step": 13915
    },
    {
      "entropy": 5.198943996429444,
      "epoch": 1.1627129969929837,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004869794769864296,
      "loss": 4.977,
      "mean_token_accuracy": 0.20553614497184752,
      "num_tokens": 25601230.0,
      "step": 13920
    },
    {
      "entropy": 5.218483114242554,
      "epoch": 1.1631306381556965,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004869694948969787,
      "loss": 4.9514,
      "mean_token_accuracy": 0.20510891526937486,
      "num_tokens": 25611139.0,
      "step": 13925
    },
    {
      "entropy": 5.225070285797119,
      "epoch": 1.1635482793184095,
      "grad_norm": 1.375,
      "learning_rate": 0.00048695950909674666,
      "loss": 5.002,
      "mean_token_accuracy": 0.19451773464679717,
      "num_tokens": 25620255.0,
      "step": 13930
    },
    {
      "entropy": 5.101096773147583,
      "epoch": 1.1639659204811226,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048694951958590835,
      "loss": 4.866,
      "mean_token_accuracy": 0.20948225408792495,
      "num_tokens": 25629127.0,
      "step": 13935
    },
    {
      "entropy": 5.133993816375733,
      "epoch": 1.1643835616438356,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048693952636463854,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2066566452383995,
      "num_tokens": 25638561.0,
      "step": 13940
    },
    {
      "entropy": 5.094225978851318,
      "epoch": 1.1648012028065486,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048692952943311214,
      "loss": 4.9422,
      "mean_token_accuracy": 0.19875528961420058,
      "num_tokens": 25647268.0,
      "step": 13945
    },
    {
      "entropy": 5.184319305419922,
      "epoch": 1.1652188439692617,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004869195287915043,
      "loss": 5.0397,
      "mean_token_accuracy": 0.19910807609558107,
      "num_tokens": 25656804.0,
      "step": 13950
    },
    {
      "entropy": 5.214507675170898,
      "epoch": 1.1656364851319747,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048690952443999,
      "loss": 4.939,
      "mean_token_accuracy": 0.20769357234239577,
      "num_tokens": 25665960.0,
      "step": 13955
    },
    {
      "entropy": 5.304385423660278,
      "epoch": 1.1660541262946875,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048689951637874445,
      "loss": 5.159,
      "mean_token_accuracy": 0.18981515914201735,
      "num_tokens": 25676836.0,
      "step": 13960
    },
    {
      "entropy": 5.16274790763855,
      "epoch": 1.1664717674574006,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004868895046079428,
      "loss": 4.8792,
      "mean_token_accuracy": 0.20243281573057176,
      "num_tokens": 25685751.0,
      "step": 13965
    },
    {
      "entropy": 5.10176568031311,
      "epoch": 1.1668894086201136,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004868794891277602,
      "loss": 4.8889,
      "mean_token_accuracy": 0.2062852129340172,
      "num_tokens": 25694948.0,
      "step": 13970
    },
    {
      "entropy": 5.20370192527771,
      "epoch": 1.1673070497828266,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004868694699383722,
      "loss": 5.1107,
      "mean_token_accuracy": 0.2013367608189583,
      "num_tokens": 25704135.0,
      "step": 13975
    },
    {
      "entropy": 5.345385265350342,
      "epoch": 1.1677246909455397,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004868594470399541,
      "loss": 5.1209,
      "mean_token_accuracy": 0.19421169608831407,
      "num_tokens": 25714300.0,
      "step": 13980
    },
    {
      "entropy": 5.203368663787842,
      "epoch": 1.1681423321082525,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048684942043268137,
      "loss": 4.9033,
      "mean_token_accuracy": 0.2070540592074394,
      "num_tokens": 25723275.0,
      "step": 13985
    },
    {
      "entropy": 5.215776252746582,
      "epoch": 1.1685599732709655,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048683939011672953,
      "loss": 5.13,
      "mean_token_accuracy": 0.19044512659311294,
      "num_tokens": 25732432.0,
      "step": 13990
    },
    {
      "entropy": 5.109826135635376,
      "epoch": 1.1689776144336785,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048682935609227414,
      "loss": 4.9218,
      "mean_token_accuracy": 0.1989915043115616,
      "num_tokens": 25741001.0,
      "step": 13995
    },
    {
      "entropy": 5.2178901672363285,
      "epoch": 1.1693952555963916,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048681931835949083,
      "loss": 5.0328,
      "mean_token_accuracy": 0.1976896345615387,
      "num_tokens": 25751632.0,
      "step": 14000
    },
    {
      "entropy": 5.198111534118652,
      "epoch": 1.1698128967591046,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004868092769185555,
      "loss": 4.981,
      "mean_token_accuracy": 0.19836863428354262,
      "num_tokens": 25760963.0,
      "step": 14005
    },
    {
      "entropy": 5.142030239105225,
      "epoch": 1.1702305379218176,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048679923176964384,
      "loss": 4.9494,
      "mean_token_accuracy": 0.2051076650619507,
      "num_tokens": 25769974.0,
      "step": 14010
    },
    {
      "entropy": 5.185855674743652,
      "epoch": 1.1706481790845307,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004867891829129316,
      "loss": 4.963,
      "mean_token_accuracy": 0.2020861253142357,
      "num_tokens": 25778311.0,
      "step": 14015
    },
    {
      "entropy": 5.223141813278199,
      "epoch": 1.1710658202472435,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004867791303485948,
      "loss": 4.9964,
      "mean_token_accuracy": 0.19930825978517533,
      "num_tokens": 25787811.0,
      "step": 14020
    },
    {
      "entropy": 5.20784821510315,
      "epoch": 1.1714834614099565,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004867690740768094,
      "loss": 4.9509,
      "mean_token_accuracy": 0.2062822014093399,
      "num_tokens": 25796651.0,
      "step": 14025
    },
    {
      "entropy": 5.1913830757141115,
      "epoch": 1.1719011025726696,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048675901409775147,
      "loss": 5.0055,
      "mean_token_accuracy": 0.20319464355707167,
      "num_tokens": 25805995.0,
      "step": 14030
    },
    {
      "entropy": 5.193567705154419,
      "epoch": 1.1723187437353826,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048674895041159714,
      "loss": 4.9847,
      "mean_token_accuracy": 0.20811574757099152,
      "num_tokens": 25814833.0,
      "step": 14035
    },
    {
      "entropy": 5.246817541122437,
      "epoch": 1.1727363848980956,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048673888301852247,
      "loss": 5.096,
      "mean_token_accuracy": 0.20544013530015945,
      "num_tokens": 25823224.0,
      "step": 14040
    },
    {
      "entropy": 5.13305344581604,
      "epoch": 1.1731540260608084,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004867288119187039,
      "loss": 4.9175,
      "mean_token_accuracy": 0.20710639506578446,
      "num_tokens": 25832583.0,
      "step": 14045
    },
    {
      "entropy": 5.1052168846130375,
      "epoch": 1.1735716672235215,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004867187371123175,
      "loss": 4.986,
      "mean_token_accuracy": 0.2038514420390129,
      "num_tokens": 25841233.0,
      "step": 14050
    },
    {
      "entropy": 5.159973192214966,
      "epoch": 1.1739893083862345,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048670865859953984,
      "loss": 4.916,
      "mean_token_accuracy": 0.2012063130736351,
      "num_tokens": 25849794.0,
      "step": 14055
    },
    {
      "entropy": 5.160331106185913,
      "epoch": 1.1744069495489475,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004866985763805472,
      "loss": 4.8701,
      "mean_token_accuracy": 0.20307471454143525,
      "num_tokens": 25858629.0,
      "step": 14060
    },
    {
      "entropy": 5.156593275070191,
      "epoch": 1.1748245907116606,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004866884904555162,
      "loss": 4.9812,
      "mean_token_accuracy": 0.20620591789484025,
      "num_tokens": 25868382.0,
      "step": 14065
    },
    {
      "entropy": 5.173895215988159,
      "epoch": 1.1752422318743736,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004866784008246233,
      "loss": 4.9946,
      "mean_token_accuracy": 0.19952823966741562,
      "num_tokens": 25877565.0,
      "step": 14070
    },
    {
      "entropy": 5.2475786209106445,
      "epoch": 1.1756598730370866,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004866683074880453,
      "loss": 5.1016,
      "mean_token_accuracy": 0.19166408777236937,
      "num_tokens": 25886791.0,
      "step": 14075
    },
    {
      "entropy": 5.2184288024902346,
      "epoch": 1.1760775141997994,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048665821044595884,
      "loss": 4.9522,
      "mean_token_accuracy": 0.20274678915739058,
      "num_tokens": 25896190.0,
      "step": 14080
    },
    {
      "entropy": 5.100507736206055,
      "epoch": 1.1764951553625125,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004866481096985405,
      "loss": 4.8988,
      "mean_token_accuracy": 0.20330509096384047,
      "num_tokens": 25904702.0,
      "step": 14085
    },
    {
      "entropy": 5.21715350151062,
      "epoch": 1.1769127965252255,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004866380052459673,
      "loss": 5.0353,
      "mean_token_accuracy": 0.20566550493240357,
      "num_tokens": 25914513.0,
      "step": 14090
    },
    {
      "entropy": 5.088445615768433,
      "epoch": 1.1773304376879385,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004866278970884161,
      "loss": 4.9096,
      "mean_token_accuracy": 0.20272248834371567,
      "num_tokens": 25923436.0,
      "step": 14095
    },
    {
      "entropy": 5.14292540550232,
      "epoch": 1.1777480788506516,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048661778522606376,
      "loss": 4.9228,
      "mean_token_accuracy": 0.2077519714832306,
      "num_tokens": 25931883.0,
      "step": 14100
    },
    {
      "entropy": 5.169765615463257,
      "epoch": 1.1781657200133644,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048660766965908736,
      "loss": 4.9941,
      "mean_token_accuracy": 0.19358734786510468,
      "num_tokens": 25942095.0,
      "step": 14105
    },
    {
      "entropy": 5.1053839206695555,
      "epoch": 1.1785833611760774,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048659755038766406,
      "loss": 4.8299,
      "mean_token_accuracy": 0.2170640230178833,
      "num_tokens": 25951265.0,
      "step": 14110
    },
    {
      "entropy": 5.1353813171386715,
      "epoch": 1.1790010023387905,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048658742741197086,
      "loss": 4.9241,
      "mean_token_accuracy": 0.199820576608181,
      "num_tokens": 25960621.0,
      "step": 14115
    },
    {
      "entropy": 5.130825519561768,
      "epoch": 1.1794186435015035,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048657730073218517,
      "loss": 4.9845,
      "mean_token_accuracy": 0.19452319592237471,
      "num_tokens": 25970769.0,
      "step": 14120
    },
    {
      "entropy": 5.2152868747711185,
      "epoch": 1.1798362846642165,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004865671703484841,
      "loss": 5.0872,
      "mean_token_accuracy": 0.1986862078309059,
      "num_tokens": 25980928.0,
      "step": 14125
    },
    {
      "entropy": 5.250247669219971,
      "epoch": 1.1802539258269296,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048655703626104493,
      "loss": 5.0275,
      "mean_token_accuracy": 0.19748563319444656,
      "num_tokens": 25991085.0,
      "step": 14130
    },
    {
      "entropy": 5.114477729797363,
      "epoch": 1.1806715669896426,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048654689847004535,
      "loss": 4.8399,
      "mean_token_accuracy": 0.21031931191682815,
      "num_tokens": 25999862.0,
      "step": 14135
    },
    {
      "entropy": 5.2166191101074215,
      "epoch": 1.1810892081523554,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004865367569756626,
      "loss": 4.9553,
      "mean_token_accuracy": 0.20333696901798248,
      "num_tokens": 26009075.0,
      "step": 14140
    },
    {
      "entropy": 5.1617128372192385,
      "epoch": 1.1815068493150684,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048652661177807426,
      "loss": 4.8576,
      "mean_token_accuracy": 0.20697477459907532,
      "num_tokens": 26017822.0,
      "step": 14145
    },
    {
      "entropy": 5.041386413574219,
      "epoch": 1.1819244904777815,
      "grad_norm": 1.25,
      "learning_rate": 0.00048651646287745803,
      "loss": 4.9036,
      "mean_token_accuracy": 0.20254568457603456,
      "num_tokens": 26026327.0,
      "step": 14150
    },
    {
      "entropy": 5.244654893875122,
      "epoch": 1.1823421316404945,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048650631027399155,
      "loss": 5.0916,
      "mean_token_accuracy": 0.18919677585363387,
      "num_tokens": 26036252.0,
      "step": 14155
    },
    {
      "entropy": 5.283205604553222,
      "epoch": 1.1827597728032075,
      "grad_norm": 1.25,
      "learning_rate": 0.00048649615396785246,
      "loss": 5.0252,
      "mean_token_accuracy": 0.20073236674070358,
      "num_tokens": 26045596.0,
      "step": 14160
    },
    {
      "entropy": 5.1950407981872555,
      "epoch": 1.1831774139659206,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048648599395921863,
      "loss": 4.933,
      "mean_token_accuracy": 0.2103711649775505,
      "num_tokens": 26054824.0,
      "step": 14165
    },
    {
      "entropy": 5.161082553863525,
      "epoch": 1.1835950551286334,
      "grad_norm": 1.25,
      "learning_rate": 0.00048647583024826786,
      "loss": 4.9377,
      "mean_token_accuracy": 0.1982040971517563,
      "num_tokens": 26064414.0,
      "step": 14170
    },
    {
      "entropy": 5.188171768188477,
      "epoch": 1.1840126962913464,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048646566283517824,
      "loss": 4.9541,
      "mean_token_accuracy": 0.20044438540935516,
      "num_tokens": 26073597.0,
      "step": 14175
    },
    {
      "entropy": 5.182464361190796,
      "epoch": 1.1844303374540595,
      "grad_norm": 1.25,
      "learning_rate": 0.00048645549172012756,
      "loss": 4.9487,
      "mean_token_accuracy": 0.20975352376699447,
      "num_tokens": 26082026.0,
      "step": 14180
    },
    {
      "entropy": 5.057018089294433,
      "epoch": 1.1848479786167725,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000486445316903294,
      "loss": 4.8882,
      "mean_token_accuracy": 0.21007515192031861,
      "num_tokens": 26090112.0,
      "step": 14185
    },
    {
      "entropy": 5.1141139507293705,
      "epoch": 1.1852656197794855,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004864351383848556,
      "loss": 4.9078,
      "mean_token_accuracy": 0.2010620579123497,
      "num_tokens": 26098551.0,
      "step": 14190
    },
    {
      "entropy": 5.235504007339477,
      "epoch": 1.1856832609421986,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004864249561649906,
      "loss": 5.0191,
      "mean_token_accuracy": 0.1924814134836197,
      "num_tokens": 26109201.0,
      "step": 14195
    },
    {
      "entropy": 5.224775075912476,
      "epoch": 1.1861009021049114,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004864147702438773,
      "loss": 4.974,
      "mean_token_accuracy": 0.20196036696434022,
      "num_tokens": 26119893.0,
      "step": 14200
    },
    {
      "entropy": 5.107402753829956,
      "epoch": 1.1865185432676244,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000486404580621694,
      "loss": 4.9474,
      "mean_token_accuracy": 0.20614732950925826,
      "num_tokens": 26128341.0,
      "step": 14205
    },
    {
      "entropy": 5.134731912612915,
      "epoch": 1.1869361844303374,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000486394387298619,
      "loss": 4.9713,
      "mean_token_accuracy": 0.19626276344060897,
      "num_tokens": 26137823.0,
      "step": 14210
    },
    {
      "entropy": 5.192541074752808,
      "epoch": 1.1873538255930505,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048638419027483075,
      "loss": 4.9945,
      "mean_token_accuracy": 0.2004741072654724,
      "num_tokens": 26147333.0,
      "step": 14215
    },
    {
      "entropy": 5.1548408508300785,
      "epoch": 1.1877714667557635,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004863739895505079,
      "loss": 4.9263,
      "mean_token_accuracy": 0.20980468690395354,
      "num_tokens": 26156145.0,
      "step": 14220
    },
    {
      "entropy": 5.198816776275635,
      "epoch": 1.1881891079184765,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048636378512582884,
      "loss": 4.8684,
      "mean_token_accuracy": 0.20338238626718522,
      "num_tokens": 26165323.0,
      "step": 14225
    },
    {
      "entropy": 5.147923326492309,
      "epoch": 1.1886067490811894,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048635357700097236,
      "loss": 4.9257,
      "mean_token_accuracy": 0.20748998522758483,
      "num_tokens": 26174549.0,
      "step": 14230
    },
    {
      "entropy": 4.985383462905884,
      "epoch": 1.1890243902439024,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004863433651761171,
      "loss": 4.8347,
      "mean_token_accuracy": 0.21116563230752944,
      "num_tokens": 26184012.0,
      "step": 14235
    },
    {
      "entropy": 5.181051349639892,
      "epoch": 1.1894420314066154,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004863331496514418,
      "loss": 5.0466,
      "mean_token_accuracy": 0.18646551668643951,
      "num_tokens": 26193819.0,
      "step": 14240
    },
    {
      "entropy": 5.129006671905517,
      "epoch": 1.1898596725693285,
      "grad_norm": 1.25,
      "learning_rate": 0.00048632293042712545,
      "loss": 4.9099,
      "mean_token_accuracy": 0.2062527820467949,
      "num_tokens": 26202209.0,
      "step": 14245
    },
    {
      "entropy": 5.087129402160644,
      "epoch": 1.1902773137320415,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048631270750334674,
      "loss": 4.8921,
      "mean_token_accuracy": 0.20567490607500077,
      "num_tokens": 26212048.0,
      "step": 14250
    },
    {
      "entropy": 5.237024164199829,
      "epoch": 1.1906949548947545,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048630248088028474,
      "loss": 5.0282,
      "mean_token_accuracy": 0.19961101561784744,
      "num_tokens": 26220877.0,
      "step": 14255
    },
    {
      "entropy": 5.234298276901245,
      "epoch": 1.1911125960574673,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004862922505581185,
      "loss": 4.9605,
      "mean_token_accuracy": 0.2057940199971199,
      "num_tokens": 26229583.0,
      "step": 14260
    },
    {
      "entropy": 5.241326427459716,
      "epoch": 1.1915302372201804,
      "grad_norm": 1.25,
      "learning_rate": 0.000486282016537027,
      "loss": 5.052,
      "mean_token_accuracy": 0.20427430123090745,
      "num_tokens": 26239483.0,
      "step": 14265
    },
    {
      "entropy": 5.1977497100830075,
      "epoch": 1.1919478783828934,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048627177881718954,
      "loss": 5.0207,
      "mean_token_accuracy": 0.19546127915382386,
      "num_tokens": 26248608.0,
      "step": 14270
    },
    {
      "entropy": 5.127264213562012,
      "epoch": 1.1923655195456064,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004862615373987853,
      "loss": 4.8983,
      "mean_token_accuracy": 0.20256271958351135,
      "num_tokens": 26258037.0,
      "step": 14275
    },
    {
      "entropy": 5.24695520401001,
      "epoch": 1.1927831607083195,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004862512922819935,
      "loss": 5.1421,
      "mean_token_accuracy": 0.18896501809358596,
      "num_tokens": 26267890.0,
      "step": 14280
    },
    {
      "entropy": 5.175356864929199,
      "epoch": 1.1932008018710325,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004862410434669936,
      "loss": 4.9439,
      "mean_token_accuracy": 0.1990038812160492,
      "num_tokens": 26276963.0,
      "step": 14285
    },
    {
      "entropy": 5.201987361907959,
      "epoch": 1.1936184430337453,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048623079095396496,
      "loss": 4.926,
      "mean_token_accuracy": 0.2005222484469414,
      "num_tokens": 26286667.0,
      "step": 14290
    },
    {
      "entropy": 5.190170335769653,
      "epoch": 1.1940360841964583,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000486220534743087,
      "loss": 4.9342,
      "mean_token_accuracy": 0.20352500528097153,
      "num_tokens": 26295575.0,
      "step": 14295
    },
    {
      "entropy": 5.096679067611694,
      "epoch": 1.1944537253591714,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004862102748345394,
      "loss": 4.9524,
      "mean_token_accuracy": 0.20471881926059723,
      "num_tokens": 26305813.0,
      "step": 14300
    },
    {
      "entropy": 5.11940631866455,
      "epoch": 1.1948713665218844,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048620001122850164,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21031437814235687,
      "num_tokens": 26314744.0,
      "step": 14305
    },
    {
      "entropy": 5.271763324737549,
      "epoch": 1.1952890076845974,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004861897439251535,
      "loss": 5.1011,
      "mean_token_accuracy": 0.18897767961025239,
      "num_tokens": 26324469.0,
      "step": 14310
    },
    {
      "entropy": 5.204616355895996,
      "epoch": 1.1957066488473105,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048617947292467473,
      "loss": 4.9616,
      "mean_token_accuracy": 0.20237270146608352,
      "num_tokens": 26333850.0,
      "step": 14315
    },
    {
      "entropy": 5.114865112304687,
      "epoch": 1.1961242900100233,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048616919822724506,
      "loss": 4.8416,
      "mean_token_accuracy": 0.21607604920864104,
      "num_tokens": 26342327.0,
      "step": 14320
    },
    {
      "entropy": 5.1408744812011715,
      "epoch": 1.1965419311727363,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048615891983304436,
      "loss": 5.0325,
      "mean_token_accuracy": 0.20221305042505264,
      "num_tokens": 26351162.0,
      "step": 14325
    },
    {
      "entropy": 5.137059593200684,
      "epoch": 1.1969595723354494,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004861486377422526,
      "loss": 4.9544,
      "mean_token_accuracy": 0.2003741055727005,
      "num_tokens": 26361452.0,
      "step": 14330
    },
    {
      "entropy": 5.185440540313721,
      "epoch": 1.1973772134981624,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004861383519550498,
      "loss": 4.9173,
      "mean_token_accuracy": 0.21113307178020477,
      "num_tokens": 26370646.0,
      "step": 14335
    },
    {
      "entropy": 5.263523054122925,
      "epoch": 1.1977948546608754,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048612806247161594,
      "loss": 5.1105,
      "mean_token_accuracy": 0.18975039422512055,
      "num_tokens": 26379829.0,
      "step": 14340
    },
    {
      "entropy": 5.186210632324219,
      "epoch": 1.1982124958235885,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004861177692921313,
      "loss": 5.0339,
      "mean_token_accuracy": 0.19998425096273423,
      "num_tokens": 26389578.0,
      "step": 14345
    },
    {
      "entropy": 5.2332946300506595,
      "epoch": 1.1986301369863013,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048610747241677595,
      "loss": 4.932,
      "mean_token_accuracy": 0.20971637666225434,
      "num_tokens": 26398586.0,
      "step": 14350
    },
    {
      "entropy": 5.183649253845215,
      "epoch": 1.1990477781490143,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004860971718457302,
      "loss": 4.9731,
      "mean_token_accuracy": 0.2007386326789856,
      "num_tokens": 26407379.0,
      "step": 14355
    },
    {
      "entropy": 5.119349193572998,
      "epoch": 1.1994654193117273,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004860868675791743,
      "loss": 4.8488,
      "mean_token_accuracy": 0.21587797105312348,
      "num_tokens": 26416489.0,
      "step": 14360
    },
    {
      "entropy": 5.182852029800415,
      "epoch": 1.1998830604744404,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048607655961728876,
      "loss": 4.9818,
      "mean_token_accuracy": 0.19983260035514833,
      "num_tokens": 26424768.0,
      "step": 14365
    },
    {
      "entropy": 5.20806565284729,
      "epoch": 1.2003007016371534,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000486066247960254,
      "loss": 4.9199,
      "mean_token_accuracy": 0.19932166039943694,
      "num_tokens": 26435175.0,
      "step": 14370
    },
    {
      "entropy": 5.070564317703247,
      "epoch": 1.2007183427998664,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048605593260825044,
      "loss": 4.857,
      "mean_token_accuracy": 0.2043012872338295,
      "num_tokens": 26443492.0,
      "step": 14375
    },
    {
      "entropy": 5.14607048034668,
      "epoch": 1.2011359839625793,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048604561356145875,
      "loss": 4.8979,
      "mean_token_accuracy": 0.21427195966243745,
      "num_tokens": 26452239.0,
      "step": 14380
    },
    {
      "entropy": 5.147463321685791,
      "epoch": 1.2015536251252923,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048603529082005955,
      "loss": 5.0522,
      "mean_token_accuracy": 0.20564057379961015,
      "num_tokens": 26461010.0,
      "step": 14385
    },
    {
      "entropy": 5.0993140697479244,
      "epoch": 1.2019712662880053,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048602496438423354,
      "loss": 4.9301,
      "mean_token_accuracy": 0.20684233158826829,
      "num_tokens": 26470762.0,
      "step": 14390
    },
    {
      "entropy": 5.101467227935791,
      "epoch": 1.2023889074507184,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004860146342541616,
      "loss": 4.8815,
      "mean_token_accuracy": 0.21211108416318894,
      "num_tokens": 26479592.0,
      "step": 14395
    },
    {
      "entropy": 5.085744953155517,
      "epoch": 1.2028065486134314,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004860043004300244,
      "loss": 4.8834,
      "mean_token_accuracy": 0.208307246863842,
      "num_tokens": 26488245.0,
      "step": 14400
    },
    {
      "entropy": 5.1721954345703125,
      "epoch": 1.2032241897761444,
      "grad_norm": 1.28125,
      "learning_rate": 0.000485993962912003,
      "loss": 4.9457,
      "mean_token_accuracy": 0.20525965094566345,
      "num_tokens": 26497039.0,
      "step": 14405
    },
    {
      "entropy": 5.156922149658203,
      "epoch": 1.2036418309388572,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004859836217002783,
      "loss": 4.97,
      "mean_token_accuracy": 0.20513388216495515,
      "num_tokens": 26506116.0,
      "step": 14410
    },
    {
      "entropy": 5.112368154525757,
      "epoch": 1.2040594721015703,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004859732767950313,
      "loss": 4.8143,
      "mean_token_accuracy": 0.21666593849658966,
      "num_tokens": 26514309.0,
      "step": 14415
    },
    {
      "entropy": 5.182245302200317,
      "epoch": 1.2044771132642833,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048596292819644315,
      "loss": 4.9604,
      "mean_token_accuracy": 0.20693174600601197,
      "num_tokens": 26523425.0,
      "step": 14420
    },
    {
      "entropy": 5.180951833724976,
      "epoch": 1.2048947544269963,
      "grad_norm": 1.25,
      "learning_rate": 0.000485952575904695,
      "loss": 5.0656,
      "mean_token_accuracy": 0.20073693096637726,
      "num_tokens": 26532723.0,
      "step": 14425
    },
    {
      "entropy": 5.169610214233399,
      "epoch": 1.2053123955897094,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048594221991996814,
      "loss": 4.9855,
      "mean_token_accuracy": 0.2023228272795677,
      "num_tokens": 26541933.0,
      "step": 14430
    },
    {
      "entropy": 5.251264524459839,
      "epoch": 1.2057300367524224,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048593186024244373,
      "loss": 5.083,
      "mean_token_accuracy": 0.1960674747824669,
      "num_tokens": 26552071.0,
      "step": 14435
    },
    {
      "entropy": 5.256617307662964,
      "epoch": 1.2061476779151352,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004859214968723033,
      "loss": 5.0874,
      "mean_token_accuracy": 0.1978855237364769,
      "num_tokens": 26561470.0,
      "step": 14440
    },
    {
      "entropy": 5.191929960250855,
      "epoch": 1.2065653190778483,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048591112980972813,
      "loss": 4.9154,
      "mean_token_accuracy": 0.21403641104698182,
      "num_tokens": 26570448.0,
      "step": 14445
    },
    {
      "entropy": 5.2367161273956295,
      "epoch": 1.2069829602405613,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004859007590548997,
      "loss": 4.9865,
      "mean_token_accuracy": 0.20111777037382125,
      "num_tokens": 26579206.0,
      "step": 14450
    },
    {
      "entropy": 5.150507211685181,
      "epoch": 1.2074006014032743,
      "grad_norm": 1.25,
      "learning_rate": 0.00048589038460799974,
      "loss": 4.9703,
      "mean_token_accuracy": 0.2019663095474243,
      "num_tokens": 26587955.0,
      "step": 14455
    },
    {
      "entropy": 5.155042695999145,
      "epoch": 1.2078182425659874,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004858800064692097,
      "loss": 4.9539,
      "mean_token_accuracy": 0.202371346950531,
      "num_tokens": 26597635.0,
      "step": 14460
    },
    {
      "entropy": 5.177419281005859,
      "epoch": 1.2082358837287004,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048586962463871124,
      "loss": 4.907,
      "mean_token_accuracy": 0.21092153936624528,
      "num_tokens": 26606245.0,
      "step": 14465
    },
    {
      "entropy": 5.238356590270996,
      "epoch": 1.2086535248914134,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048585923911668625,
      "loss": 5.013,
      "mean_token_accuracy": 0.19515984505414963,
      "num_tokens": 26614464.0,
      "step": 14470
    },
    {
      "entropy": 5.178532886505127,
      "epoch": 1.2090711660541262,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048584884990331644,
      "loss": 5.0089,
      "mean_token_accuracy": 0.2051496610045433,
      "num_tokens": 26623436.0,
      "step": 14475
    },
    {
      "entropy": 5.1742331981658936,
      "epoch": 1.2094888072168393,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048583845699878375,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2055835485458374,
      "num_tokens": 26632415.0,
      "step": 14480
    },
    {
      "entropy": 5.111305189132691,
      "epoch": 1.2099064483795523,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004858280604032701,
      "loss": 4.846,
      "mean_token_accuracy": 0.21569220274686812,
      "num_tokens": 26641618.0,
      "step": 14485
    },
    {
      "entropy": 5.2370934009552,
      "epoch": 1.2103240895422653,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048581766011695745,
      "loss": 5.065,
      "mean_token_accuracy": 0.20003195405006408,
      "num_tokens": 26651290.0,
      "step": 14490
    },
    {
      "entropy": 5.185987186431885,
      "epoch": 1.2107417307049784,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048580725614002785,
      "loss": 4.941,
      "mean_token_accuracy": 0.20754511803388595,
      "num_tokens": 26660945.0,
      "step": 14495
    },
    {
      "entropy": 5.172185754776001,
      "epoch": 1.2111593718676912,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048579684847266353,
      "loss": 4.9387,
      "mean_token_accuracy": 0.2089068904519081,
      "num_tokens": 26670221.0,
      "step": 14500
    },
    {
      "entropy": 5.14262433052063,
      "epoch": 1.2115770130304042,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004857864371150467,
      "loss": 4.8822,
      "mean_token_accuracy": 0.20941120386123657,
      "num_tokens": 26679074.0,
      "step": 14505
    },
    {
      "entropy": 5.225486946105957,
      "epoch": 1.2119946541931172,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048577602206735943,
      "loss": 5.051,
      "mean_token_accuracy": 0.1984088972210884,
      "num_tokens": 26689124.0,
      "step": 14510
    },
    {
      "entropy": 5.06153564453125,
      "epoch": 1.2124122953558303,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004857656033297843,
      "loss": 4.8075,
      "mean_token_accuracy": 0.21065133512020112,
      "num_tokens": 26698421.0,
      "step": 14515
    },
    {
      "entropy": 5.161602735519409,
      "epoch": 1.2128299365185433,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048575518090250347,
      "loss": 4.9105,
      "mean_token_accuracy": 0.20899197161197663,
      "num_tokens": 26706432.0,
      "step": 14520
    },
    {
      "entropy": 5.173313570022583,
      "epoch": 1.2132475776812564,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004857447547856996,
      "loss": 4.9489,
      "mean_token_accuracy": 0.20769215524196624,
      "num_tokens": 26715224.0,
      "step": 14525
    },
    {
      "entropy": 5.179033422470093,
      "epoch": 1.2136652188439694,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004857343249795552,
      "loss": 4.9502,
      "mean_token_accuracy": 0.20486061722040178,
      "num_tokens": 26724311.0,
      "step": 14530
    },
    {
      "entropy": 5.149508476257324,
      "epoch": 1.2140828600066822,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004857238914842526,
      "loss": 4.9822,
      "mean_token_accuracy": 0.1939324215054512,
      "num_tokens": 26734012.0,
      "step": 14535
    },
    {
      "entropy": 5.1732506275177,
      "epoch": 1.2145005011693952,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048571345429997473,
      "loss": 4.8799,
      "mean_token_accuracy": 0.2117844045162201,
      "num_tokens": 26742745.0,
      "step": 14540
    },
    {
      "entropy": 5.135331392288208,
      "epoch": 1.2149181423321083,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048570301342690416,
      "loss": 4.9194,
      "mean_token_accuracy": 0.21378224492073059,
      "num_tokens": 26751856.0,
      "step": 14545
    },
    {
      "entropy": 5.028673887252808,
      "epoch": 1.2153357834948213,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004856925688652238,
      "loss": 4.8678,
      "mean_token_accuracy": 0.2136292040348053,
      "num_tokens": 26761359.0,
      "step": 14550
    },
    {
      "entropy": 5.112123823165893,
      "epoch": 1.2157534246575343,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048568212061511634,
      "loss": 4.8615,
      "mean_token_accuracy": 0.20854734778404235,
      "num_tokens": 26770588.0,
      "step": 14555
    },
    {
      "entropy": 5.222061347961426,
      "epoch": 1.2161710658202471,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048567166867676475,
      "loss": 4.9905,
      "mean_token_accuracy": 0.2023475781083107,
      "num_tokens": 26780637.0,
      "step": 14560
    },
    {
      "entropy": 5.13651556968689,
      "epoch": 1.2165887069829602,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048566121305035207,
      "loss": 4.9003,
      "mean_token_accuracy": 0.21043167859315873,
      "num_tokens": 26789387.0,
      "step": 14565
    },
    {
      "entropy": 5.188025760650635,
      "epoch": 1.2170063481456732,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048565075373606126,
      "loss": 4.9884,
      "mean_token_accuracy": 0.20440658628940583,
      "num_tokens": 26798677.0,
      "step": 14570
    },
    {
      "entropy": 5.21405258178711,
      "epoch": 1.2174239893083862,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048564029073407545,
      "loss": 4.9732,
      "mean_token_accuracy": 0.20520322769880295,
      "num_tokens": 26807457.0,
      "step": 14575
    },
    {
      "entropy": 5.198657894134522,
      "epoch": 1.2178416304710993,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004856298240445778,
      "loss": 4.952,
      "mean_token_accuracy": 0.20507543981075288,
      "num_tokens": 26816832.0,
      "step": 14580
    },
    {
      "entropy": 5.118776416778564,
      "epoch": 1.2182592716338123,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004856193536677515,
      "loss": 4.9254,
      "mean_token_accuracy": 0.20378464311361313,
      "num_tokens": 26826534.0,
      "step": 14585
    },
    {
      "entropy": 5.180170154571533,
      "epoch": 1.2186769127965253,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048560887960377995,
      "loss": 5.0413,
      "mean_token_accuracy": 0.20114791095256807,
      "num_tokens": 26835981.0,
      "step": 14590
    },
    {
      "entropy": 5.164838790893555,
      "epoch": 1.2190945539592382,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048559840185284647,
      "loss": 4.8785,
      "mean_token_accuracy": 0.20781095772981645,
      "num_tokens": 26845321.0,
      "step": 14595
    },
    {
      "entropy": 5.135073566436768,
      "epoch": 1.2195121951219512,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048558792041513446,
      "loss": 4.8284,
      "mean_token_accuracy": 0.20998603403568267,
      "num_tokens": 26853866.0,
      "step": 14600
    },
    {
      "entropy": 5.093303155899048,
      "epoch": 1.2199298362846642,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004855774352908273,
      "loss": 4.841,
      "mean_token_accuracy": 0.20932035595178605,
      "num_tokens": 26863581.0,
      "step": 14605
    },
    {
      "entropy": 5.1862140655517575,
      "epoch": 1.2203474774473773,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004855669464801088,
      "loss": 4.9504,
      "mean_token_accuracy": 0.20711832493543625,
      "num_tokens": 26873481.0,
      "step": 14610
    },
    {
      "entropy": 5.227762126922608,
      "epoch": 1.2207651186100903,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048555645398316237,
      "loss": 5.0736,
      "mean_token_accuracy": 0.19968123137950897,
      "num_tokens": 26882222.0,
      "step": 14615
    },
    {
      "entropy": 5.205530405044556,
      "epoch": 1.221182759772803,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004855459578001718,
      "loss": 4.9955,
      "mean_token_accuracy": 0.1995491400361061,
      "num_tokens": 26891305.0,
      "step": 14620
    },
    {
      "entropy": 5.178498888015747,
      "epoch": 1.2216004009355161,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004855354579313208,
      "loss": 4.9353,
      "mean_token_accuracy": 0.20364181101322174,
      "num_tokens": 26901759.0,
      "step": 14625
    },
    {
      "entropy": 5.218368148803711,
      "epoch": 1.2220180420982292,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004855249543767932,
      "loss": 5.005,
      "mean_token_accuracy": 0.19963036924600602,
      "num_tokens": 26911183.0,
      "step": 14630
    },
    {
      "entropy": 5.192747640609741,
      "epoch": 1.2224356832609422,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048551444713677285,
      "loss": 5.0054,
      "mean_token_accuracy": 0.1953968659043312,
      "num_tokens": 26921185.0,
      "step": 14635
    },
    {
      "entropy": 5.154981517791748,
      "epoch": 1.2228533244236552,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048550393621144374,
      "loss": 4.8211,
      "mean_token_accuracy": 0.21093014776706695,
      "num_tokens": 26930034.0,
      "step": 14640
    },
    {
      "entropy": 5.105802536010742,
      "epoch": 1.2232709655863683,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004854934216009898,
      "loss": 4.8898,
      "mean_token_accuracy": 0.21005679219961165,
      "num_tokens": 26939273.0,
      "step": 14645
    },
    {
      "entropy": 5.125222253799438,
      "epoch": 1.2236886067490813,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004854829033055951,
      "loss": 4.9156,
      "mean_token_accuracy": 0.20484028160572051,
      "num_tokens": 26948924.0,
      "step": 14650
    },
    {
      "entropy": 5.115103101730346,
      "epoch": 1.2241062479117941,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004854723813254439,
      "loss": 4.8273,
      "mean_token_accuracy": 0.20963678658008575,
      "num_tokens": 26957492.0,
      "step": 14655
    },
    {
      "entropy": 5.166096878051758,
      "epoch": 1.2245238890745072,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004854618556607203,
      "loss": 5.0377,
      "mean_token_accuracy": 0.19760544300079347,
      "num_tokens": 26967114.0,
      "step": 14660
    },
    {
      "entropy": 5.290770149230957,
      "epoch": 1.2249415302372202,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048545132631160856,
      "loss": 4.9942,
      "mean_token_accuracy": 0.1987449496984482,
      "num_tokens": 26976661.0,
      "step": 14665
    },
    {
      "entropy": 5.227971982955933,
      "epoch": 1.2253591713999332,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004854407932782931,
      "loss": 5.0161,
      "mean_token_accuracy": 0.20316352993249892,
      "num_tokens": 26986044.0,
      "step": 14670
    },
    {
      "entropy": 5.11852126121521,
      "epoch": 1.2257768125626463,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004854302565609582,
      "loss": 4.9945,
      "mean_token_accuracy": 0.20548736304044724,
      "num_tokens": 26995166.0,
      "step": 14675
    },
    {
      "entropy": 5.170224046707153,
      "epoch": 1.226194453725359,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048541971615978833,
      "loss": 4.9704,
      "mean_token_accuracy": 0.19872014224529266,
      "num_tokens": 27004245.0,
      "step": 14680
    },
    {
      "entropy": 5.228067350387573,
      "epoch": 1.226612094888072,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048540917207496806,
      "loss": 5.0647,
      "mean_token_accuracy": 0.19724346250295638,
      "num_tokens": 27013278.0,
      "step": 14685
    },
    {
      "entropy": 5.194910383224487,
      "epoch": 1.2270297360507851,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048539862430668204,
      "loss": 4.9042,
      "mean_token_accuracy": 0.20330731719732284,
      "num_tokens": 27022387.0,
      "step": 14690
    },
    {
      "entropy": 5.1644142150878904,
      "epoch": 1.2274473772134982,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048538807285511474,
      "loss": 4.9458,
      "mean_token_accuracy": 0.2004190430045128,
      "num_tokens": 27031760.0,
      "step": 14695
    },
    {
      "entropy": 5.20043683052063,
      "epoch": 1.2278650183762112,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048537751772045103,
      "loss": 4.9103,
      "mean_token_accuracy": 0.21603277027606965,
      "num_tokens": 27040908.0,
      "step": 14700
    },
    {
      "entropy": 5.181299161911011,
      "epoch": 1.2282826595389242,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004853669589028757,
      "loss": 4.9265,
      "mean_token_accuracy": 0.20820981860160828,
      "num_tokens": 27049989.0,
      "step": 14705
    },
    {
      "entropy": 5.098298692703247,
      "epoch": 1.2287003007016373,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004853563964025734,
      "loss": 4.9144,
      "mean_token_accuracy": 0.20996073633432388,
      "num_tokens": 27058706.0,
      "step": 14710
    },
    {
      "entropy": 5.096966505050659,
      "epoch": 1.22911794186435,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048534583021972926,
      "loss": 4.892,
      "mean_token_accuracy": 0.20566961914300919,
      "num_tokens": 27067604.0,
      "step": 14715
    },
    {
      "entropy": 5.153396654129028,
      "epoch": 1.2295355830270631,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004853352603545281,
      "loss": 4.9365,
      "mean_token_accuracy": 0.20968307703733444,
      "num_tokens": 27076489.0,
      "step": 14720
    },
    {
      "entropy": 5.1602373123168945,
      "epoch": 1.2299532241897762,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004853246868071551,
      "loss": 4.9111,
      "mean_token_accuracy": 0.20618190318346025,
      "num_tokens": 27085750.0,
      "step": 14725
    },
    {
      "entropy": 5.150932931900025,
      "epoch": 1.2303708653524892,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048531410957779533,
      "loss": 4.9084,
      "mean_token_accuracy": 0.20786479860544205,
      "num_tokens": 27095509.0,
      "step": 14730
    },
    {
      "entropy": 5.09764552116394,
      "epoch": 1.2307885065152022,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004853035286666338,
      "loss": 4.9186,
      "mean_token_accuracy": 0.20642659664154053,
      "num_tokens": 27105054.0,
      "step": 14735
    },
    {
      "entropy": 5.199010229110717,
      "epoch": 1.231206147677915,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000485292944073856,
      "loss": 5.0302,
      "mean_token_accuracy": 0.2037755846977234,
      "num_tokens": 27113497.0,
      "step": 14740
    },
    {
      "entropy": 5.2527931213378904,
      "epoch": 1.231623788840628,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000485282355799647,
      "loss": 5.0051,
      "mean_token_accuracy": 0.2051698848605156,
      "num_tokens": 27122718.0,
      "step": 14745
    },
    {
      "entropy": 5.16599817276001,
      "epoch": 1.232041430003341,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048527176384419217,
      "loss": 4.9304,
      "mean_token_accuracy": 0.20960622280836105,
      "num_tokens": 27132361.0,
      "step": 14750
    },
    {
      "entropy": 5.2240269660949705,
      "epoch": 1.2324590711660541,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048526116820767713,
      "loss": 4.9237,
      "mean_token_accuracy": 0.20689452886581422,
      "num_tokens": 27141430.0,
      "step": 14755
    },
    {
      "entropy": 5.243683576583862,
      "epoch": 1.2328767123287672,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004852505688902872,
      "loss": 5.0129,
      "mean_token_accuracy": 0.20217214971780778,
      "num_tokens": 27150154.0,
      "step": 14760
    },
    {
      "entropy": 5.149538993835449,
      "epoch": 1.2332943534914802,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004852399658922079,
      "loss": 4.9632,
      "mean_token_accuracy": 0.20948652625083924,
      "num_tokens": 27159314.0,
      "step": 14765
    },
    {
      "entropy": 5.210212755203247,
      "epoch": 1.2337119946541932,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004852293592136251,
      "loss": 4.9656,
      "mean_token_accuracy": 0.197274249792099,
      "num_tokens": 27167537.0,
      "step": 14770
    },
    {
      "entropy": 5.248903226852417,
      "epoch": 1.234129635816906,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004852187488547242,
      "loss": 4.9892,
      "mean_token_accuracy": 0.19964949935674667,
      "num_tokens": 27177037.0,
      "step": 14775
    },
    {
      "entropy": 5.196665191650391,
      "epoch": 1.234547276979619,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048520813481569116,
      "loss": 5.0221,
      "mean_token_accuracy": 0.1987486720085144,
      "num_tokens": 27186498.0,
      "step": 14780
    },
    {
      "entropy": 5.154570627212524,
      "epoch": 1.2349649181423321,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004851975170967116,
      "loss": 4.8659,
      "mean_token_accuracy": 0.21129745692014695,
      "num_tokens": 27195170.0,
      "step": 14785
    },
    {
      "entropy": 5.167276287078858,
      "epoch": 1.2353825593050451,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004851868956979715,
      "loss": 4.9015,
      "mean_token_accuracy": 0.20844382494688035,
      "num_tokens": 27204756.0,
      "step": 14790
    },
    {
      "entropy": 4.999532985687256,
      "epoch": 1.2358002004677582,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004851762706196568,
      "loss": 4.8781,
      "mean_token_accuracy": 0.20051447302103043,
      "num_tokens": 27213607.0,
      "step": 14795
    },
    {
      "entropy": 5.167258262634277,
      "epoch": 1.236217841630471,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048516564186195353,
      "loss": 4.9575,
      "mean_token_accuracy": 0.2076592355966568,
      "num_tokens": 27222887.0,
      "step": 14800
    },
    {
      "entropy": 5.231129741668701,
      "epoch": 1.236635482793184,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048515500942504763,
      "loss": 4.976,
      "mean_token_accuracy": 0.2033357560634613,
      "num_tokens": 27231374.0,
      "step": 14805
    },
    {
      "entropy": 5.2369263648986815,
      "epoch": 1.237053123955897,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048514437330912544,
      "loss": 5.0387,
      "mean_token_accuracy": 0.19605533480644227,
      "num_tokens": 27239929.0,
      "step": 14810
    },
    {
      "entropy": 5.165923738479615,
      "epoch": 1.23747076511861,
      "grad_norm": 1.5,
      "learning_rate": 0.00048513373351437294,
      "loss": 4.9247,
      "mean_token_accuracy": 0.20035431236028672,
      "num_tokens": 27249789.0,
      "step": 14815
    },
    {
      "entropy": 5.170178461074829,
      "epoch": 1.2378884062813231,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048512309004097653,
      "loss": 4.9381,
      "mean_token_accuracy": 0.20190029591321945,
      "num_tokens": 27259773.0,
      "step": 14820
    },
    {
      "entropy": 5.29471230506897,
      "epoch": 1.2383060474440362,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004851124428891226,
      "loss": 5.073,
      "mean_token_accuracy": 0.19736886024475098,
      "num_tokens": 27268494.0,
      "step": 14825
    },
    {
      "entropy": 5.220165205001831,
      "epoch": 1.2387236886067492,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004851017920589973,
      "loss": 4.9622,
      "mean_token_accuracy": 0.2003882810473442,
      "num_tokens": 27276914.0,
      "step": 14830
    },
    {
      "entropy": 5.129614210128784,
      "epoch": 1.239141329769462,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004850911375507873,
      "loss": 4.9694,
      "mean_token_accuracy": 0.20690634250640869,
      "num_tokens": 27286091.0,
      "step": 14835
    },
    {
      "entropy": 5.210960721969604,
      "epoch": 1.239558970932175,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048508047936467907,
      "loss": 4.9972,
      "mean_token_accuracy": 0.2051364302635193,
      "num_tokens": 27295201.0,
      "step": 14840
    },
    {
      "entropy": 5.228882312774658,
      "epoch": 1.239976612094888,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048506981750085914,
      "loss": 5.0369,
      "mean_token_accuracy": 0.2096265971660614,
      "num_tokens": 27303900.0,
      "step": 14845
    },
    {
      "entropy": 5.159246015548706,
      "epoch": 1.240394253257601,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048505915195951423,
      "loss": 4.9386,
      "mean_token_accuracy": 0.20610670000314713,
      "num_tokens": 27312613.0,
      "step": 14850
    },
    {
      "entropy": 5.145587539672851,
      "epoch": 1.2408118944203141,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048504848274083095,
      "loss": 4.9686,
      "mean_token_accuracy": 0.20542112737894058,
      "num_tokens": 27322047.0,
      "step": 14855
    },
    {
      "entropy": 5.280626583099365,
      "epoch": 1.241229535583027,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048503780984499625,
      "loss": 5.193,
      "mean_token_accuracy": 0.18792821019887923,
      "num_tokens": 27332163.0,
      "step": 14860
    },
    {
      "entropy": 5.268613195419311,
      "epoch": 1.24164717674574,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004850271332721968,
      "loss": 4.9933,
      "mean_token_accuracy": 0.19999616146087645,
      "num_tokens": 27343838.0,
      "step": 14865
    },
    {
      "entropy": 5.254788970947265,
      "epoch": 1.242064817908453,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004850164530226196,
      "loss": 4.9468,
      "mean_token_accuracy": 0.2130851849913597,
      "num_tokens": 27352362.0,
      "step": 14870
    },
    {
      "entropy": 5.178209066390991,
      "epoch": 1.242482459071166,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004850057690964515,
      "loss": 5.0498,
      "mean_token_accuracy": 0.19359085857868194,
      "num_tokens": 27361444.0,
      "step": 14875
    },
    {
      "entropy": 5.096546363830567,
      "epoch": 1.242900100233879,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048499508149387973,
      "loss": 4.8604,
      "mean_token_accuracy": 0.2037036493420601,
      "num_tokens": 27370872.0,
      "step": 14880
    },
    {
      "entropy": 5.145865154266358,
      "epoch": 1.2433177413965921,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048498439021509124,
      "loss": 4.9422,
      "mean_token_accuracy": 0.20644918829202652,
      "num_tokens": 27380628.0,
      "step": 14885
    },
    {
      "entropy": 5.210357427597046,
      "epoch": 1.2437353825593052,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004849736952602732,
      "loss": 4.8997,
      "mean_token_accuracy": 0.20614794343709947,
      "num_tokens": 27390187.0,
      "step": 14890
    },
    {
      "entropy": 5.2617920398712155,
      "epoch": 1.244153023722018,
      "grad_norm": 1.1875,
      "learning_rate": 0.000484962996629613,
      "loss": 5.1667,
      "mean_token_accuracy": 0.19085199534893035,
      "num_tokens": 27400307.0,
      "step": 14895
    },
    {
      "entropy": 5.233146238327026,
      "epoch": 1.244570664884731,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048495229432329775,
      "loss": 4.976,
      "mean_token_accuracy": 0.20135580897331237,
      "num_tokens": 27410060.0,
      "step": 14900
    },
    {
      "entropy": 5.219910287857056,
      "epoch": 1.244988306047444,
      "grad_norm": 1.25,
      "learning_rate": 0.00048494158834151485,
      "loss": 5.011,
      "mean_token_accuracy": 0.1999530240893364,
      "num_tokens": 27419119.0,
      "step": 14905
    },
    {
      "entropy": 5.182317018508911,
      "epoch": 1.245405947210157,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048493087868445176,
      "loss": 4.8978,
      "mean_token_accuracy": 0.21152960360050202,
      "num_tokens": 27428052.0,
      "step": 14910
    },
    {
      "entropy": 5.165277910232544,
      "epoch": 1.24582358837287,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048492016535229594,
      "loss": 4.8936,
      "mean_token_accuracy": 0.2063406303524971,
      "num_tokens": 27436418.0,
      "step": 14915
    },
    {
      "entropy": 5.179647541046142,
      "epoch": 1.246241229535583,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004849094483452349,
      "loss": 5.0616,
      "mean_token_accuracy": 0.20440271347761155,
      "num_tokens": 27445609.0,
      "step": 14920
    },
    {
      "entropy": 5.171057796478271,
      "epoch": 1.246658870698296,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004848987276634564,
      "loss": 5.0225,
      "mean_token_accuracy": 0.2003096967935562,
      "num_tokens": 27456250.0,
      "step": 14925
    },
    {
      "entropy": 5.197917842864991,
      "epoch": 1.247076511861009,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048488800330714795,
      "loss": 4.9383,
      "mean_token_accuracy": 0.19898485094308854,
      "num_tokens": 27466134.0,
      "step": 14930
    },
    {
      "entropy": 5.154957008361817,
      "epoch": 1.247494153023722,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048487727527649733,
      "loss": 4.9505,
      "mean_token_accuracy": 0.20494175255298613,
      "num_tokens": 27476097.0,
      "step": 14935
    },
    {
      "entropy": 5.1492591381073,
      "epoch": 1.247911794186435,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048486654357169246,
      "loss": 4.9449,
      "mean_token_accuracy": 0.2057437241077423,
      "num_tokens": 27485850.0,
      "step": 14940
    },
    {
      "entropy": 5.199787569046021,
      "epoch": 1.248329435349148,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004848558081929211,
      "loss": 4.9387,
      "mean_token_accuracy": 0.20572056770324706,
      "num_tokens": 27494169.0,
      "step": 14945
    },
    {
      "entropy": 5.152124214172363,
      "epoch": 1.2487470765118611,
      "grad_norm": 1.25,
      "learning_rate": 0.0004848450691403712,
      "loss": 4.9615,
      "mean_token_accuracy": 0.20099812746047974,
      "num_tokens": 27502599.0,
      "step": 14950
    },
    {
      "entropy": 5.055132341384888,
      "epoch": 1.249164717674574,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004848343264142307,
      "loss": 4.7231,
      "mean_token_accuracy": 0.22839573323726653,
      "num_tokens": 27511278.0,
      "step": 14955
    },
    {
      "entropy": 5.195808172225952,
      "epoch": 1.249582358837287,
      "grad_norm": 1.125,
      "learning_rate": 0.0004848235800146879,
      "loss": 4.9832,
      "mean_token_accuracy": 0.19828079342842103,
      "num_tokens": 27521040.0,
      "step": 14960
    },
    {
      "entropy": 5.190809535980224,
      "epoch": 1.25,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004848128299419306,
      "loss": 4.9781,
      "mean_token_accuracy": 0.1986464723944664,
      "num_tokens": 27530893.0,
      "step": 14965
    },
    {
      "entropy": 5.187432146072387,
      "epoch": 1.250417641162713,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048480207619614726,
      "loss": 4.9961,
      "mean_token_accuracy": 0.2017314299941063,
      "num_tokens": 27540401.0,
      "step": 14970
    },
    {
      "entropy": 5.081004905700683,
      "epoch": 1.250835282325426,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000484791318777526,
      "loss": 4.7966,
      "mean_token_accuracy": 0.21301444619894028,
      "num_tokens": 27549951.0,
      "step": 14975
    },
    {
      "entropy": 5.165010023117065,
      "epoch": 1.2512529234881389,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048478055768625517,
      "loss": 4.9409,
      "mean_token_accuracy": 0.2008342146873474,
      "num_tokens": 27559170.0,
      "step": 14980
    },
    {
      "entropy": 5.0890114307403564,
      "epoch": 1.251670564650852,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048476979292252323,
      "loss": 4.7998,
      "mean_token_accuracy": 0.21791396141052247,
      "num_tokens": 27567094.0,
      "step": 14985
    },
    {
      "entropy": 5.144490766525268,
      "epoch": 1.252088205813565,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004847590244865185,
      "loss": 4.9308,
      "mean_token_accuracy": 0.20808619409799575,
      "num_tokens": 27575141.0,
      "step": 14990
    },
    {
      "entropy": 5.185410737991333,
      "epoch": 1.252505846976278,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048474825237842964,
      "loss": 4.9797,
      "mean_token_accuracy": 0.20510129630565643,
      "num_tokens": 27584094.0,
      "step": 14995
    },
    {
      "entropy": 5.240324258804321,
      "epoch": 1.252923488138991,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000484737476598445,
      "loss": 5.078,
      "mean_token_accuracy": 0.1942172572016716,
      "num_tokens": 27593157.0,
      "step": 15000
    },
    {
      "epoch": 1.252923488138991,
      "eval_entropy": 4.972418664143616,
      "eval_loss": 4.976927280426025,
      "eval_mean_token_accuracy": 0.2142851393337637,
      "eval_num_tokens": 27593157.0,
      "eval_runtime": 27.6167,
      "eval_samples_per_second": 1367.399,
      "eval_steps_per_second": 170.948,
      "step": 15000
    },
    {
      "entropy": 5.172480583190918,
      "epoch": 1.253341129301704,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048472669714675357,
      "loss": 4.9736,
      "mean_token_accuracy": 0.20670702904462815,
      "num_tokens": 27601833.0,
      "step": 15005
    },
    {
      "entropy": 5.215618991851807,
      "epoch": 1.253758770464417,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048471591402354374,
      "loss": 4.9864,
      "mean_token_accuracy": 0.2062656745314598,
      "num_tokens": 27610969.0,
      "step": 15010
    },
    {
      "entropy": 5.148627281188965,
      "epoch": 1.25417641162713,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048470512722900434,
      "loss": 4.9347,
      "mean_token_accuracy": 0.2141784131526947,
      "num_tokens": 27620479.0,
      "step": 15015
    },
    {
      "entropy": 5.12378888130188,
      "epoch": 1.254594052789843,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048469433676332434,
      "loss": 4.9391,
      "mean_token_accuracy": 0.2096070185303688,
      "num_tokens": 27629558.0,
      "step": 15020
    },
    {
      "entropy": 5.218819236755371,
      "epoch": 1.255011693952556,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004846835426266926,
      "loss": 4.9061,
      "mean_token_accuracy": 0.2019507870078087,
      "num_tokens": 27639070.0,
      "step": 15025
    },
    {
      "entropy": 5.111401987075806,
      "epoch": 1.255429335115269,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004846727448192981,
      "loss": 4.8613,
      "mean_token_accuracy": 0.20690663307905197,
      "num_tokens": 27647839.0,
      "step": 15030
    },
    {
      "entropy": 5.149005174636841,
      "epoch": 1.255846976277982,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048466194334132964,
      "loss": 4.8749,
      "mean_token_accuracy": 0.2056342840194702,
      "num_tokens": 27656361.0,
      "step": 15035
    },
    {
      "entropy": 5.1054881572723385,
      "epoch": 1.2562646174406948,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048465113819297654,
      "loss": 4.8198,
      "mean_token_accuracy": 0.2073519542813301,
      "num_tokens": 27664783.0,
      "step": 15040
    },
    {
      "entropy": 5.174130868911743,
      "epoch": 1.2566822586034079,
      "grad_norm": 1.40625,
      "learning_rate": 0.000484640329374428,
      "loss": 4.919,
      "mean_token_accuracy": 0.20937900096178055,
      "num_tokens": 27673414.0,
      "step": 15045
    },
    {
      "entropy": 5.083298921585083,
      "epoch": 1.257099899766121,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048462951688587317,
      "loss": 4.9171,
      "mean_token_accuracy": 0.20891739279031754,
      "num_tokens": 27682304.0,
      "step": 15050
    },
    {
      "entropy": 5.167435789108277,
      "epoch": 1.257517540928834,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048461870072750117,
      "loss": 4.9876,
      "mean_token_accuracy": 0.1990368500351906,
      "num_tokens": 27691989.0,
      "step": 15055
    },
    {
      "entropy": 5.172483444213867,
      "epoch": 1.257935182091547,
      "grad_norm": 1.25,
      "learning_rate": 0.00048460788089950164,
      "loss": 4.9415,
      "mean_token_accuracy": 0.20058891028165818,
      "num_tokens": 27701201.0,
      "step": 15060
    },
    {
      "entropy": 5.216622066497803,
      "epoch": 1.25835282325426,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004845970574020637,
      "loss": 4.9682,
      "mean_token_accuracy": 0.19893250167369841,
      "num_tokens": 27710027.0,
      "step": 15065
    },
    {
      "entropy": 5.17182092666626,
      "epoch": 1.258770464416973,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004845862302353771,
      "loss": 4.9467,
      "mean_token_accuracy": 0.20222676694393157,
      "num_tokens": 27719180.0,
      "step": 15070
    },
    {
      "entropy": 5.0666579246521,
      "epoch": 1.2591881055796859,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048457539939963126,
      "loss": 4.9293,
      "mean_token_accuracy": 0.21089064031839372,
      "num_tokens": 27728560.0,
      "step": 15075
    },
    {
      "entropy": 5.1163383483886715,
      "epoch": 1.259605746742399,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048456456489501574,
      "loss": 4.8864,
      "mean_token_accuracy": 0.20969958752393722,
      "num_tokens": 27737660.0,
      "step": 15080
    },
    {
      "entropy": 5.138727235794067,
      "epoch": 1.260023387905112,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004845537267217202,
      "loss": 4.8795,
      "mean_token_accuracy": 0.21030199974775315,
      "num_tokens": 27746857.0,
      "step": 15085
    },
    {
      "entropy": 5.162420272827148,
      "epoch": 1.260441029067825,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004845428848799346,
      "loss": 4.9377,
      "mean_token_accuracy": 0.20588705390691758,
      "num_tokens": 27754492.0,
      "step": 15090
    },
    {
      "entropy": 5.102679061889648,
      "epoch": 1.260858670230538,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004845320393698484,
      "loss": 4.8432,
      "mean_token_accuracy": 0.2114707350730896,
      "num_tokens": 27762481.0,
      "step": 15095
    },
    {
      "entropy": 5.24645767211914,
      "epoch": 1.2612763113932508,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004845211901916517,
      "loss": 5.0891,
      "mean_token_accuracy": 0.19396835565567017,
      "num_tokens": 27772616.0,
      "step": 15100
    },
    {
      "entropy": 5.156600046157837,
      "epoch": 1.2616939525559638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048451033734553444,
      "loss": 4.9219,
      "mean_token_accuracy": 0.211461578309536,
      "num_tokens": 27781245.0,
      "step": 15105
    },
    {
      "entropy": 5.092680025100708,
      "epoch": 1.2621115937186769,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004844994808316864,
      "loss": 4.8885,
      "mean_token_accuracy": 0.2097795471549034,
      "num_tokens": 27789900.0,
      "step": 15110
    },
    {
      "entropy": 5.122084903717041,
      "epoch": 1.26252923488139,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004844886206502979,
      "loss": 4.8595,
      "mean_token_accuracy": 0.20764111578464509,
      "num_tokens": 27798405.0,
      "step": 15115
    },
    {
      "entropy": 5.121844959259033,
      "epoch": 1.262946876044103,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048447775680155884,
      "loss": 4.905,
      "mean_token_accuracy": 0.21103364676237107,
      "num_tokens": 27808727.0,
      "step": 15120
    },
    {
      "entropy": 5.137041425704956,
      "epoch": 1.263364517206816,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048446688928565956,
      "loss": 4.9544,
      "mean_token_accuracy": 0.20573309510946275,
      "num_tokens": 27817427.0,
      "step": 15125
    },
    {
      "entropy": 5.15964822769165,
      "epoch": 1.263782158369529,
      "grad_norm": 1.5,
      "learning_rate": 0.0004844560181027903,
      "loss": 4.9093,
      "mean_token_accuracy": 0.2007589280605316,
      "num_tokens": 27826160.0,
      "step": 15130
    },
    {
      "entropy": 5.145121908187866,
      "epoch": 1.2641997995322418,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004844451432531412,
      "loss": 4.9986,
      "mean_token_accuracy": 0.20236769020557405,
      "num_tokens": 27836045.0,
      "step": 15135
    },
    {
      "entropy": 5.1401451587677,
      "epoch": 1.2646174406949549,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004844342647369028,
      "loss": 4.9544,
      "mean_token_accuracy": 0.20828503519296646,
      "num_tokens": 27844995.0,
      "step": 15140
    },
    {
      "entropy": 5.120144701004028,
      "epoch": 1.2650350818576679,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048442338255426555,
      "loss": 4.9129,
      "mean_token_accuracy": 0.20336648523807527,
      "num_tokens": 27854554.0,
      "step": 15145
    },
    {
      "entropy": 5.206351375579834,
      "epoch": 1.265452723020381,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004844124967054199,
      "loss": 4.9645,
      "mean_token_accuracy": 0.20465039014816283,
      "num_tokens": 27864860.0,
      "step": 15150
    },
    {
      "entropy": 5.25334062576294,
      "epoch": 1.265870364183094,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004844016071905564,
      "loss": 5.1129,
      "mean_token_accuracy": 0.19819784462451934,
      "num_tokens": 27875614.0,
      "step": 15155
    },
    {
      "entropy": 5.149796485900879,
      "epoch": 1.2662880053458068,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004843907140098656,
      "loss": 4.9277,
      "mean_token_accuracy": 0.21064305752515794,
      "num_tokens": 27884424.0,
      "step": 15160
    },
    {
      "entropy": 5.141860914230347,
      "epoch": 1.2667056465085198,
      "grad_norm": 2.25,
      "learning_rate": 0.0004843798171635384,
      "loss": 4.7859,
      "mean_token_accuracy": 0.2145609974861145,
      "num_tokens": 27892916.0,
      "step": 15165
    },
    {
      "entropy": 5.197795009613037,
      "epoch": 1.2671232876712328,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004843689166517655,
      "loss": 5.0668,
      "mean_token_accuracy": 0.20023233890533448,
      "num_tokens": 27902902.0,
      "step": 15170
    },
    {
      "entropy": 5.131577777862549,
      "epoch": 1.2675409288339459,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004843580124747377,
      "loss": 5.037,
      "mean_token_accuracy": 0.19589757919311523,
      "num_tokens": 27912650.0,
      "step": 15175
    },
    {
      "entropy": 5.255009174346924,
      "epoch": 1.267958569996659,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048434710463264585,
      "loss": 4.9934,
      "mean_token_accuracy": 0.20222398191690444,
      "num_tokens": 27921281.0,
      "step": 15180
    },
    {
      "entropy": 5.1282798767089846,
      "epoch": 1.268376211159372,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048433619312568095,
      "loss": 4.8188,
      "mean_token_accuracy": 0.22084284722805023,
      "num_tokens": 27929266.0,
      "step": 15185
    },
    {
      "entropy": 5.186333179473877,
      "epoch": 1.268793852322085,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048432527795403405,
      "loss": 5.0068,
      "mean_token_accuracy": 0.2023696020245552,
      "num_tokens": 27938107.0,
      "step": 15190
    },
    {
      "entropy": 5.2662323951721195,
      "epoch": 1.2692114934847978,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004843143591178962,
      "loss": 5.051,
      "mean_token_accuracy": 0.19933224320411683,
      "num_tokens": 27947621.0,
      "step": 15195
    },
    {
      "entropy": 5.215515375137329,
      "epoch": 1.2696291346475108,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048430343661745844,
      "loss": 4.9969,
      "mean_token_accuracy": 0.20185237973928452,
      "num_tokens": 27956603.0,
      "step": 15200
    },
    {
      "entropy": 5.209321451187134,
      "epoch": 1.2700467758102238,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004842925104529122,
      "loss": 5.0353,
      "mean_token_accuracy": 0.20360014587640762,
      "num_tokens": 27966291.0,
      "step": 15205
    },
    {
      "entropy": 5.18613634109497,
      "epoch": 1.2704644169729369,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048428158062444857,
      "loss": 4.8906,
      "mean_token_accuracy": 0.2059626594185829,
      "num_tokens": 27975377.0,
      "step": 15210
    },
    {
      "entropy": 5.146964025497437,
      "epoch": 1.27088205813565,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048427064713225894,
      "loss": 4.9463,
      "mean_token_accuracy": 0.20589026510715486,
      "num_tokens": 27985429.0,
      "step": 15215
    },
    {
      "entropy": 5.193218183517456,
      "epoch": 1.2712996992983627,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048425970997653477,
      "loss": 4.9695,
      "mean_token_accuracy": 0.19826629757881165,
      "num_tokens": 27995247.0,
      "step": 15220
    },
    {
      "entropy": 5.138306427001953,
      "epoch": 1.2717173404610758,
      "grad_norm": 1.25,
      "learning_rate": 0.0004842487691574675,
      "loss": 4.8647,
      "mean_token_accuracy": 0.20721751153469087,
      "num_tokens": 28004638.0,
      "step": 15225
    },
    {
      "entropy": 5.146691799163818,
      "epoch": 1.2721349816237888,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048423782467524865,
      "loss": 5.0394,
      "mean_token_accuracy": 0.19834436029195784,
      "num_tokens": 28014149.0,
      "step": 15230
    },
    {
      "entropy": 5.240814065933227,
      "epoch": 1.2725526227865018,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004842268765300698,
      "loss": 4.9176,
      "mean_token_accuracy": 0.2011452481150627,
      "num_tokens": 28022651.0,
      "step": 15235
    },
    {
      "entropy": 5.152206659317017,
      "epoch": 1.2729702639492149,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048421592472212263,
      "loss": 4.9862,
      "mean_token_accuracy": 0.20226186364889145,
      "num_tokens": 28031397.0,
      "step": 15240
    },
    {
      "entropy": 5.168397998809814,
      "epoch": 1.273387905111928,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048420496925159886,
      "loss": 5.0007,
      "mean_token_accuracy": 0.20066383779048919,
      "num_tokens": 28041147.0,
      "step": 15245
    },
    {
      "entropy": 5.192602252960205,
      "epoch": 1.273805546274641,
      "grad_norm": 1.5,
      "learning_rate": 0.0004841940101186903,
      "loss": 4.9527,
      "mean_token_accuracy": 0.20332321226596833,
      "num_tokens": 28051216.0,
      "step": 15250
    },
    {
      "entropy": 5.194561862945557,
      "epoch": 1.2742231874373537,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004841830473235888,
      "loss": 4.9648,
      "mean_token_accuracy": 0.20185057520866395,
      "num_tokens": 28061849.0,
      "step": 15255
    },
    {
      "entropy": 5.240706729888916,
      "epoch": 1.2746408286000668,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048417208086648625,
      "loss": 4.9748,
      "mean_token_accuracy": 0.20331989526748656,
      "num_tokens": 28071538.0,
      "step": 15260
    },
    {
      "entropy": 5.214382648468018,
      "epoch": 1.2750584697627798,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004841611107475746,
      "loss": 4.9712,
      "mean_token_accuracy": 0.19684422612190247,
      "num_tokens": 28081566.0,
      "step": 15265
    },
    {
      "entropy": 5.205614280700684,
      "epoch": 1.2754761109254928,
      "grad_norm": 1.359375,
      "learning_rate": 0.000484150136967046,
      "loss": 4.9531,
      "mean_token_accuracy": 0.20160872489213943,
      "num_tokens": 28091031.0,
      "step": 15270
    },
    {
      "entropy": 5.140332841873169,
      "epoch": 1.2758937520882059,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004841391595250925,
      "loss": 4.8985,
      "mean_token_accuracy": 0.20952629148960114,
      "num_tokens": 28099583.0,
      "step": 15275
    },
    {
      "entropy": 5.146846723556519,
      "epoch": 1.2763113932509187,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004841281784219063,
      "loss": 4.9675,
      "mean_token_accuracy": 0.19680272787809372,
      "num_tokens": 28108809.0,
      "step": 15280
    },
    {
      "entropy": 5.17894401550293,
      "epoch": 1.2767290344136317,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048411719365767947,
      "loss": 4.8389,
      "mean_token_accuracy": 0.21552435159683228,
      "num_tokens": 28117587.0,
      "step": 15285
    },
    {
      "entropy": 5.1626606464385985,
      "epoch": 1.2771466755763448,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048410620523260455,
      "loss": 4.9302,
      "mean_token_accuracy": 0.20801685899496078,
      "num_tokens": 28127061.0,
      "step": 15290
    },
    {
      "entropy": 5.084664678573608,
      "epoch": 1.2775643167390578,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048409521314687383,
      "loss": 4.8901,
      "mean_token_accuracy": 0.20512267351150512,
      "num_tokens": 28136316.0,
      "step": 15295
    },
    {
      "entropy": 5.1433408737182615,
      "epoch": 1.2779819579017708,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004840842174006797,
      "loss": 4.93,
      "mean_token_accuracy": 0.19832748174667358,
      "num_tokens": 28144842.0,
      "step": 15300
    },
    {
      "entropy": 5.125684547424316,
      "epoch": 1.2783995990644839,
      "grad_norm": 1.5,
      "learning_rate": 0.0004840732179942146,
      "loss": 4.9265,
      "mean_token_accuracy": 0.20860806703567505,
      "num_tokens": 28154286.0,
      "step": 15305
    },
    {
      "entropy": 5.233809280395508,
      "epoch": 1.278817240227197,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004840622149276712,
      "loss": 4.9769,
      "mean_token_accuracy": 0.20038667023181916,
      "num_tokens": 28162995.0,
      "step": 15310
    },
    {
      "entropy": 5.2084534645080565,
      "epoch": 1.2792348813899097,
      "grad_norm": 1.25,
      "learning_rate": 0.0004840512082012421,
      "loss": 4.9392,
      "mean_token_accuracy": 0.19906319975852965,
      "num_tokens": 28172617.0,
      "step": 15315
    },
    {
      "entropy": 5.272152090072632,
      "epoch": 1.2796525225526227,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048404019781511995,
      "loss": 5.025,
      "mean_token_accuracy": 0.20001850575208663,
      "num_tokens": 28181944.0,
      "step": 15320
    },
    {
      "entropy": 5.082116985321045,
      "epoch": 1.2800701637153358,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048402918376949753,
      "loss": 4.8402,
      "mean_token_accuracy": 0.21299617886543273,
      "num_tokens": 28190704.0,
      "step": 15325
    },
    {
      "entropy": 5.136901807785034,
      "epoch": 1.2804878048780488,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004840181660645676,
      "loss": 4.9445,
      "mean_token_accuracy": 0.19952504187822342,
      "num_tokens": 28200042.0,
      "step": 15330
    },
    {
      "entropy": 5.218760824203491,
      "epoch": 1.2809054460407618,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004840071447005231,
      "loss": 5.0119,
      "mean_token_accuracy": 0.19723379760980606,
      "num_tokens": 28209989.0,
      "step": 15335
    },
    {
      "entropy": 5.276055240631104,
      "epoch": 1.2813230872034747,
      "grad_norm": 1.203125,
      "learning_rate": 0.000483996119677557,
      "loss": 4.9415,
      "mean_token_accuracy": 0.20475793182849883,
      "num_tokens": 28219018.0,
      "step": 15340
    },
    {
      "entropy": 5.13262677192688,
      "epoch": 1.281740728366188,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048398509099586225,
      "loss": 4.9297,
      "mean_token_accuracy": 0.20474273711442947,
      "num_tokens": 28227912.0,
      "step": 15345
    },
    {
      "entropy": 5.107831048965454,
      "epoch": 1.2821583695289007,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048397405865563186,
      "loss": 4.921,
      "mean_token_accuracy": 0.20204878449440003,
      "num_tokens": 28237314.0,
      "step": 15350
    },
    {
      "entropy": 5.089810180664062,
      "epoch": 1.2825760106916138,
      "grad_norm": 1.25,
      "learning_rate": 0.0004839630226570591,
      "loss": 4.8724,
      "mean_token_accuracy": 0.21178796291351318,
      "num_tokens": 28246490.0,
      "step": 15355
    },
    {
      "entropy": 5.102705430984497,
      "epoch": 1.2829936518543268,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000483951983000337,
      "loss": 4.8806,
      "mean_token_accuracy": 0.21255181729793549,
      "num_tokens": 28255243.0,
      "step": 15360
    },
    {
      "entropy": 5.198540592193604,
      "epoch": 1.2834112930170398,
      "grad_norm": 1.28125,
      "learning_rate": 0.000483940939685659,
      "loss": 4.8787,
      "mean_token_accuracy": 0.20986103564500808,
      "num_tokens": 28265106.0,
      "step": 15365
    },
    {
      "entropy": 5.212856340408325,
      "epoch": 1.2838289341797529,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048392989271321836,
      "loss": 4.9932,
      "mean_token_accuracy": 0.19659313261508943,
      "num_tokens": 28275214.0,
      "step": 15370
    },
    {
      "entropy": 5.145081996917725,
      "epoch": 1.2842465753424657,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004839188420832084,
      "loss": 4.8869,
      "mean_token_accuracy": 0.2127483978867531,
      "num_tokens": 28284391.0,
      "step": 15375
    },
    {
      "entropy": 5.150554180145264,
      "epoch": 1.2846642165051787,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048390778779582266,
      "loss": 5.0005,
      "mean_token_accuracy": 0.1975470796227455,
      "num_tokens": 28294070.0,
      "step": 15380
    },
    {
      "entropy": 5.140177774429321,
      "epoch": 1.2850818576678917,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048389672985125467,
      "loss": 4.8938,
      "mean_token_accuracy": 0.20421373397111892,
      "num_tokens": 28302936.0,
      "step": 15385
    },
    {
      "entropy": 5.222402572631836,
      "epoch": 1.2854994988306048,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004838856682496979,
      "loss": 4.9756,
      "mean_token_accuracy": 0.20209104418754578,
      "num_tokens": 28312487.0,
      "step": 15390
    },
    {
      "entropy": 5.1427253723144535,
      "epoch": 1.2859171399933178,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004838746029913461,
      "loss": 4.7474,
      "mean_token_accuracy": 0.22176881283521652,
      "num_tokens": 28321331.0,
      "step": 15395
    },
    {
      "entropy": 5.157711124420166,
      "epoch": 1.2863347811560306,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000483863534076393,
      "loss": 4.9062,
      "mean_token_accuracy": 0.21372569352388382,
      "num_tokens": 28329459.0,
      "step": 15400
    },
    {
      "entropy": 5.164652633666992,
      "epoch": 1.2867524223187439,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004838524615050323,
      "loss": 4.9979,
      "mean_token_accuracy": 0.19753903597593309,
      "num_tokens": 28338128.0,
      "step": 15405
    },
    {
      "entropy": 5.175483226776123,
      "epoch": 1.2871700634814567,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004838413852774579,
      "loss": 4.9392,
      "mean_token_accuracy": 0.20209489166736602,
      "num_tokens": 28347697.0,
      "step": 15410
    },
    {
      "entropy": 5.251401758193969,
      "epoch": 1.2875877046441697,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048383030539386355,
      "loss": 5.0279,
      "mean_token_accuracy": 0.2009572431445122,
      "num_tokens": 28357146.0,
      "step": 15415
    },
    {
      "entropy": 5.109013795852661,
      "epoch": 1.2880053458068828,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004838192218544434,
      "loss": 4.9161,
      "mean_token_accuracy": 0.20834251046180724,
      "num_tokens": 28366399.0,
      "step": 15420
    },
    {
      "entropy": 5.107675457000733,
      "epoch": 1.2884229869695958,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004838081346593915,
      "loss": 4.9675,
      "mean_token_accuracy": 0.20038881599903108,
      "num_tokens": 28375030.0,
      "step": 15425
    },
    {
      "entropy": 5.134133672714233,
      "epoch": 1.2888406281323088,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048379704380890177,
      "loss": 4.8617,
      "mean_token_accuracy": 0.21200687885284425,
      "num_tokens": 28383704.0,
      "step": 15430
    },
    {
      "entropy": 5.303136157989502,
      "epoch": 1.2892582692950216,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004837859493031685,
      "loss": 5.0253,
      "mean_token_accuracy": 0.20004342198371888,
      "num_tokens": 28393270.0,
      "step": 15435
    },
    {
      "entropy": 5.179622936248779,
      "epoch": 1.2896759104577347,
      "grad_norm": 1.5,
      "learning_rate": 0.00048377485114238576,
      "loss": 4.9119,
      "mean_token_accuracy": 0.20213798880577089,
      "num_tokens": 28402757.0,
      "step": 15440
    },
    {
      "entropy": 5.08172459602356,
      "epoch": 1.2900935516204477,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004837637493267481,
      "loss": 4.8709,
      "mean_token_accuracy": 0.20826005935668945,
      "num_tokens": 28411888.0,
      "step": 15445
    },
    {
      "entropy": 5.215762710571289,
      "epoch": 1.2905111927831607,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004837526438564496,
      "loss": 5.021,
      "mean_token_accuracy": 0.19225906133651732,
      "num_tokens": 28421631.0,
      "step": 15450
    },
    {
      "entropy": 5.138888263702393,
      "epoch": 1.2909288339458738,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004837415347316849,
      "loss": 4.9398,
      "mean_token_accuracy": 0.20212435424327851,
      "num_tokens": 28431375.0,
      "step": 15455
    },
    {
      "entropy": 5.163145351409912,
      "epoch": 1.2913464751085866,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048373042195264836,
      "loss": 4.9131,
      "mean_token_accuracy": 0.20443625152111053,
      "num_tokens": 28440555.0,
      "step": 15460
    },
    {
      "entropy": 5.237237787246704,
      "epoch": 1.2917641162712998,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004837193055195344,
      "loss": 5.0398,
      "mean_token_accuracy": 0.19929395020008087,
      "num_tokens": 28449362.0,
      "step": 15465
    },
    {
      "entropy": 5.2098273754119875,
      "epoch": 1.2921817574340126,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048370818543253793,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21008391678333282,
      "num_tokens": 28459285.0,
      "step": 15470
    },
    {
      "entropy": 5.124650144577027,
      "epoch": 1.2925993985967257,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048369706169185334,
      "loss": 4.9237,
      "mean_token_accuracy": 0.20810714662075042,
      "num_tokens": 28469091.0,
      "step": 15475
    },
    {
      "entropy": 5.163948822021484,
      "epoch": 1.2930170397594387,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048368593429767555,
      "loss": 4.9231,
      "mean_token_accuracy": 0.20571576654911042,
      "num_tokens": 28478127.0,
      "step": 15480
    },
    {
      "entropy": 5.183759593963623,
      "epoch": 1.2934346809221517,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048367480325019923,
      "loss": 4.9621,
      "mean_token_accuracy": 0.20311406701803209,
      "num_tokens": 28487823.0,
      "step": 15485
    },
    {
      "entropy": 5.1497221946716305,
      "epoch": 1.2938523220848648,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048366366854961937,
      "loss": 5.0019,
      "mean_token_accuracy": 0.20113380402326583,
      "num_tokens": 28497924.0,
      "step": 15490
    },
    {
      "entropy": 5.216079473495483,
      "epoch": 1.2942699632475776,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004836525301961307,
      "loss": 4.9073,
      "mean_token_accuracy": 0.20981784611940385,
      "num_tokens": 28506994.0,
      "step": 15495
    },
    {
      "entropy": 5.080442905426025,
      "epoch": 1.2946876044102906,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004836413881899285,
      "loss": 4.8522,
      "mean_token_accuracy": 0.21079850494861602,
      "num_tokens": 28516919.0,
      "step": 15500
    },
    {
      "entropy": 5.113883447647095,
      "epoch": 1.2951052455730037,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004836302425312075,
      "loss": 4.9444,
      "mean_token_accuracy": 0.21432480812072754,
      "num_tokens": 28524902.0,
      "step": 15505
    },
    {
      "entropy": 5.226777839660644,
      "epoch": 1.2955228867357167,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004836190932201631,
      "loss": 4.995,
      "mean_token_accuracy": 0.20123831182718277,
      "num_tokens": 28534257.0,
      "step": 15510
    },
    {
      "entropy": 5.1801433086395265,
      "epoch": 1.2959405278984297,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048360794025699026,
      "loss": 4.8972,
      "mean_token_accuracy": 0.20669442415237427,
      "num_tokens": 28544056.0,
      "step": 15515
    },
    {
      "entropy": 5.154482460021972,
      "epoch": 1.2963581690611425,
      "grad_norm": 1.25,
      "learning_rate": 0.0004835967836418844,
      "loss": 4.946,
      "mean_token_accuracy": 0.20038562268018723,
      "num_tokens": 28552858.0,
      "step": 15520
    },
    {
      "entropy": 5.155268764495849,
      "epoch": 1.2967758102238558,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004835856233750407,
      "loss": 4.942,
      "mean_token_accuracy": 0.20941556841135026,
      "num_tokens": 28561548.0,
      "step": 15525
    },
    {
      "entropy": 5.159929943084717,
      "epoch": 1.2971934513865686,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048357445945665465,
      "loss": 4.8669,
      "mean_token_accuracy": 0.21086709201335907,
      "num_tokens": 28570931.0,
      "step": 15530
    },
    {
      "entropy": 5.167643451690674,
      "epoch": 1.2976110925492816,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004835632918869215,
      "loss": 5.0007,
      "mean_token_accuracy": 0.19450924843549727,
      "num_tokens": 28580298.0,
      "step": 15535
    },
    {
      "entropy": 5.13236813545227,
      "epoch": 1.2980287337119947,
      "grad_norm": 1.9375,
      "learning_rate": 0.000483552120666037,
      "loss": 4.8103,
      "mean_token_accuracy": 0.21437878012657166,
      "num_tokens": 28589740.0,
      "step": 15540
    },
    {
      "entropy": 5.167823410034179,
      "epoch": 1.2984463748747077,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048354094579419656,
      "loss": 4.8507,
      "mean_token_accuracy": 0.21602003425359725,
      "num_tokens": 28598538.0,
      "step": 15545
    },
    {
      "entropy": 5.17964391708374,
      "epoch": 1.2988640160374207,
      "grad_norm": 1.25,
      "learning_rate": 0.0004835297672715958,
      "loss": 4.9903,
      "mean_token_accuracy": 0.20086732804775237,
      "num_tokens": 28608132.0,
      "step": 15550
    },
    {
      "entropy": 5.208846092224121,
      "epoch": 1.2992816572001336,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048351858509843054,
      "loss": 4.9775,
      "mean_token_accuracy": 0.20724613070487977,
      "num_tokens": 28617502.0,
      "step": 15555
    },
    {
      "entropy": 5.154063606262207,
      "epoch": 1.2996992983628466,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004835073992748964,
      "loss": 4.9028,
      "mean_token_accuracy": 0.2091211512684822,
      "num_tokens": 28626798.0,
      "step": 15560
    },
    {
      "entropy": 5.14700837135315,
      "epoch": 1.3001169395255596,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048349620980118926,
      "loss": 4.8768,
      "mean_token_accuracy": 0.20239893794059755,
      "num_tokens": 28636599.0,
      "step": 15565
    },
    {
      "entropy": 5.145308113098144,
      "epoch": 1.3005345806882727,
      "grad_norm": 1.90625,
      "learning_rate": 0.000483485016677505,
      "loss": 4.9027,
      "mean_token_accuracy": 0.21225001215934752,
      "num_tokens": 28646156.0,
      "step": 15570
    },
    {
      "entropy": 5.32040433883667,
      "epoch": 1.3009522218509857,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048347381990403957,
      "loss": 5.0706,
      "mean_token_accuracy": 0.19877982437610625,
      "num_tokens": 28654492.0,
      "step": 15575
    },
    {
      "entropy": 5.171299457550049,
      "epoch": 1.3013698630136985,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048346261948098895,
      "loss": 4.9317,
      "mean_token_accuracy": 0.2061738207936287,
      "num_tokens": 28663606.0,
      "step": 15580
    },
    {
      "entropy": 5.206483697891235,
      "epoch": 1.3017875041764118,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048345141540854933,
      "loss": 5.0411,
      "mean_token_accuracy": 0.1972224310040474,
      "num_tokens": 28671924.0,
      "step": 15585
    },
    {
      "entropy": 5.164387559890747,
      "epoch": 1.3022051453391246,
      "grad_norm": 1.25,
      "learning_rate": 0.00048344020768691673,
      "loss": 4.9005,
      "mean_token_accuracy": 0.2069829061627388,
      "num_tokens": 28681428.0,
      "step": 15590
    },
    {
      "entropy": 5.078217267990112,
      "epoch": 1.3026227865018376,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004834289963162874,
      "loss": 4.884,
      "mean_token_accuracy": 0.21305157542228698,
      "num_tokens": 28689894.0,
      "step": 15595
    },
    {
      "entropy": 5.138427495956421,
      "epoch": 1.3030404276645506,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004834177812968576,
      "loss": 4.8609,
      "mean_token_accuracy": 0.2154493436217308,
      "num_tokens": 28698084.0,
      "step": 15600
    },
    {
      "entropy": 5.166884899139404,
      "epoch": 1.3034580688272637,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048340656262882366,
      "loss": 4.8861,
      "mean_token_accuracy": 0.20612238943576813,
      "num_tokens": 28706817.0,
      "step": 15605
    },
    {
      "entropy": 5.130106258392334,
      "epoch": 1.3038757099899767,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048339534031238197,
      "loss": 4.8047,
      "mean_token_accuracy": 0.21336934417486192,
      "num_tokens": 28715020.0,
      "step": 15610
    },
    {
      "entropy": 5.142653274536133,
      "epoch": 1.3042933511526895,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000483384114347729,
      "loss": 4.8915,
      "mean_token_accuracy": 0.20867228806018828,
      "num_tokens": 28724570.0,
      "step": 15615
    },
    {
      "entropy": 5.169595861434937,
      "epoch": 1.3047109923154026,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004833728847350613,
      "loss": 4.8903,
      "mean_token_accuracy": 0.20630472302436828,
      "num_tokens": 28732832.0,
      "step": 15620
    },
    {
      "entropy": 5.151440954208374,
      "epoch": 1.3051286334781156,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048336165147457537,
      "loss": 4.9449,
      "mean_token_accuracy": 0.20442404448986054,
      "num_tokens": 28742335.0,
      "step": 15625
    },
    {
      "entropy": 5.17317271232605,
      "epoch": 1.3055462746408286,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004833504145664681,
      "loss": 4.8867,
      "mean_token_accuracy": 0.21068256348371506,
      "num_tokens": 28751596.0,
      "step": 15630
    },
    {
      "entropy": 5.115510416030884,
      "epoch": 1.3059639158035417,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048333917401093587,
      "loss": 4.9432,
      "mean_token_accuracy": 0.20476183146238328,
      "num_tokens": 28761485.0,
      "step": 15635
    },
    {
      "entropy": 5.08582706451416,
      "epoch": 1.3063815569662545,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048332792980817563,
      "loss": 4.8695,
      "mean_token_accuracy": 0.20488111972808837,
      "num_tokens": 28771421.0,
      "step": 15640
    },
    {
      "entropy": 5.206784963607788,
      "epoch": 1.3067991981289677,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004833166819583843,
      "loss": 4.9628,
      "mean_token_accuracy": 0.20690949410200118,
      "num_tokens": 28780392.0,
      "step": 15645
    },
    {
      "entropy": 5.255417060852051,
      "epoch": 1.3072168392916805,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004833054304617586,
      "loss": 4.9457,
      "mean_token_accuracy": 0.20677572041749953,
      "num_tokens": 28788898.0,
      "step": 15650
    },
    {
      "entropy": 5.213649272918701,
      "epoch": 1.3076344804543936,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048329417531849567,
      "loss": 4.9485,
      "mean_token_accuracy": 0.20104088485240937,
      "num_tokens": 28797434.0,
      "step": 15655
    },
    {
      "entropy": 5.2232763290405275,
      "epoch": 1.3080521216171066,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048328291652879245,
      "loss": 5.009,
      "mean_token_accuracy": 0.20068462640047074,
      "num_tokens": 28806842.0,
      "step": 15660
    },
    {
      "entropy": 5.142750358581543,
      "epoch": 1.3084697627798196,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048327165409284614,
      "loss": 4.9143,
      "mean_token_accuracy": 0.21081365942955016,
      "num_tokens": 28816088.0,
      "step": 15665
    },
    {
      "entropy": 5.182534646987915,
      "epoch": 1.3088874039425327,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004832603880108537,
      "loss": 4.8873,
      "mean_token_accuracy": 0.20924682766199112,
      "num_tokens": 28824380.0,
      "step": 15670
    },
    {
      "entropy": 5.069730758666992,
      "epoch": 1.3093050451052455,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048324911828301256,
      "loss": 4.8393,
      "mean_token_accuracy": 0.20908746123313904,
      "num_tokens": 28833374.0,
      "step": 15675
    },
    {
      "entropy": 5.125282573699951,
      "epoch": 1.3097226862679585,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004832378449095199,
      "loss": 4.9461,
      "mean_token_accuracy": 0.20797738134860994,
      "num_tokens": 28842446.0,
      "step": 15680
    },
    {
      "entropy": 5.217948055267334,
      "epoch": 1.3101403274306715,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004832265678905731,
      "loss": 5.0614,
      "mean_token_accuracy": 0.1971353068947792,
      "num_tokens": 28851807.0,
      "step": 15685
    },
    {
      "entropy": 5.26354603767395,
      "epoch": 1.3105579685933846,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004832152872263697,
      "loss": 4.9823,
      "mean_token_accuracy": 0.1993308499455452,
      "num_tokens": 28860825.0,
      "step": 15690
    },
    {
      "entropy": 5.300148057937622,
      "epoch": 1.3109756097560976,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004832040029171069,
      "loss": 5.1325,
      "mean_token_accuracy": 0.19986274987459182,
      "num_tokens": 28870071.0,
      "step": 15695
    },
    {
      "entropy": 5.227228355407715,
      "epoch": 1.3113932509188104,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048319271496298255,
      "loss": 4.9403,
      "mean_token_accuracy": 0.20283003002405167,
      "num_tokens": 28879207.0,
      "step": 15700
    },
    {
      "entropy": 5.18164758682251,
      "epoch": 1.3118108920815237,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048318142336419413,
      "loss": 5.0174,
      "mean_token_accuracy": 0.20084186494350434,
      "num_tokens": 28888696.0,
      "step": 15705
    },
    {
      "entropy": 5.121413087844848,
      "epoch": 1.3122285332442365,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048317012812093926,
      "loss": 4.8967,
      "mean_token_accuracy": 0.2108782336115837,
      "num_tokens": 28897499.0,
      "step": 15710
    },
    {
      "entropy": 5.20941424369812,
      "epoch": 1.3126461744069495,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004831588292334157,
      "loss": 4.9515,
      "mean_token_accuracy": 0.20170468538999559,
      "num_tokens": 28907459.0,
      "step": 15715
    },
    {
      "entropy": 5.155849599838257,
      "epoch": 1.3130638155696626,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004831475267018213,
      "loss": 4.9319,
      "mean_token_accuracy": 0.201112100481987,
      "num_tokens": 28917828.0,
      "step": 15720
    },
    {
      "entropy": 5.1737621307373045,
      "epoch": 1.3134814567323756,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004831362205263538,
      "loss": 4.8966,
      "mean_token_accuracy": 0.20234486311674119,
      "num_tokens": 28927289.0,
      "step": 15725
    },
    {
      "entropy": 5.183673667907715,
      "epoch": 1.3138990978950886,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004831249107072114,
      "loss": 4.9014,
      "mean_token_accuracy": 0.21270525306463242,
      "num_tokens": 28936210.0,
      "step": 15730
    },
    {
      "entropy": 5.2109921932220455,
      "epoch": 1.3143167390578014,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048311359724459175,
      "loss": 5.0153,
      "mean_token_accuracy": 0.19991078227758408,
      "num_tokens": 28945942.0,
      "step": 15735
    },
    {
      "entropy": 5.102045917510987,
      "epoch": 1.3147343802205145,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048310228013869323,
      "loss": 4.8534,
      "mean_token_accuracy": 0.2143813982605934,
      "num_tokens": 28955023.0,
      "step": 15740
    },
    {
      "entropy": 5.184965467453003,
      "epoch": 1.3151520213832275,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004830909593897137,
      "loss": 4.9477,
      "mean_token_accuracy": 0.20453268438577651,
      "num_tokens": 28964454.0,
      "step": 15745
    },
    {
      "entropy": 5.117368078231811,
      "epoch": 1.3155696625459405,
      "grad_norm": 1.25,
      "learning_rate": 0.0004830796349978515,
      "loss": 4.8168,
      "mean_token_accuracy": 0.21450940072536467,
      "num_tokens": 28973364.0,
      "step": 15750
    },
    {
      "entropy": 5.1757598400115965,
      "epoch": 1.3159873037086536,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004830683069633048,
      "loss": 5.0436,
      "mean_token_accuracy": 0.2026698499917984,
      "num_tokens": 28982641.0,
      "step": 15755
    },
    {
      "entropy": 5.141507196426391,
      "epoch": 1.3164049448713664,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004830569752862719,
      "loss": 4.8518,
      "mean_token_accuracy": 0.21419608145952224,
      "num_tokens": 28991149.0,
      "step": 15760
    },
    {
      "entropy": 5.157705640792846,
      "epoch": 1.3168225860340796,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048304563996695127,
      "loss": 5.0079,
      "mean_token_accuracy": 0.19541564285755159,
      "num_tokens": 29001063.0,
      "step": 15765
    },
    {
      "entropy": 5.17944598197937,
      "epoch": 1.3172402271967925,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048303430100554126,
      "loss": 4.9652,
      "mean_token_accuracy": 0.20714571177959443,
      "num_tokens": 29009921.0,
      "step": 15770
    },
    {
      "entropy": 5.1805980682373045,
      "epoch": 1.3176578683595055,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004830229584022404,
      "loss": 4.9644,
      "mean_token_accuracy": 0.20314323902130127,
      "num_tokens": 29019247.0,
      "step": 15775
    },
    {
      "entropy": 5.139682912826538,
      "epoch": 1.3180755095222185,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004830116121572472,
      "loss": 4.9533,
      "mean_token_accuracy": 0.20579672306776048,
      "num_tokens": 29028299.0,
      "step": 15780
    },
    {
      "entropy": 5.15749831199646,
      "epoch": 1.3184931506849316,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004830002622707604,
      "loss": 4.8882,
      "mean_token_accuracy": 0.2055453136563301,
      "num_tokens": 29037501.0,
      "step": 15785
    },
    {
      "entropy": 5.263999557495117,
      "epoch": 1.3189107918476446,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048298890874297866,
      "loss": 5.0502,
      "mean_token_accuracy": 0.20791687816381454,
      "num_tokens": 29046987.0,
      "step": 15790
    },
    {
      "entropy": 5.209774971008301,
      "epoch": 1.3193284330103574,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004829775515741007,
      "loss": 4.9578,
      "mean_token_accuracy": 0.2072492778301239,
      "num_tokens": 29057280.0,
      "step": 15795
    },
    {
      "entropy": 5.124493503570557,
      "epoch": 1.3197460741730704,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004829661907643253,
      "loss": 4.9812,
      "mean_token_accuracy": 0.19855135977268218,
      "num_tokens": 29067908.0,
      "step": 15800
    },
    {
      "entropy": 5.198938035964966,
      "epoch": 1.3201637153357835,
      "grad_norm": 1.25,
      "learning_rate": 0.0004829548263138515,
      "loss": 5.0413,
      "mean_token_accuracy": 0.1971461594104767,
      "num_tokens": 29077207.0,
      "step": 15805
    },
    {
      "entropy": 5.210603189468384,
      "epoch": 1.3205813564984965,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004829434582228781,
      "loss": 4.8746,
      "mean_token_accuracy": 0.20705655068159104,
      "num_tokens": 29085795.0,
      "step": 15810
    },
    {
      "entropy": 5.163326072692871,
      "epoch": 1.3209989976612095,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004829320864916042,
      "loss": 4.9923,
      "mean_token_accuracy": 0.20575217604637147,
      "num_tokens": 29095291.0,
      "step": 15815
    },
    {
      "entropy": 5.157334804534912,
      "epoch": 1.3214166388239224,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048292071112022886,
      "loss": 4.8396,
      "mean_token_accuracy": 0.2083418369293213,
      "num_tokens": 29103993.0,
      "step": 15820
    },
    {
      "entropy": 5.1076263904571535,
      "epoch": 1.3218342799866356,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004829093321089511,
      "loss": 4.952,
      "mean_token_accuracy": 0.20847168415784836,
      "num_tokens": 29113047.0,
      "step": 15825
    },
    {
      "entropy": 5.155609655380249,
      "epoch": 1.3222519211493484,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004828979494579703,
      "loss": 4.984,
      "mean_token_accuracy": 0.20783765763044357,
      "num_tokens": 29122637.0,
      "step": 15830
    },
    {
      "entropy": 5.133995103836059,
      "epoch": 1.3226695623120615,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048288656316748554,
      "loss": 4.9043,
      "mean_token_accuracy": 0.2103823110461235,
      "num_tokens": 29132237.0,
      "step": 15835
    },
    {
      "entropy": 5.238634014129639,
      "epoch": 1.3230872034747745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048287517323769634,
      "loss": 4.9831,
      "mean_token_accuracy": 0.20392991304397584,
      "num_tokens": 29141317.0,
      "step": 15840
    },
    {
      "entropy": 5.262897634506226,
      "epoch": 1.3235048446374875,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048286377966880206,
      "loss": 5.0485,
      "mean_token_accuracy": 0.19479888826608657,
      "num_tokens": 29150993.0,
      "step": 15845
    },
    {
      "entropy": 5.1340655326843265,
      "epoch": 1.3239224858002006,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048285238246100203,
      "loss": 4.7886,
      "mean_token_accuracy": 0.21515662372112274,
      "num_tokens": 29159126.0,
      "step": 15850
    },
    {
      "entropy": 5.12099666595459,
      "epoch": 1.3243401269629134,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004828409816144959,
      "loss": 4.9332,
      "mean_token_accuracy": 0.20742589682340623,
      "num_tokens": 29167262.0,
      "step": 15855
    },
    {
      "entropy": 5.099379682540894,
      "epoch": 1.3247577681256264,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004828295771294831,
      "loss": 4.8631,
      "mean_token_accuracy": 0.2114694148302078,
      "num_tokens": 29177634.0,
      "step": 15860
    },
    {
      "entropy": 5.171880435943604,
      "epoch": 1.3251754092883394,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048281816900616354,
      "loss": 4.8962,
      "mean_token_accuracy": 0.20466845780611037,
      "num_tokens": 29185922.0,
      "step": 15865
    },
    {
      "entropy": 5.105803442001343,
      "epoch": 1.3255930504510525,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004828067572447367,
      "loss": 4.8865,
      "mean_token_accuracy": 0.20808068960905074,
      "num_tokens": 29195178.0,
      "step": 15870
    },
    {
      "entropy": 5.095398855209351,
      "epoch": 1.3260106916137655,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048279534184540246,
      "loss": 4.8965,
      "mean_token_accuracy": 0.21891288012266158,
      "num_tokens": 29203203.0,
      "step": 15875
    },
    {
      "entropy": 5.138548278808594,
      "epoch": 1.3264283327764783,
      "grad_norm": 1.25,
      "learning_rate": 0.0004827839228083606,
      "loss": 4.9537,
      "mean_token_accuracy": 0.20312877893447875,
      "num_tokens": 29213135.0,
      "step": 15880
    },
    {
      "entropy": 5.17853045463562,
      "epoch": 1.3268459739391916,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004827725001338111,
      "loss": 4.9495,
      "mean_token_accuracy": 0.20396195352077484,
      "num_tokens": 29222666.0,
      "step": 15885
    },
    {
      "entropy": 5.201378297805786,
      "epoch": 1.3272636151019044,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048276107382195393,
      "loss": 4.9462,
      "mean_token_accuracy": 0.20542955845594407,
      "num_tokens": 29232215.0,
      "step": 15890
    },
    {
      "entropy": 5.206853151321411,
      "epoch": 1.3276812562646174,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048274964387298903,
      "loss": 4.9545,
      "mean_token_accuracy": 0.2004047691822052,
      "num_tokens": 29242073.0,
      "step": 15895
    },
    {
      "entropy": 5.0807394027709964,
      "epoch": 1.3280988974273304,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004827382102871166,
      "loss": 4.8727,
      "mean_token_accuracy": 0.2123336002230644,
      "num_tokens": 29251135.0,
      "step": 15900
    },
    {
      "entropy": 5.240659379959107,
      "epoch": 1.3285165385900435,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048272677306453665,
      "loss": 4.9502,
      "mean_token_accuracy": 0.2110212191939354,
      "num_tokens": 29260746.0,
      "step": 15905
    },
    {
      "entropy": 5.170893764495849,
      "epoch": 1.3289341797527565,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048271533220544955,
      "loss": 4.9168,
      "mean_token_accuracy": 0.2044968619942665,
      "num_tokens": 29269102.0,
      "step": 15910
    },
    {
      "entropy": 5.077240610122681,
      "epoch": 1.3293518209154693,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048270388771005555,
      "loss": 4.8651,
      "mean_token_accuracy": 0.2116764545440674,
      "num_tokens": 29277474.0,
      "step": 15915
    },
    {
      "entropy": 5.162775039672852,
      "epoch": 1.3297694620781824,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048269243957855493,
      "loss": 4.9017,
      "mean_token_accuracy": 0.20935402661561966,
      "num_tokens": 29286744.0,
      "step": 15920
    },
    {
      "entropy": 5.172525691986084,
      "epoch": 1.3301871032408954,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004826809878111481,
      "loss": 4.8961,
      "mean_token_accuracy": 0.2040143519639969,
      "num_tokens": 29296039.0,
      "step": 15925
    },
    {
      "entropy": 5.07702898979187,
      "epoch": 1.3306047444036084,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048266953240803576,
      "loss": 4.9253,
      "mean_token_accuracy": 0.20901413410902023,
      "num_tokens": 29304777.0,
      "step": 15930
    },
    {
      "entropy": 5.147122049331665,
      "epoch": 1.3310223855663215,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004826580733694182,
      "loss": 4.897,
      "mean_token_accuracy": 0.21184486895799637,
      "num_tokens": 29313161.0,
      "step": 15935
    },
    {
      "entropy": 5.186066484451294,
      "epoch": 1.3314400267290343,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004826466106954961,
      "loss": 4.9503,
      "mean_token_accuracy": 0.21401528120040894,
      "num_tokens": 29322620.0,
      "step": 15940
    },
    {
      "entropy": 5.1588826179504395,
      "epoch": 1.3318576678917475,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048263514438647015,
      "loss": 4.9897,
      "mean_token_accuracy": 0.19165965020656586,
      "num_tokens": 29332641.0,
      "step": 15945
    },
    {
      "entropy": 5.136185836791992,
      "epoch": 1.3322753090544603,
      "grad_norm": 1.296875,
      "learning_rate": 0.000482623674442541,
      "loss": 5.0251,
      "mean_token_accuracy": 0.2028067484498024,
      "num_tokens": 29342582.0,
      "step": 15950
    },
    {
      "entropy": 5.15098729133606,
      "epoch": 1.3326929502171734,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004826122008639096,
      "loss": 4.9636,
      "mean_token_accuracy": 0.20500448048114778,
      "num_tokens": 29351982.0,
      "step": 15955
    },
    {
      "entropy": 5.241524076461792,
      "epoch": 1.3331105913798864,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048260072365077675,
      "loss": 4.9354,
      "mean_token_accuracy": 0.20503380447626113,
      "num_tokens": 29361564.0,
      "step": 15960
    },
    {
      "entropy": 5.190820646286011,
      "epoch": 1.3335282325425994,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004825892428033432,
      "loss": 5.0094,
      "mean_token_accuracy": 0.19378644973039627,
      "num_tokens": 29371026.0,
      "step": 15965
    },
    {
      "entropy": 5.194439649581909,
      "epoch": 1.3339458737053125,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004825777583218102,
      "loss": 4.9341,
      "mean_token_accuracy": 0.19876884967088698,
      "num_tokens": 29380304.0,
      "step": 15970
    },
    {
      "entropy": 5.14125623703003,
      "epoch": 1.3343635148680253,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004825662702063787,
      "loss": 4.8068,
      "mean_token_accuracy": 0.21402783393859864,
      "num_tokens": 29388567.0,
      "step": 15975
    },
    {
      "entropy": 5.131200885772705,
      "epoch": 1.3347811560307383,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004825547784572497,
      "loss": 4.9038,
      "mean_token_accuracy": 0.20977475196123124,
      "num_tokens": 29397523.0,
      "step": 15980
    },
    {
      "entropy": 5.136656045913696,
      "epoch": 1.3351987971934514,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004825432830746245,
      "loss": 4.8884,
      "mean_token_accuracy": 0.2090964287519455,
      "num_tokens": 29405556.0,
      "step": 15985
    },
    {
      "entropy": 5.162516927719116,
      "epoch": 1.3356164383561644,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004825317840587044,
      "loss": 4.9526,
      "mean_token_accuracy": 0.20380874276161193,
      "num_tokens": 29416043.0,
      "step": 15990
    },
    {
      "entropy": 5.144733667373657,
      "epoch": 1.3360340795188774,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004825202814096905,
      "loss": 4.9591,
      "mean_token_accuracy": 0.20723665058612822,
      "num_tokens": 29425410.0,
      "step": 15995
    },
    {
      "entropy": 5.236337423324585,
      "epoch": 1.3364517206815905,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004825087751277843,
      "loss": 4.991,
      "mean_token_accuracy": 0.20957676619291304,
      "num_tokens": 29434879.0,
      "step": 16000
    },
    {
      "entropy": 5.110996723175049,
      "epoch": 1.3368693618443035,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048249726521318716,
      "loss": 4.8353,
      "mean_token_accuracy": 0.21886272132396697,
      "num_tokens": 29443844.0,
      "step": 16005
    },
    {
      "entropy": 5.05059232711792,
      "epoch": 1.3372870030070163,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004824857516661008,
      "loss": 4.8045,
      "mean_token_accuracy": 0.2158799186348915,
      "num_tokens": 29452437.0,
      "step": 16010
    },
    {
      "entropy": 5.079374647140503,
      "epoch": 1.3377046441697293,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048247423448672647,
      "loss": 4.8411,
      "mean_token_accuracy": 0.21246036291122436,
      "num_tokens": 29461733.0,
      "step": 16015
    },
    {
      "entropy": 5.184403944015503,
      "epoch": 1.3381222853324424,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004824627136752659,
      "loss": 4.9712,
      "mean_token_accuracy": 0.197035051882267,
      "num_tokens": 29472287.0,
      "step": 16020
    },
    {
      "entropy": 5.105896902084351,
      "epoch": 1.3385399264951554,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004824511892319209,
      "loss": 4.8407,
      "mean_token_accuracy": 0.21075997203588487,
      "num_tokens": 29482587.0,
      "step": 16025
    },
    {
      "entropy": 5.091841793060302,
      "epoch": 1.3389575676578684,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048243966115689305,
      "loss": 4.8566,
      "mean_token_accuracy": 0.21704034358263016,
      "num_tokens": 29492005.0,
      "step": 16030
    },
    {
      "entropy": 5.119936323165893,
      "epoch": 1.3393752088205813,
      "grad_norm": 1.25,
      "learning_rate": 0.0004824281294503843,
      "loss": 4.8813,
      "mean_token_accuracy": 0.2122072696685791,
      "num_tokens": 29501167.0,
      "step": 16035
    },
    {
      "entropy": 5.0829850196838375,
      "epoch": 1.3397928499832943,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004824165941125964,
      "loss": 4.9225,
      "mean_token_accuracy": 0.20412594825029373,
      "num_tokens": 29509631.0,
      "step": 16040
    },
    {
      "entropy": 5.251471853256225,
      "epoch": 1.3402104911460073,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004824050551437315,
      "loss": 4.9276,
      "mean_token_accuracy": 0.2117255598306656,
      "num_tokens": 29518453.0,
      "step": 16045
    },
    {
      "entropy": 5.211207008361816,
      "epoch": 1.3406281323087204,
      "grad_norm": 1.25,
      "learning_rate": 0.0004823935125439912,
      "loss": 4.9325,
      "mean_token_accuracy": 0.204210102558136,
      "num_tokens": 29527618.0,
      "step": 16050
    },
    {
      "entropy": 5.137830018997192,
      "epoch": 1.3410457734714334,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000482381966313578,
      "loss": 4.9469,
      "mean_token_accuracy": 0.20344625860452653,
      "num_tokens": 29536433.0,
      "step": 16055
    },
    {
      "entropy": 5.137209033966064,
      "epoch": 1.3414634146341464,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048237041645269387,
      "loss": 4.8341,
      "mean_token_accuracy": 0.21025864034891129,
      "num_tokens": 29545831.0,
      "step": 16060
    },
    {
      "entropy": 5.0668919563293455,
      "epoch": 1.3418810557968595,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004823588629615409,
      "loss": 4.8876,
      "mean_token_accuracy": 0.20891664773225785,
      "num_tokens": 29554259.0,
      "step": 16065
    },
    {
      "entropy": 5.195721578598023,
      "epoch": 1.3422986969595723,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048234730584032143,
      "loss": 5.0198,
      "mean_token_accuracy": 0.19688054621219636,
      "num_tokens": 29563626.0,
      "step": 16070
    },
    {
      "entropy": 5.240046501159668,
      "epoch": 1.3427163381222853,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004823357450892379,
      "loss": 4.9888,
      "mean_token_accuracy": 0.19834381341934204,
      "num_tokens": 29573598.0,
      "step": 16075
    },
    {
      "entropy": 5.079400825500488,
      "epoch": 1.3431339792849983,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004823241807084925,
      "loss": 4.8062,
      "mean_token_accuracy": 0.21316552758216858,
      "num_tokens": 29582031.0,
      "step": 16080
    },
    {
      "entropy": 5.114629459381104,
      "epoch": 1.3435516204477114,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048231261269828776,
      "loss": 5.0072,
      "mean_token_accuracy": 0.1971979558467865,
      "num_tokens": 29590910.0,
      "step": 16085
    },
    {
      "entropy": 5.168224763870239,
      "epoch": 1.3439692616104244,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048230104105882624,
      "loss": 4.9431,
      "mean_token_accuracy": 0.2062482625246048,
      "num_tokens": 29600094.0,
      "step": 16090
    },
    {
      "entropy": 5.227047872543335,
      "epoch": 1.3443869027731372,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004822894657903104,
      "loss": 4.95,
      "mean_token_accuracy": 0.20703855156898499,
      "num_tokens": 29608976.0,
      "step": 16095
    },
    {
      "entropy": 5.094570350646973,
      "epoch": 1.3448045439358502,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000482277886892943,
      "loss": 4.8647,
      "mean_token_accuracy": 0.20958521217107773,
      "num_tokens": 29617993.0,
      "step": 16100
    },
    {
      "entropy": 5.160855722427368,
      "epoch": 1.3452221850985633,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004822663043669267,
      "loss": 4.9157,
      "mean_token_accuracy": 0.2085632026195526,
      "num_tokens": 29627301.0,
      "step": 16105
    },
    {
      "entropy": 5.178886556625367,
      "epoch": 1.3456398262612763,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048225471821246416,
      "loss": 4.9511,
      "mean_token_accuracy": 0.20282212644815445,
      "num_tokens": 29636464.0,
      "step": 16110
    },
    {
      "entropy": 5.113752603530884,
      "epoch": 1.3460574674239894,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004822431284297584,
      "loss": 4.8938,
      "mean_token_accuracy": 0.20975109934806824,
      "num_tokens": 29645449.0,
      "step": 16115
    },
    {
      "entropy": 5.177691316604614,
      "epoch": 1.3464751085867024,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004822315350190122,
      "loss": 4.9222,
      "mean_token_accuracy": 0.21248852759599685,
      "num_tokens": 29654488.0,
      "step": 16120
    },
    {
      "entropy": 5.18659234046936,
      "epoch": 1.3468927497494154,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048221993798042856,
      "loss": 4.9307,
      "mean_token_accuracy": 0.20108346939086913,
      "num_tokens": 29664100.0,
      "step": 16125
    },
    {
      "entropy": 5.305808258056641,
      "epoch": 1.3473103909121282,
      "grad_norm": 1.25,
      "learning_rate": 0.00048220833731421045,
      "loss": 5.0125,
      "mean_token_accuracy": 0.2013043001294136,
      "num_tokens": 29673083.0,
      "step": 16130
    },
    {
      "entropy": 5.152714967727661,
      "epoch": 1.3477280320748413,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004821967330205609,
      "loss": 4.8259,
      "mean_token_accuracy": 0.21110471785068513,
      "num_tokens": 29681818.0,
      "step": 16135
    },
    {
      "entropy": 5.079560089111328,
      "epoch": 1.3481456732375543,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004821851250996832,
      "loss": 4.885,
      "mean_token_accuracy": 0.20244283825159073,
      "num_tokens": 29690884.0,
      "step": 16140
    },
    {
      "entropy": 5.157465410232544,
      "epoch": 1.3485633144002673,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004821735135517805,
      "loss": 4.8772,
      "mean_token_accuracy": 0.20906876176595687,
      "num_tokens": 29699697.0,
      "step": 16145
    },
    {
      "entropy": 5.1897481918334964,
      "epoch": 1.3489809555629804,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004821618983770561,
      "loss": 4.9437,
      "mean_token_accuracy": 0.20448482632637024,
      "num_tokens": 29708914.0,
      "step": 16150
    },
    {
      "entropy": 5.0875598907470705,
      "epoch": 1.3493985967256932,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048215027957571326,
      "loss": 4.8531,
      "mean_token_accuracy": 0.2099583104252815,
      "num_tokens": 29717108.0,
      "step": 16155
    },
    {
      "entropy": 5.1992264747619625,
      "epoch": 1.3498162378884062,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004821386571479555,
      "loss": 4.9372,
      "mean_token_accuracy": 0.20413416624069214,
      "num_tokens": 29727264.0,
      "step": 16160
    },
    {
      "entropy": 5.160131168365479,
      "epoch": 1.3502338790511192,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004821270310939861,
      "loss": 4.8629,
      "mean_token_accuracy": 0.2072824716567993,
      "num_tokens": 29735746.0,
      "step": 16165
    },
    {
      "entropy": 5.192693138122559,
      "epoch": 1.3506515202138323,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004821154014140088,
      "loss": 4.9988,
      "mean_token_accuracy": 0.20295075923204423,
      "num_tokens": 29744877.0,
      "step": 16170
    },
    {
      "entropy": 5.188033151626587,
      "epoch": 1.3510691613765453,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004821037681082271,
      "loss": 4.9363,
      "mean_token_accuracy": 0.19716689884662628,
      "num_tokens": 29753268.0,
      "step": 16175
    },
    {
      "entropy": 5.162658548355102,
      "epoch": 1.3514868025392583,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004820921311768446,
      "loss": 4.8176,
      "mean_token_accuracy": 0.20503176301717757,
      "num_tokens": 29762242.0,
      "step": 16180
    },
    {
      "entropy": 5.125366115570069,
      "epoch": 1.3519044437019714,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048208049062006505,
      "loss": 4.943,
      "mean_token_accuracy": 0.20199419558048248,
      "num_tokens": 29771267.0,
      "step": 16185
    },
    {
      "entropy": 5.142292642593384,
      "epoch": 1.3523220848646842,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048206884643809224,
      "loss": 4.8926,
      "mean_token_accuracy": 0.21183663755655288,
      "num_tokens": 29779883.0,
      "step": 16190
    },
    {
      "entropy": 5.2082562923431395,
      "epoch": 1.3527397260273972,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048205719863113014,
      "loss": 4.9376,
      "mean_token_accuracy": 0.20122295171022414,
      "num_tokens": 29788588.0,
      "step": 16195
    },
    {
      "entropy": 5.084494256973267,
      "epoch": 1.3531573671901103,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004820455471993824,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2173047825694084,
      "num_tokens": 29797023.0,
      "step": 16200
    },
    {
      "entropy": 5.15872712135315,
      "epoch": 1.3535750083528233,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048203389214305326,
      "loss": 4.923,
      "mean_token_accuracy": 0.20640652030706405,
      "num_tokens": 29805735.0,
      "step": 16205
    },
    {
      "entropy": 5.131176137924195,
      "epoch": 1.3539926495155363,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004820222334623465,
      "loss": 4.9106,
      "mean_token_accuracy": 0.2055223822593689,
      "num_tokens": 29814564.0,
      "step": 16210
    },
    {
      "entropy": 5.158227491378784,
      "epoch": 1.3544102906782491,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048201057115746643,
      "loss": 4.9158,
      "mean_token_accuracy": 0.2086961954832077,
      "num_tokens": 29823801.0,
      "step": 16215
    },
    {
      "entropy": 5.179012298583984,
      "epoch": 1.3548279318409622,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004819989052286172,
      "loss": 5.0021,
      "mean_token_accuracy": 0.2044561043381691,
      "num_tokens": 29832846.0,
      "step": 16220
    },
    {
      "entropy": 5.177942132949829,
      "epoch": 1.3552455730036752,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004819872356760028,
      "loss": 4.9137,
      "mean_token_accuracy": 0.20588994920253753,
      "num_tokens": 29841573.0,
      "step": 16225
    },
    {
      "entropy": 5.163326454162598,
      "epoch": 1.3556632141663882,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048197556249982774,
      "loss": 5.0466,
      "mean_token_accuracy": 0.19953929036855697,
      "num_tokens": 29850669.0,
      "step": 16230
    },
    {
      "entropy": 5.145029735565186,
      "epoch": 1.3560808553291013,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004819638857002965,
      "loss": 4.8697,
      "mean_token_accuracy": 0.21525536179542543,
      "num_tokens": 29859389.0,
      "step": 16235
    },
    {
      "entropy": 5.148257875442505,
      "epoch": 1.3564984964918143,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048195220527761306,
      "loss": 4.9224,
      "mean_token_accuracy": 0.2039658844470978,
      "num_tokens": 29867891.0,
      "step": 16240
    },
    {
      "entropy": 5.126183748245239,
      "epoch": 1.3569161376545273,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048194052123198236,
      "loss": 4.9146,
      "mean_token_accuracy": 0.2075326770544052,
      "num_tokens": 29876839.0,
      "step": 16245
    },
    {
      "entropy": 5.142038583755493,
      "epoch": 1.3573337788172402,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048192883356360867,
      "loss": 4.9013,
      "mean_token_accuracy": 0.2068633943796158,
      "num_tokens": 29885992.0,
      "step": 16250
    },
    {
      "entropy": 5.113740015029907,
      "epoch": 1.3577514199799532,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004819171422726966,
      "loss": 4.817,
      "mean_token_accuracy": 0.21507459431886672,
      "num_tokens": 29894156.0,
      "step": 16255
    },
    {
      "entropy": 5.185928916931152,
      "epoch": 1.3581690611426662,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048190544735945094,
      "loss": 4.9903,
      "mean_token_accuracy": 0.1992097720503807,
      "num_tokens": 29903892.0,
      "step": 16260
    },
    {
      "entropy": 5.1530365467071535,
      "epoch": 1.3585867023053793,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004818937488240764,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21017599999904632,
      "num_tokens": 29913516.0,
      "step": 16265
    },
    {
      "entropy": 5.1330641746521,
      "epoch": 1.3590043434680923,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048188204666677775,
      "loss": 4.9798,
      "mean_token_accuracy": 0.2015740305185318,
      "num_tokens": 29922590.0,
      "step": 16270
    },
    {
      "entropy": 5.142926025390625,
      "epoch": 1.359421984630805,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004818703408877598,
      "loss": 4.8942,
      "mean_token_accuracy": 0.21273099035024642,
      "num_tokens": 29931423.0,
      "step": 16275
    },
    {
      "entropy": 5.115842533111572,
      "epoch": 1.3598396257935181,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004818586314872276,
      "loss": 4.9147,
      "mean_token_accuracy": 0.21202665865421294,
      "num_tokens": 29940386.0,
      "step": 16280
    },
    {
      "entropy": 5.220868968963623,
      "epoch": 1.3602572669562312,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048184691846538596,
      "loss": 5.0232,
      "mean_token_accuracy": 0.20771065652370452,
      "num_tokens": 29948939.0,
      "step": 16285
    },
    {
      "entropy": 5.186430644989014,
      "epoch": 1.3606749081189442,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048183520182244006,
      "loss": 4.9501,
      "mean_token_accuracy": 0.2085529386997223,
      "num_tokens": 29957892.0,
      "step": 16290
    },
    {
      "entropy": 5.192347860336303,
      "epoch": 1.3610925492816572,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000481823481558595,
      "loss": 4.9618,
      "mean_token_accuracy": 0.1972670763731003,
      "num_tokens": 29967419.0,
      "step": 16295
    },
    {
      "entropy": 5.283215522766113,
      "epoch": 1.3615101904443703,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048181175767405594,
      "loss": 5.0625,
      "mean_token_accuracy": 0.20170041024684907,
      "num_tokens": 29976817.0,
      "step": 16300
    },
    {
      "entropy": 5.224116373062134,
      "epoch": 1.3619278316070833,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004818000301690282,
      "loss": 4.9156,
      "mean_token_accuracy": 0.21054738909006118,
      "num_tokens": 29985572.0,
      "step": 16305
    },
    {
      "entropy": 5.041603994369507,
      "epoch": 1.3623454727697961,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004817882990437169,
      "loss": 4.8501,
      "mean_token_accuracy": 0.2129541128873825,
      "num_tokens": 29994753.0,
      "step": 16310
    },
    {
      "entropy": 5.072797012329102,
      "epoch": 1.3627631139325092,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048177656429832763,
      "loss": 4.8989,
      "mean_token_accuracy": 0.20649760514497756,
      "num_tokens": 30003840.0,
      "step": 16315
    },
    {
      "entropy": 5.132313346862793,
      "epoch": 1.3631807550952222,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048176482593306557,
      "loss": 4.8904,
      "mean_token_accuracy": 0.21488573998212815,
      "num_tokens": 30013880.0,
      "step": 16320
    },
    {
      "entropy": 5.201842880249023,
      "epoch": 1.3635983962579352,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048175308394813644,
      "loss": 4.9616,
      "mean_token_accuracy": 0.2059384688735008,
      "num_tokens": 30022636.0,
      "step": 16325
    },
    {
      "entropy": 5.190887928009033,
      "epoch": 1.3640160374206483,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004817413383437456,
      "loss": 4.949,
      "mean_token_accuracy": 0.21349097341299056,
      "num_tokens": 30031547.0,
      "step": 16330
    },
    {
      "entropy": 5.099081993103027,
      "epoch": 1.364433678583361,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048172958912009886,
      "loss": 4.9533,
      "mean_token_accuracy": 0.20203142911195754,
      "num_tokens": 30041603.0,
      "step": 16335
    },
    {
      "entropy": 5.163622283935547,
      "epoch": 1.364851319746074,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048171783627740176,
      "loss": 4.8996,
      "mean_token_accuracy": 0.20617176592350006,
      "num_tokens": 30051894.0,
      "step": 16340
    },
    {
      "entropy": 5.1732151985168455,
      "epoch": 1.3652689609087871,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048170607981586023,
      "loss": 4.9584,
      "mean_token_accuracy": 0.20691109001636504,
      "num_tokens": 30061782.0,
      "step": 16345
    },
    {
      "entropy": 5.180705976486206,
      "epoch": 1.3656866020715002,
      "grad_norm": 1.625,
      "learning_rate": 0.0004816943197356799,
      "loss": 4.9612,
      "mean_token_accuracy": 0.19670221954584122,
      "num_tokens": 30072051.0,
      "step": 16350
    },
    {
      "entropy": 5.165250396728515,
      "epoch": 1.3661042432342132,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004816825560370667,
      "loss": 4.9412,
      "mean_token_accuracy": 0.20063096433877944,
      "num_tokens": 30082077.0,
      "step": 16355
    },
    {
      "entropy": 5.147458791732788,
      "epoch": 1.3665218843969262,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048167078872022657,
      "loss": 4.9026,
      "mean_token_accuracy": 0.20955607891082764,
      "num_tokens": 30090982.0,
      "step": 16360
    },
    {
      "entropy": 5.233877182006836,
      "epoch": 1.3669395255596393,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048165901778536547,
      "loss": 5.0983,
      "mean_token_accuracy": 0.1993892788887024,
      "num_tokens": 30100688.0,
      "step": 16365
    },
    {
      "entropy": 5.191025018692017,
      "epoch": 1.367357166722352,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048164724323268956,
      "loss": 4.9242,
      "mean_token_accuracy": 0.2080811530351639,
      "num_tokens": 30109248.0,
      "step": 16370
    },
    {
      "entropy": 5.168928861618042,
      "epoch": 1.3677748078850651,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048163546506240493,
      "loss": 4.9428,
      "mean_token_accuracy": 0.20328947454690932,
      "num_tokens": 30118850.0,
      "step": 16375
    },
    {
      "entropy": 5.256982612609863,
      "epoch": 1.3681924490477781,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004816236832747177,
      "loss": 4.984,
      "mean_token_accuracy": 0.1989699751138687,
      "num_tokens": 30128007.0,
      "step": 16380
    },
    {
      "entropy": 5.167699193954467,
      "epoch": 1.3686100902104912,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004816118978698342,
      "loss": 4.9812,
      "mean_token_accuracy": 0.20199739336967468,
      "num_tokens": 30137944.0,
      "step": 16385
    },
    {
      "entropy": 5.1321436882019045,
      "epoch": 1.3690277313732042,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004816001088479608,
      "loss": 4.9334,
      "mean_token_accuracy": 0.2082753449678421,
      "num_tokens": 30147378.0,
      "step": 16390
    },
    {
      "entropy": 5.057757043838501,
      "epoch": 1.369445372535917,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004815883162093037,
      "loss": 4.7572,
      "mean_token_accuracy": 0.21449211686849595,
      "num_tokens": 30156443.0,
      "step": 16395
    },
    {
      "entropy": 5.116669082641602,
      "epoch": 1.36986301369863,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048157651995406954,
      "loss": 4.8446,
      "mean_token_accuracy": 0.2158801957964897,
      "num_tokens": 30164937.0,
      "step": 16400
    },
    {
      "entropy": 5.096645450592041,
      "epoch": 1.370280654861343,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004815647200824647,
      "loss": 4.8147,
      "mean_token_accuracy": 0.2149915024638176,
      "num_tokens": 30173696.0,
      "step": 16405
    },
    {
      "entropy": 5.199629068374634,
      "epoch": 1.3706982960240561,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048155291659469587,
      "loss": 4.9736,
      "mean_token_accuracy": 0.20166016072034837,
      "num_tokens": 30183371.0,
      "step": 16410
    },
    {
      "entropy": 5.087919139862061,
      "epoch": 1.3711159371867692,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048154110949096945,
      "loss": 4.8531,
      "mean_token_accuracy": 0.21351614594459534,
      "num_tokens": 30191987.0,
      "step": 16415
    },
    {
      "entropy": 5.102304601669312,
      "epoch": 1.3715335783494822,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004815292987714924,
      "loss": 4.8631,
      "mean_token_accuracy": 0.20869213789701463,
      "num_tokens": 30200936.0,
      "step": 16420
    },
    {
      "entropy": 5.209830570220947,
      "epoch": 1.3719512195121952,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004815174844364714,
      "loss": 4.9927,
      "mean_token_accuracy": 0.21001094579696655,
      "num_tokens": 30210371.0,
      "step": 16425
    },
    {
      "entropy": 5.08905668258667,
      "epoch": 1.372368860674908,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004815056664861133,
      "loss": 4.8752,
      "mean_token_accuracy": 0.20841034501791,
      "num_tokens": 30220426.0,
      "step": 16430
    },
    {
      "entropy": 5.070647144317627,
      "epoch": 1.372786501837621,
      "grad_norm": 1.375,
      "learning_rate": 0.0004814938449206248,
      "loss": 4.8487,
      "mean_token_accuracy": 0.2123531922698021,
      "num_tokens": 30230042.0,
      "step": 16435
    },
    {
      "entropy": 5.144211196899414,
      "epoch": 1.373204143000334,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048148201974021297,
      "loss": 4.8198,
      "mean_token_accuracy": 0.2074206218123436,
      "num_tokens": 30238356.0,
      "step": 16440
    },
    {
      "entropy": 5.17895655632019,
      "epoch": 1.3736217841630471,
      "grad_norm": 1.1875,
      "learning_rate": 0.000481470190945085,
      "loss": 4.9984,
      "mean_token_accuracy": 0.19720254838466644,
      "num_tokens": 30247260.0,
      "step": 16445
    },
    {
      "entropy": 5.11223874092102,
      "epoch": 1.3740394253257602,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004814583585354477,
      "loss": 4.8288,
      "mean_token_accuracy": 0.21131580471992492,
      "num_tokens": 30255892.0,
      "step": 16450
    },
    {
      "entropy": 5.1979148387908936,
      "epoch": 1.374457066488473,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048144652251150836,
      "loss": 4.966,
      "mean_token_accuracy": 0.1944601833820343,
      "num_tokens": 30265970.0,
      "step": 16455
    },
    {
      "entropy": 5.168201017379761,
      "epoch": 1.374874707651186,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000481434682873474,
      "loss": 4.9075,
      "mean_token_accuracy": 0.2093770310282707,
      "num_tokens": 30276339.0,
      "step": 16460
    },
    {
      "entropy": 5.144068717956543,
      "epoch": 1.375292348813899,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004814228396215522,
      "loss": 4.9193,
      "mean_token_accuracy": 0.2039335012435913,
      "num_tokens": 30285748.0,
      "step": 16465
    },
    {
      "entropy": 5.136648559570313,
      "epoch": 1.375709989976612,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004814109927559501,
      "loss": 4.8666,
      "mean_token_accuracy": 0.20718628615140916,
      "num_tokens": 30294953.0,
      "step": 16470
    },
    {
      "entropy": 5.1966756820678714,
      "epoch": 1.3761276311393251,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004813991422768751,
      "loss": 4.9421,
      "mean_token_accuracy": 0.20977271646261214,
      "num_tokens": 30303889.0,
      "step": 16475
    },
    {
      "entropy": 5.115855264663696,
      "epoch": 1.3765452723020382,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048138728818453456,
      "loss": 4.932,
      "mean_token_accuracy": 0.210543717443943,
      "num_tokens": 30312541.0,
      "step": 16480
    },
    {
      "entropy": 5.159768342971802,
      "epoch": 1.3769629134647512,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048137543047913636,
      "loss": 5.0215,
      "mean_token_accuracy": 0.201439069211483,
      "num_tokens": 30322804.0,
      "step": 16485
    },
    {
      "entropy": 5.2314528465271,
      "epoch": 1.377380554627464,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048136356916088764,
      "loss": 5.015,
      "mean_token_accuracy": 0.19645354002714158,
      "num_tokens": 30333398.0,
      "step": 16490
    },
    {
      "entropy": 5.120488214492798,
      "epoch": 1.377798195790177,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048135170422999624,
      "loss": 4.8199,
      "mean_token_accuracy": 0.20942801833152772,
      "num_tokens": 30342719.0,
      "step": 16495
    },
    {
      "entropy": 5.192204427719116,
      "epoch": 1.37821583695289,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048133983568666994,
      "loss": 4.9433,
      "mean_token_accuracy": 0.20340341776609422,
      "num_tokens": 30352314.0,
      "step": 16500
    },
    {
      "entropy": 5.143529653549194,
      "epoch": 1.378633478115603,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004813279635311165,
      "loss": 4.8841,
      "mean_token_accuracy": 0.21334562003612517,
      "num_tokens": 30361425.0,
      "step": 16505
    },
    {
      "entropy": 5.139241409301758,
      "epoch": 1.3790511192783161,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004813160877635436,
      "loss": 4.9738,
      "mean_token_accuracy": 0.2085084155201912,
      "num_tokens": 30370437.0,
      "step": 16510
    },
    {
      "entropy": 5.159690999984742,
      "epoch": 1.379468760441029,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048130420838415926,
      "loss": 4.9172,
      "mean_token_accuracy": 0.20858239531517028,
      "num_tokens": 30379479.0,
      "step": 16515
    },
    {
      "entropy": 5.167208051681518,
      "epoch": 1.379886401603742,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048129232539317147,
      "loss": 4.9356,
      "mean_token_accuracy": 0.20677457600831986,
      "num_tokens": 30388426.0,
      "step": 16520
    },
    {
      "entropy": 5.138595008850098,
      "epoch": 1.380304042766455,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048128043879078816,
      "loss": 4.9492,
      "mean_token_accuracy": 0.2117907702922821,
      "num_tokens": 30397359.0,
      "step": 16525
    },
    {
      "entropy": 5.135281276702881,
      "epoch": 1.380721683929168,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004812685485772175,
      "loss": 4.9803,
      "mean_token_accuracy": 0.20798729360103607,
      "num_tokens": 30407639.0,
      "step": 16530
    },
    {
      "entropy": 5.114239311218261,
      "epoch": 1.381139325091881,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048125665475266756,
      "loss": 4.8129,
      "mean_token_accuracy": 0.22871217578649522,
      "num_tokens": 30416563.0,
      "step": 16535
    },
    {
      "entropy": 5.245465469360352,
      "epoch": 1.3815569662545941,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048124475731734667,
      "loss": 4.9503,
      "mean_token_accuracy": 0.20682310312986374,
      "num_tokens": 30425611.0,
      "step": 16540
    },
    {
      "entropy": 5.183871412277222,
      "epoch": 1.3819746074173072,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000481232856271463,
      "loss": 5.0086,
      "mean_token_accuracy": 0.20188721120357514,
      "num_tokens": 30434775.0,
      "step": 16545
    },
    {
      "entropy": 5.27626371383667,
      "epoch": 1.38239224858002,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048122095161522496,
      "loss": 4.9901,
      "mean_token_accuracy": 0.2015072822570801,
      "num_tokens": 30444655.0,
      "step": 16550
    },
    {
      "entropy": 5.141908168792725,
      "epoch": 1.382809889742733,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004812090433488409,
      "loss": 4.8981,
      "mean_token_accuracy": 0.21208438724279405,
      "num_tokens": 30454418.0,
      "step": 16555
    },
    {
      "entropy": 5.0903397560119625,
      "epoch": 1.383227530905446,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048119713147251944,
      "loss": 4.8842,
      "mean_token_accuracy": 0.21464357823133468,
      "num_tokens": 30463234.0,
      "step": 16560
    },
    {
      "entropy": 5.246651315689087,
      "epoch": 1.383645172068159,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048118521598646886,
      "loss": 4.9694,
      "mean_token_accuracy": 0.20048714876174928,
      "num_tokens": 30471864.0,
      "step": 16565
    },
    {
      "entropy": 5.083123254776001,
      "epoch": 1.384062813230872,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048117329689089793,
      "loss": 4.8521,
      "mean_token_accuracy": 0.21679978519678117,
      "num_tokens": 30481047.0,
      "step": 16570
    },
    {
      "entropy": 5.167449951171875,
      "epoch": 1.384480454393585,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048116137418601526,
      "loss": 5.029,
      "mean_token_accuracy": 0.1988441288471222,
      "num_tokens": 30491129.0,
      "step": 16575
    },
    {
      "entropy": 5.272493505477906,
      "epoch": 1.384898095556298,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048114944787202955,
      "loss": 5.0492,
      "mean_token_accuracy": 0.20064319670200348,
      "num_tokens": 30500466.0,
      "step": 16580
    },
    {
      "entropy": 5.152998447418213,
      "epoch": 1.385315736719011,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048113751794914974,
      "loss": 4.8343,
      "mean_token_accuracy": 0.2110416427254677,
      "num_tokens": 30509124.0,
      "step": 16585
    },
    {
      "entropy": 5.093537187576294,
      "epoch": 1.385733377881724,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048112558441758437,
      "loss": 4.9096,
      "mean_token_accuracy": 0.20436349511146545,
      "num_tokens": 30518251.0,
      "step": 16590
    },
    {
      "entropy": 5.093700504302978,
      "epoch": 1.386151019044437,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004811136472775426,
      "loss": 4.8564,
      "mean_token_accuracy": 0.20317798405885695,
      "num_tokens": 30526939.0,
      "step": 16595
    },
    {
      "entropy": 5.145337629318237,
      "epoch": 1.38656866020715,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004811017065292334,
      "loss": 4.9236,
      "mean_token_accuracy": 0.2116194859147072,
      "num_tokens": 30535867.0,
      "step": 16600
    },
    {
      "entropy": 5.129948139190674,
      "epoch": 1.3869863013698631,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048108976217286576,
      "loss": 4.9115,
      "mean_token_accuracy": 0.2057232990860939,
      "num_tokens": 30544194.0,
      "step": 16605
    },
    {
      "entropy": 5.207217264175415,
      "epoch": 1.387403942532576,
      "grad_norm": 1.25,
      "learning_rate": 0.0004810778142086487,
      "loss": 4.9322,
      "mean_token_accuracy": 0.20967212468385696,
      "num_tokens": 30553373.0,
      "step": 16610
    },
    {
      "entropy": 5.117929601669312,
      "epoch": 1.387821583695289,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048106586263679146,
      "loss": 4.8497,
      "mean_token_accuracy": 0.2179664596915245,
      "num_tokens": 30562413.0,
      "step": 16615
    },
    {
      "entropy": 5.216776657104492,
      "epoch": 1.388239224858002,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048105390745750324,
      "loss": 5.0039,
      "mean_token_accuracy": 0.20191051959991455,
      "num_tokens": 30572525.0,
      "step": 16620
    },
    {
      "entropy": 5.117004919052124,
      "epoch": 1.388656866020715,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048104194867099336,
      "loss": 4.9762,
      "mean_token_accuracy": 0.203702612221241,
      "num_tokens": 30581616.0,
      "step": 16625
    },
    {
      "entropy": 5.143327617645264,
      "epoch": 1.389074507183428,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048102998627747116,
      "loss": 4.922,
      "mean_token_accuracy": 0.20919372141361237,
      "num_tokens": 30590834.0,
      "step": 16630
    },
    {
      "entropy": 5.159092760086059,
      "epoch": 1.3894921483461409,
      "grad_norm": 1.296875,
      "learning_rate": 0.000481018020277146,
      "loss": 4.9098,
      "mean_token_accuracy": 0.20935335904359817,
      "num_tokens": 30600326.0,
      "step": 16635
    },
    {
      "entropy": 5.084617710113525,
      "epoch": 1.389909789508854,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048100605067022753,
      "loss": 4.7842,
      "mean_token_accuracy": 0.21784369498491288,
      "num_tokens": 30608757.0,
      "step": 16640
    },
    {
      "entropy": 5.135858392715454,
      "epoch": 1.390327430671567,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048099407745692504,
      "loss": 4.924,
      "mean_token_accuracy": 0.2098824143409729,
      "num_tokens": 30617359.0,
      "step": 16645
    },
    {
      "entropy": 5.16234073638916,
      "epoch": 1.39074507183428,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048098210063744827,
      "loss": 4.9963,
      "mean_token_accuracy": 0.19755117297172547,
      "num_tokens": 30626687.0,
      "step": 16650
    },
    {
      "entropy": 5.1057390689849855,
      "epoch": 1.391162712996993,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048097012021200696,
      "loss": 4.9075,
      "mean_token_accuracy": 0.21017294824123384,
      "num_tokens": 30636665.0,
      "step": 16655
    },
    {
      "entropy": 5.1467071056365965,
      "epoch": 1.391580354159706,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004809581361808108,
      "loss": 4.9519,
      "mean_token_accuracy": 0.2055586352944374,
      "num_tokens": 30646140.0,
      "step": 16660
    },
    {
      "entropy": 5.231961488723755,
      "epoch": 1.391997995322419,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004809461485440694,
      "loss": 4.9304,
      "mean_token_accuracy": 0.20902494639158248,
      "num_tokens": 30655483.0,
      "step": 16665
    },
    {
      "entropy": 5.135117149353027,
      "epoch": 1.392415636485132,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004809341573019929,
      "loss": 4.9096,
      "mean_token_accuracy": 0.20480920970439911,
      "num_tokens": 30664706.0,
      "step": 16670
    },
    {
      "entropy": 5.1258604526519775,
      "epoch": 1.392833277647845,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004809221624547911,
      "loss": 4.8158,
      "mean_token_accuracy": 0.21376144289970397,
      "num_tokens": 30672979.0,
      "step": 16675
    },
    {
      "entropy": 5.147611999511719,
      "epoch": 1.393250918810558,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000480910164002674,
      "loss": 4.9448,
      "mean_token_accuracy": 0.20309599339962006,
      "num_tokens": 30683345.0,
      "step": 16680
    },
    {
      "entropy": 5.152550125122071,
      "epoch": 1.393668559973271,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048089816194585153,
      "loss": 4.9031,
      "mean_token_accuracy": 0.21067062467336656,
      "num_tokens": 30692492.0,
      "step": 16685
    },
    {
      "entropy": 5.202511739730835,
      "epoch": 1.394086201135984,
      "grad_norm": 1.25,
      "learning_rate": 0.0004808861562845339,
      "loss": 4.93,
      "mean_token_accuracy": 0.21248593628406526,
      "num_tokens": 30701575.0,
      "step": 16690
    },
    {
      "entropy": 5.156600046157837,
      "epoch": 1.3945038422986968,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004808741470189313,
      "loss": 4.9047,
      "mean_token_accuracy": 0.20895194113254548,
      "num_tokens": 30711905.0,
      "step": 16695
    },
    {
      "entropy": 5.1302077293396,
      "epoch": 1.39492148346141,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004808621341492541,
      "loss": 4.9503,
      "mean_token_accuracy": 0.20465278178453444,
      "num_tokens": 30721705.0,
      "step": 16700
    },
    {
      "entropy": 5.126604700088501,
      "epoch": 1.395339124624123,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004808501176757122,
      "loss": 4.8738,
      "mean_token_accuracy": 0.20884339660406112,
      "num_tokens": 30730123.0,
      "step": 16705
    },
    {
      "entropy": 5.1525962352752686,
      "epoch": 1.395756765786836,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004808380975985163,
      "loss": 4.8896,
      "mean_token_accuracy": 0.20468977987766265,
      "num_tokens": 30738851.0,
      "step": 16710
    },
    {
      "entropy": 5.100785207748413,
      "epoch": 1.396174406949549,
      "grad_norm": 1.25,
      "learning_rate": 0.0004808260739178768,
      "loss": 4.8264,
      "mean_token_accuracy": 0.20919316709041597,
      "num_tokens": 30748324.0,
      "step": 16715
    },
    {
      "entropy": 5.1314221858978275,
      "epoch": 1.396592048112262,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004808140466340041,
      "loss": 4.8638,
      "mean_token_accuracy": 0.21346567571163177,
      "num_tokens": 30757100.0,
      "step": 16720
    },
    {
      "entropy": 5.135165691375732,
      "epoch": 1.397009689274975,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004808020157471087,
      "loss": 4.9478,
      "mean_token_accuracy": 0.2074833407998085,
      "num_tokens": 30766968.0,
      "step": 16725
    },
    {
      "entropy": 5.151693296432495,
      "epoch": 1.3974273304376879,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048078998125740136,
      "loss": 4.909,
      "mean_token_accuracy": 0.20336152762174606,
      "num_tokens": 30776323.0,
      "step": 16730
    },
    {
      "entropy": 5.247948884963989,
      "epoch": 1.3978449716004009,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004807779431650927,
      "loss": 4.9751,
      "mean_token_accuracy": 0.20901384204626083,
      "num_tokens": 30785730.0,
      "step": 16735
    },
    {
      "entropy": 5.193715286254883,
      "epoch": 1.398262612763114,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004807659014703934,
      "loss": 4.8932,
      "mean_token_accuracy": 0.21295734792947768,
      "num_tokens": 30795172.0,
      "step": 16740
    },
    {
      "entropy": 5.13029375076294,
      "epoch": 1.398680253925827,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004807538561735143,
      "loss": 4.9037,
      "mean_token_accuracy": 0.20978702753782272,
      "num_tokens": 30803747.0,
      "step": 16745
    },
    {
      "entropy": 5.128589630126953,
      "epoch": 1.39909789508854,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048074180727466636,
      "loss": 4.8997,
      "mean_token_accuracy": 0.21181560605764388,
      "num_tokens": 30813687.0,
      "step": 16750
    },
    {
      "entropy": 5.141436529159546,
      "epoch": 1.3995155362512528,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048072975477406036,
      "loss": 4.9352,
      "mean_token_accuracy": 0.21264829188585282,
      "num_tokens": 30823284.0,
      "step": 16755
    },
    {
      "entropy": 5.13640832901001,
      "epoch": 1.399933177413966,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004807176986719074,
      "loss": 4.9152,
      "mean_token_accuracy": 0.2029614508152008,
      "num_tokens": 30832152.0,
      "step": 16760
    },
    {
      "entropy": 5.096826839447021,
      "epoch": 1.4003508185766789,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048070563896841843,
      "loss": 4.9053,
      "mean_token_accuracy": 0.21078179776668549,
      "num_tokens": 30841275.0,
      "step": 16765
    },
    {
      "entropy": 5.1680107593536375,
      "epoch": 1.400768459739392,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048069357566380474,
      "loss": 4.8962,
      "mean_token_accuracy": 0.21673545390367507,
      "num_tokens": 30850698.0,
      "step": 16770
    },
    {
      "entropy": 5.148619318008423,
      "epoch": 1.401186100902105,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048068150875827737,
      "loss": 4.9591,
      "mean_token_accuracy": 0.20866574198007584,
      "num_tokens": 30860212.0,
      "step": 16775
    },
    {
      "entropy": 5.1729597568511965,
      "epoch": 1.401603742064818,
      "grad_norm": 1.25,
      "learning_rate": 0.00048066943825204764,
      "loss": 5.0027,
      "mean_token_accuracy": 0.19625453352928163,
      "num_tokens": 30870005.0,
      "step": 16780
    },
    {
      "entropy": 5.171029186248779,
      "epoch": 1.402021383227531,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004806573641453268,
      "loss": 4.7866,
      "mean_token_accuracy": 0.20917522460222243,
      "num_tokens": 30878081.0,
      "step": 16785
    },
    {
      "entropy": 5.098481035232544,
      "epoch": 1.4024390243902438,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048064528643832627,
      "loss": 4.8968,
      "mean_token_accuracy": 0.20847069919109346,
      "num_tokens": 30886798.0,
      "step": 16790
    },
    {
      "entropy": 5.140516138076782,
      "epoch": 1.4028566655529568,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048063320513125747,
      "loss": 4.9179,
      "mean_token_accuracy": 0.20525728911161423,
      "num_tokens": 30897154.0,
      "step": 16795
    },
    {
      "entropy": 5.156975746154785,
      "epoch": 1.4032743067156699,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004806211202243319,
      "loss": 4.9029,
      "mean_token_accuracy": 0.2110048934817314,
      "num_tokens": 30905920.0,
      "step": 16800
    },
    {
      "entropy": 5.087568759918213,
      "epoch": 1.403691947878383,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048060903171776113,
      "loss": 5.0163,
      "mean_token_accuracy": 0.20063271522521972,
      "num_tokens": 30915539.0,
      "step": 16805
    },
    {
      "entropy": 5.224270391464233,
      "epoch": 1.404109589041096,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004805969396117567,
      "loss": 5.0036,
      "mean_token_accuracy": 0.20184151083230972,
      "num_tokens": 30926283.0,
      "step": 16810
    },
    {
      "entropy": 5.1911403179168705,
      "epoch": 1.4045272302038088,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048058484390653037,
      "loss": 4.8415,
      "mean_token_accuracy": 0.21320489048957825,
      "num_tokens": 30934953.0,
      "step": 16815
    },
    {
      "entropy": 5.137839317321777,
      "epoch": 1.404944871366522,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004805727446022939,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21202368587255477,
      "num_tokens": 30943848.0,
      "step": 16820
    },
    {
      "entropy": 5.086958169937134,
      "epoch": 1.4053625125292348,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048056064169925904,
      "loss": 4.8947,
      "mean_token_accuracy": 0.21113825142383574,
      "num_tokens": 30952226.0,
      "step": 16825
    },
    {
      "entropy": 5.142522478103638,
      "epoch": 1.4057801536919479,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048054853519763786,
      "loss": 4.9233,
      "mean_token_accuracy": 0.2073330655694008,
      "num_tokens": 30961783.0,
      "step": 16830
    },
    {
      "entropy": 5.222301435470581,
      "epoch": 1.406197794854661,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004805364250976421,
      "loss": 4.9668,
      "mean_token_accuracy": 0.2048247680068016,
      "num_tokens": 30971055.0,
      "step": 16835
    },
    {
      "entropy": 5.189017200469971,
      "epoch": 1.406615436017374,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004805243113994837,
      "loss": 4.8953,
      "mean_token_accuracy": 0.21142600178718568,
      "num_tokens": 30979906.0,
      "step": 16840
    },
    {
      "entropy": 5.139331579208374,
      "epoch": 1.407033077180087,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004805121941033749,
      "loss": 4.9121,
      "mean_token_accuracy": 0.20673353523015975,
      "num_tokens": 30989430.0,
      "step": 16845
    },
    {
      "entropy": 5.1089293479919435,
      "epoch": 1.4074507183427998,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004805000732095278,
      "loss": 4.8789,
      "mean_token_accuracy": 0.2150939628481865,
      "num_tokens": 30999060.0,
      "step": 16850
    },
    {
      "entropy": 5.102806282043457,
      "epoch": 1.4078683595055128,
      "grad_norm": 1.25,
      "learning_rate": 0.0004804879487181546,
      "loss": 4.9039,
      "mean_token_accuracy": 0.20722644925117492,
      "num_tokens": 31008840.0,
      "step": 16855
    },
    {
      "entropy": 5.1845386028289795,
      "epoch": 1.4082860006682258,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004804758206294675,
      "loss": 4.9023,
      "mean_token_accuracy": 0.21171188801527024,
      "num_tokens": 31018409.0,
      "step": 16860
    },
    {
      "entropy": 5.120317792892456,
      "epoch": 1.4087036418309389,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004804636889436787,
      "loss": 4.8924,
      "mean_token_accuracy": 0.2057185247540474,
      "num_tokens": 31026896.0,
      "step": 16865
    },
    {
      "entropy": 5.095272397994995,
      "epoch": 1.409121282993652,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048045155366100085,
      "loss": 4.8725,
      "mean_token_accuracy": 0.20386771261692047,
      "num_tokens": 31037181.0,
      "step": 16870
    },
    {
      "entropy": 5.117877578735351,
      "epoch": 1.4095389241563647,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004804394147816462,
      "loss": 4.8015,
      "mean_token_accuracy": 0.21431331038475038,
      "num_tokens": 31046534.0,
      "step": 16875
    },
    {
      "entropy": 5.1610987186431885,
      "epoch": 1.409956565319078,
      "grad_norm": 1.375,
      "learning_rate": 0.00048042727230582726,
      "loss": 4.9315,
      "mean_token_accuracy": 0.21004485189914704,
      "num_tokens": 31055076.0,
      "step": 16880
    },
    {
      "entropy": 5.120778322219849,
      "epoch": 1.4103742064817908,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004804151262337568,
      "loss": 4.9848,
      "mean_token_accuracy": 0.1962988942861557,
      "num_tokens": 31064792.0,
      "step": 16885
    },
    {
      "entropy": 5.233624076843261,
      "epoch": 1.4107918476445038,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004804029765656471,
      "loss": 5.0345,
      "mean_token_accuracy": 0.20329535007476807,
      "num_tokens": 31073808.0,
      "step": 16890
    },
    {
      "entropy": 5.236541795730591,
      "epoch": 1.4112094888072169,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004803908233017111,
      "loss": 4.9533,
      "mean_token_accuracy": 0.20905306339263915,
      "num_tokens": 31083718.0,
      "step": 16895
    },
    {
      "entropy": 5.2223473072052,
      "epoch": 1.41162712996993,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004803786664421616,
      "loss": 4.9361,
      "mean_token_accuracy": 0.20355068743228913,
      "num_tokens": 31093269.0,
      "step": 16900
    },
    {
      "entropy": 5.1160272598266605,
      "epoch": 1.412044771132643,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048036650598721125,
      "loss": 4.8918,
      "mean_token_accuracy": 0.21980983912944793,
      "num_tokens": 31101848.0,
      "step": 16905
    },
    {
      "entropy": 5.157998132705688,
      "epoch": 1.4124624122953557,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048035434193707306,
      "loss": 5.0028,
      "mean_token_accuracy": 0.20985425412654876,
      "num_tokens": 31111902.0,
      "step": 16910
    },
    {
      "entropy": 5.196841144561768,
      "epoch": 1.4128800534580688,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048034217429195986,
      "loss": 4.9344,
      "mean_token_accuracy": 0.21144617199897767,
      "num_tokens": 31121545.0,
      "step": 16915
    },
    {
      "entropy": 5.199971294403076,
      "epoch": 1.4132976946207818,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048033000305208485,
      "loss": 4.9879,
      "mean_token_accuracy": 0.2000981315970421,
      "num_tokens": 31130808.0,
      "step": 16920
    },
    {
      "entropy": 5.221096992492676,
      "epoch": 1.4137153357834948,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048031782821766086,
      "loss": 4.9406,
      "mean_token_accuracy": 0.20666281282901763,
      "num_tokens": 31141433.0,
      "step": 16925
    },
    {
      "entropy": 5.091422748565674,
      "epoch": 1.4141329769462079,
      "grad_norm": 1.25,
      "learning_rate": 0.00048030564978890117,
      "loss": 4.7831,
      "mean_token_accuracy": 0.21708951890468597,
      "num_tokens": 31149696.0,
      "step": 16930
    },
    {
      "entropy": 5.117171382904052,
      "epoch": 1.4145506181089207,
      "grad_norm": 1.421875,
      "learning_rate": 0.000480293467766019,
      "loss": 4.9162,
      "mean_token_accuracy": 0.2077425315976143,
      "num_tokens": 31159143.0,
      "step": 16935
    },
    {
      "entropy": 5.0835559368133545,
      "epoch": 1.414968259271634,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004802812821492276,
      "loss": 4.9172,
      "mean_token_accuracy": 0.200777505338192,
      "num_tokens": 31168459.0,
      "step": 16940
    },
    {
      "entropy": 5.119722843170166,
      "epoch": 1.4153859004343468,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048026909293874016,
      "loss": 4.8411,
      "mean_token_accuracy": 0.2142619624733925,
      "num_tokens": 31176824.0,
      "step": 16945
    },
    {
      "entropy": 5.049099683761597,
      "epoch": 1.4158035415970598,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004802569001347702,
      "loss": 4.8702,
      "mean_token_accuracy": 0.2107143983244896,
      "num_tokens": 31186235.0,
      "step": 16950
    },
    {
      "entropy": 5.1796094417572025,
      "epoch": 1.4162211827597728,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048024470373753114,
      "loss": 4.9897,
      "mean_token_accuracy": 0.20210909247398376,
      "num_tokens": 31195783.0,
      "step": 16955
    },
    {
      "entropy": 5.169140434265136,
      "epoch": 1.4166388239224859,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048023250374723656,
      "loss": 4.8926,
      "mean_token_accuracy": 0.20608078688383102,
      "num_tokens": 31204028.0,
      "step": 16960
    },
    {
      "entropy": 5.151114797592163,
      "epoch": 1.417056465085199,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004802203001640998,
      "loss": 4.91,
      "mean_token_accuracy": 0.20357382148504258,
      "num_tokens": 31213281.0,
      "step": 16965
    },
    {
      "entropy": 5.153474760055542,
      "epoch": 1.4174741062479117,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004802080929883348,
      "loss": 4.94,
      "mean_token_accuracy": 0.20463399589061737,
      "num_tokens": 31221830.0,
      "step": 16970
    },
    {
      "entropy": 5.306595754623413,
      "epoch": 1.4178917474106247,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004801958822201551,
      "loss": 5.067,
      "mean_token_accuracy": 0.19519750624895096,
      "num_tokens": 31230825.0,
      "step": 16975
    },
    {
      "entropy": 5.120306873321534,
      "epoch": 1.4183093885733378,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048018366785977445,
      "loss": 4.9205,
      "mean_token_accuracy": 0.20807593762874604,
      "num_tokens": 31239672.0,
      "step": 16980
    },
    {
      "entropy": 5.076950740814209,
      "epoch": 1.4187270297360508,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048017144990740673,
      "loss": 4.8014,
      "mean_token_accuracy": 0.21418218910694123,
      "num_tokens": 31248311.0,
      "step": 16985
    },
    {
      "entropy": 5.170657968521118,
      "epoch": 1.4191446708987638,
      "grad_norm": 1.25,
      "learning_rate": 0.0004801592283632658,
      "loss": 4.9231,
      "mean_token_accuracy": 0.21834250390529633,
      "num_tokens": 31257276.0,
      "step": 16990
    },
    {
      "entropy": 5.111931228637696,
      "epoch": 1.4195623120614766,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048014700322756567,
      "loss": 4.8149,
      "mean_token_accuracy": 0.21428578794002534,
      "num_tokens": 31266533.0,
      "step": 16995
    },
    {
      "entropy": 5.10798659324646,
      "epoch": 1.41997995322419,
      "grad_norm": 1.25,
      "learning_rate": 0.0004801347745005203,
      "loss": 4.9175,
      "mean_token_accuracy": 0.20855195373296737,
      "num_tokens": 31276003.0,
      "step": 17000
    },
    {
      "entropy": 5.192242050170899,
      "epoch": 1.4203975943869027,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004801225421823437,
      "loss": 4.9276,
      "mean_token_accuracy": 0.2084476962685585,
      "num_tokens": 31285523.0,
      "step": 17005
    },
    {
      "entropy": 5.187716293334961,
      "epoch": 1.4208152355496158,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048011030627325015,
      "loss": 4.942,
      "mean_token_accuracy": 0.20709554255008697,
      "num_tokens": 31294756.0,
      "step": 17010
    },
    {
      "entropy": 5.147516298294067,
      "epoch": 1.4212328767123288,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048009806677345394,
      "loss": 4.8307,
      "mean_token_accuracy": 0.21012722253799437,
      "num_tokens": 31303343.0,
      "step": 17015
    },
    {
      "entropy": 5.092759418487549,
      "epoch": 1.4216505178750418,
      "grad_norm": 1.71875,
      "learning_rate": 0.000480085823683169,
      "loss": 4.9137,
      "mean_token_accuracy": 0.21803715825080872,
      "num_tokens": 31312941.0,
      "step": 17020
    },
    {
      "entropy": 5.0570989608764645,
      "epoch": 1.4220681590377549,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004800735770026099,
      "loss": 4.7881,
      "mean_token_accuracy": 0.2237013101577759,
      "num_tokens": 31320797.0,
      "step": 17025
    },
    {
      "entropy": 5.18113751411438,
      "epoch": 1.4224858002004677,
      "grad_norm": 1.296875,
      "learning_rate": 0.000480061326731991,
      "loss": 4.938,
      "mean_token_accuracy": 0.20186226367950438,
      "num_tokens": 31329470.0,
      "step": 17030
    },
    {
      "entropy": 5.174644804000854,
      "epoch": 1.4229034413631807,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004800490728715267,
      "loss": 4.9395,
      "mean_token_accuracy": 0.20418516993522645,
      "num_tokens": 31339449.0,
      "step": 17035
    },
    {
      "entropy": 5.1067687511444095,
      "epoch": 1.4233210825258937,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004800368154214317,
      "loss": 4.8003,
      "mean_token_accuracy": 0.20877888947725295,
      "num_tokens": 31349403.0,
      "step": 17040
    },
    {
      "entropy": 5.1766884326934814,
      "epoch": 1.4237387236886068,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004800245543819203,
      "loss": 4.8631,
      "mean_token_accuracy": 0.20014817863702775,
      "num_tokens": 31359281.0,
      "step": 17045
    },
    {
      "entropy": 5.1558126449584964,
      "epoch": 1.4241563648513198,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048001228975320736,
      "loss": 4.8433,
      "mean_token_accuracy": 0.21299343407154084,
      "num_tokens": 31367495.0,
      "step": 17050
    },
    {
      "entropy": 5.181714487075806,
      "epoch": 1.4245740060140326,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004800000215355075,
      "loss": 4.936,
      "mean_token_accuracy": 0.21249287724494934,
      "num_tokens": 31376428.0,
      "step": 17055
    },
    {
      "entropy": 5.191418027877807,
      "epoch": 1.4249916471767459,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004799877497290355,
      "loss": 4.9878,
      "mean_token_accuracy": 0.20017764419317247,
      "num_tokens": 31385174.0,
      "step": 17060
    },
    {
      "entropy": 5.202902317047119,
      "epoch": 1.4254092883394587,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047997547433400615,
      "loss": 4.9765,
      "mean_token_accuracy": 0.20231525599956512,
      "num_tokens": 31395023.0,
      "step": 17065
    },
    {
      "entropy": 5.050139760971069,
      "epoch": 1.4258269295021717,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004799631953506345,
      "loss": 4.7701,
      "mean_token_accuracy": 0.22117212265729905,
      "num_tokens": 31404666.0,
      "step": 17070
    },
    {
      "entropy": 5.178532409667969,
      "epoch": 1.4262445706648847,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047995091277913535,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20744012147188187,
      "num_tokens": 31414297.0,
      "step": 17075
    },
    {
      "entropy": 5.118920087814331,
      "epoch": 1.4266622118275978,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047993862661972373,
      "loss": 4.8643,
      "mean_token_accuracy": 0.21148476004600525,
      "num_tokens": 31424604.0,
      "step": 17080
    },
    {
      "entropy": 5.091216707229615,
      "epoch": 1.4270798529903108,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004799263368726148,
      "loss": 4.8719,
      "mean_token_accuracy": 0.21016548126935958,
      "num_tokens": 31433066.0,
      "step": 17085
    },
    {
      "entropy": 5.1635884761810305,
      "epoch": 1.4274974941530236,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004799140435380237,
      "loss": 4.9383,
      "mean_token_accuracy": 0.20176807641983033,
      "num_tokens": 31442624.0,
      "step": 17090
    },
    {
      "entropy": 5.265144205093383,
      "epoch": 1.4279151353157367,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047990174661616556,
      "loss": 4.9402,
      "mean_token_accuracy": 0.21229350417852402,
      "num_tokens": 31450688.0,
      "step": 17095
    },
    {
      "entropy": 5.12986569404602,
      "epoch": 1.4283327764784497,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047988944610725574,
      "loss": 4.957,
      "mean_token_accuracy": 0.21222704499959946,
      "num_tokens": 31460289.0,
      "step": 17100
    },
    {
      "entropy": 5.06997447013855,
      "epoch": 1.4287504176411627,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004798771420115095,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2111150309443474,
      "num_tokens": 31468560.0,
      "step": 17105
    },
    {
      "entropy": 5.155711698532104,
      "epoch": 1.4291680588038758,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004798648343291423,
      "loss": 4.9593,
      "mean_token_accuracy": 0.20460644066333772,
      "num_tokens": 31477544.0,
      "step": 17110
    },
    {
      "entropy": 5.120484256744385,
      "epoch": 1.4295856999665886,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047985252306036966,
      "loss": 4.7442,
      "mean_token_accuracy": 0.2209637463092804,
      "num_tokens": 31485937.0,
      "step": 17115
    },
    {
      "entropy": 5.083667898178101,
      "epoch": 1.4300033411293018,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047984020820540695,
      "loss": 4.8286,
      "mean_token_accuracy": 0.21215756982564926,
      "num_tokens": 31495558.0,
      "step": 17120
    },
    {
      "entropy": 5.053993844985962,
      "epoch": 1.4304209822920146,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047982788976446984,
      "loss": 4.907,
      "mean_token_accuracy": 0.20299692302942277,
      "num_tokens": 31503974.0,
      "step": 17125
    },
    {
      "entropy": 5.143328905105591,
      "epoch": 1.4308386234547277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000479815567737774,
      "loss": 4.9094,
      "mean_token_accuracy": 0.2048829272389412,
      "num_tokens": 31513476.0,
      "step": 17130
    },
    {
      "entropy": 5.172032594680786,
      "epoch": 1.4312562646174407,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000479803242125535,
      "loss": 4.983,
      "mean_token_accuracy": 0.2106518790125847,
      "num_tokens": 31521981.0,
      "step": 17135
    },
    {
      "entropy": 5.162804841995239,
      "epoch": 1.4316739057801537,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047979091292796886,
      "loss": 4.9458,
      "mean_token_accuracy": 0.20676501840353012,
      "num_tokens": 31531363.0,
      "step": 17140
    },
    {
      "entropy": 5.179101896286011,
      "epoch": 1.4320915469428668,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004797785801452911,
      "loss": 4.934,
      "mean_token_accuracy": 0.20354807674884795,
      "num_tokens": 31540527.0,
      "step": 17145
    },
    {
      "entropy": 5.144188737869262,
      "epoch": 1.4325091881055796,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004797662437777179,
      "loss": 4.9342,
      "mean_token_accuracy": 0.20799606591463088,
      "num_tokens": 31550247.0,
      "step": 17150
    },
    {
      "entropy": 5.122576475143433,
      "epoch": 1.4329268292682926,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047975390382546523,
      "loss": 4.939,
      "mean_token_accuracy": 0.2063236266374588,
      "num_tokens": 31558982.0,
      "step": 17155
    },
    {
      "entropy": 5.167500162124634,
      "epoch": 1.4333444704310057,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047974156028874894,
      "loss": 4.9293,
      "mean_token_accuracy": 0.2091868281364441,
      "num_tokens": 31568572.0,
      "step": 17160
    },
    {
      "entropy": 5.278171491622925,
      "epoch": 1.4337621115937187,
      "grad_norm": 1.25,
      "learning_rate": 0.0004797292131677851,
      "loss": 5.1191,
      "mean_token_accuracy": 0.19622637331485748,
      "num_tokens": 31578552.0,
      "step": 17165
    },
    {
      "entropy": 5.264607954025268,
      "epoch": 1.4341797527564317,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047971686246279014,
      "loss": 5.0437,
      "mean_token_accuracy": 0.19873125851154327,
      "num_tokens": 31587140.0,
      "step": 17170
    },
    {
      "entropy": 5.25016074180603,
      "epoch": 1.4345973939191445,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004797045081739799,
      "loss": 4.9572,
      "mean_token_accuracy": 0.20538575649261476,
      "num_tokens": 31596539.0,
      "step": 17175
    },
    {
      "entropy": 5.139680480957031,
      "epoch": 1.4350150350818578,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004796921503015709,
      "loss": 4.8941,
      "mean_token_accuracy": 0.20791810601949692,
      "num_tokens": 31605669.0,
      "step": 17180
    },
    {
      "entropy": 5.189978408813476,
      "epoch": 1.4354326762445706,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047967978884577944,
      "loss": 4.8967,
      "mean_token_accuracy": 0.21323980391025543,
      "num_tokens": 31614885.0,
      "step": 17185
    },
    {
      "entropy": 5.07644739151001,
      "epoch": 1.4358503174072836,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004796674238068219,
      "loss": 4.8778,
      "mean_token_accuracy": 0.21189096122980117,
      "num_tokens": 31625300.0,
      "step": 17190
    },
    {
      "entropy": 5.062536764144897,
      "epoch": 1.4362679585699967,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047965505518491474,
      "loss": 4.7851,
      "mean_token_accuracy": 0.21505019962787628,
      "num_tokens": 31634675.0,
      "step": 17195
    },
    {
      "entropy": 5.177867269515991,
      "epoch": 1.4366855997327097,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047964268298027456,
      "loss": 4.9234,
      "mean_token_accuracy": 0.20473503321409225,
      "num_tokens": 31643959.0,
      "step": 17200
    },
    {
      "entropy": 5.146200704574585,
      "epoch": 1.4371032408954227,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047963030719311777,
      "loss": 4.9461,
      "mean_token_accuracy": 0.20366205424070358,
      "num_tokens": 31653423.0,
      "step": 17205
    },
    {
      "entropy": 5.068150377273559,
      "epoch": 1.4375208820581356,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004796179278236612,
      "loss": 4.8264,
      "mean_token_accuracy": 0.2098379760980606,
      "num_tokens": 31663154.0,
      "step": 17210
    },
    {
      "entropy": 5.211493492126465,
      "epoch": 1.4379385232208486,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004796055448721215,
      "loss": 4.8853,
      "mean_token_accuracy": 0.20910698622465135,
      "num_tokens": 31672433.0,
      "step": 17215
    },
    {
      "entropy": 5.135979318618775,
      "epoch": 1.4383561643835616,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047959315833871546,
      "loss": 4.9231,
      "mean_token_accuracy": 0.21228777319192887,
      "num_tokens": 31680794.0,
      "step": 17220
    },
    {
      "entropy": 5.132727384567261,
      "epoch": 1.4387738055462747,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047958076822365994,
      "loss": 5.0063,
      "mean_token_accuracy": 0.19754363000392913,
      "num_tokens": 31691474.0,
      "step": 17225
    },
    {
      "entropy": 5.176097345352173,
      "epoch": 1.4391914467089877,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047956837452717185,
      "loss": 4.9132,
      "mean_token_accuracy": 0.20801828503608705,
      "num_tokens": 31700821.0,
      "step": 17230
    },
    {
      "entropy": 5.16516900062561,
      "epoch": 1.4396090878717005,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047955597724946803,
      "loss": 4.9306,
      "mean_token_accuracy": 0.2138318508863449,
      "num_tokens": 31709274.0,
      "step": 17235
    },
    {
      "entropy": 5.119444179534912,
      "epoch": 1.4400267290344138,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004795435763907656,
      "loss": 4.957,
      "mean_token_accuracy": 0.2086333930492401,
      "num_tokens": 31718870.0,
      "step": 17240
    },
    {
      "entropy": 5.153617525100708,
      "epoch": 1.4404443701971266,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004795311719512817,
      "loss": 4.9506,
      "mean_token_accuracy": 0.20315176099538804,
      "num_tokens": 31728271.0,
      "step": 17245
    },
    {
      "entropy": 5.15676794052124,
      "epoch": 1.4408620113598396,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047951876393123344,
      "loss": 4.935,
      "mean_token_accuracy": 0.20354150384664535,
      "num_tokens": 31736765.0,
      "step": 17250
    },
    {
      "entropy": 5.140319585800171,
      "epoch": 1.4412796525225526,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047950635233083805,
      "loss": 4.8446,
      "mean_token_accuracy": 0.21795834302902223,
      "num_tokens": 31745529.0,
      "step": 17255
    },
    {
      "entropy": 5.142514991760254,
      "epoch": 1.4416972936852657,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047949393715031275,
      "loss": 4.968,
      "mean_token_accuracy": 0.20559622198343278,
      "num_tokens": 31755319.0,
      "step": 17260
    },
    {
      "entropy": 5.12270975112915,
      "epoch": 1.4421149348479787,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047948151838987505,
      "loss": 4.8801,
      "mean_token_accuracy": 0.20759458243846893,
      "num_tokens": 31764583.0,
      "step": 17265
    },
    {
      "entropy": 5.22823805809021,
      "epoch": 1.4425325760106915,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047946909604974205,
      "loss": 4.9042,
      "mean_token_accuracy": 0.21196184903383256,
      "num_tokens": 31774211.0,
      "step": 17270
    },
    {
      "entropy": 5.188108158111572,
      "epoch": 1.4429502171734045,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004794566701301315,
      "loss": 5.008,
      "mean_token_accuracy": 0.1954091817140579,
      "num_tokens": 31783616.0,
      "step": 17275
    },
    {
      "entropy": 5.107557153701782,
      "epoch": 1.4433678583361176,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047944424063126083,
      "loss": 4.9145,
      "mean_token_accuracy": 0.20527153015136718,
      "num_tokens": 31791972.0,
      "step": 17280
    },
    {
      "entropy": 5.107687044143677,
      "epoch": 1.4437854994988306,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004794318075533477,
      "loss": 4.9397,
      "mean_token_accuracy": 0.2001738131046295,
      "num_tokens": 31802321.0,
      "step": 17285
    },
    {
      "entropy": 5.0809088230133055,
      "epoch": 1.4442031406615436,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004794193708966096,
      "loss": 4.8091,
      "mean_token_accuracy": 0.21096540093421937,
      "num_tokens": 31811238.0,
      "step": 17290
    },
    {
      "entropy": 5.158375883102417,
      "epoch": 1.4446207818242565,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047940693066126445,
      "loss": 4.9019,
      "mean_token_accuracy": 0.20862740278244019,
      "num_tokens": 31820253.0,
      "step": 17295
    },
    {
      "entropy": 5.128254032135009,
      "epoch": 1.4450384229869697,
      "grad_norm": 1.375,
      "learning_rate": 0.0004793944868475299,
      "loss": 4.8554,
      "mean_token_accuracy": 0.21284697353839874,
      "num_tokens": 31828488.0,
      "step": 17300
    },
    {
      "entropy": 5.022374439239502,
      "epoch": 1.4454560641496825,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004793820394556238,
      "loss": 4.7705,
      "mean_token_accuracy": 0.22034941762685775,
      "num_tokens": 31836589.0,
      "step": 17305
    },
    {
      "entropy": 5.149864721298218,
      "epoch": 1.4458737053123956,
      "grad_norm": 1.25,
      "learning_rate": 0.0004793695884857641,
      "loss": 4.9509,
      "mean_token_accuracy": 0.21311169117689133,
      "num_tokens": 31845655.0,
      "step": 17310
    },
    {
      "entropy": 5.149763584136963,
      "epoch": 1.4462913464751086,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004793571339381687,
      "loss": 5.0461,
      "mean_token_accuracy": 0.19792868942022324,
      "num_tokens": 31854250.0,
      "step": 17315
    },
    {
      "entropy": 5.163375473022461,
      "epoch": 1.4467089876378216,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004793446758130558,
      "loss": 4.9058,
      "mean_token_accuracy": 0.2094845324754715,
      "num_tokens": 31863712.0,
      "step": 17320
    },
    {
      "entropy": 5.172806310653686,
      "epoch": 1.4471266288005347,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047933221411064334,
      "loss": 4.9234,
      "mean_token_accuracy": 0.20802261382341386,
      "num_tokens": 31872688.0,
      "step": 17325
    },
    {
      "entropy": 5.19673490524292,
      "epoch": 1.4475442699632475,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004793197488311494,
      "loss": 4.9607,
      "mean_token_accuracy": 0.2064736306667328,
      "num_tokens": 31882710.0,
      "step": 17330
    },
    {
      "entropy": 5.130542230606079,
      "epoch": 1.4479619111259605,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047930727997479253,
      "loss": 4.8662,
      "mean_token_accuracy": 0.20936956256628036,
      "num_tokens": 31891792.0,
      "step": 17335
    },
    {
      "entropy": 5.054149055480957,
      "epoch": 1.4483795522886735,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004792948075417906,
      "loss": 4.8761,
      "mean_token_accuracy": 0.2123418465256691,
      "num_tokens": 31900810.0,
      "step": 17340
    },
    {
      "entropy": 5.089960718154908,
      "epoch": 1.4487971934513866,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004792823315323622,
      "loss": 4.8443,
      "mean_token_accuracy": 0.2070659503340721,
      "num_tokens": 31909894.0,
      "step": 17345
    },
    {
      "entropy": 5.165180206298828,
      "epoch": 1.4492148346140996,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004792698519467258,
      "loss": 4.8969,
      "mean_token_accuracy": 0.19899687618017198,
      "num_tokens": 31918844.0,
      "step": 17350
    },
    {
      "entropy": 5.130238771438599,
      "epoch": 1.4496324757768126,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004792573687850997,
      "loss": 4.9404,
      "mean_token_accuracy": 0.20382728129625322,
      "num_tokens": 31927807.0,
      "step": 17355
    },
    {
      "entropy": 5.261888027191162,
      "epoch": 1.4500501169395257,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004792448820477025,
      "loss": 5.0509,
      "mean_token_accuracy": 0.19159342348575592,
      "num_tokens": 31936067.0,
      "step": 17360
    },
    {
      "entropy": 5.119286155700683,
      "epoch": 1.4504677581022385,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004792323917347528,
      "loss": 4.8612,
      "mean_token_accuracy": 0.21422169655561446,
      "num_tokens": 31944650.0,
      "step": 17365
    },
    {
      "entropy": 5.166160202026367,
      "epoch": 1.4508853992649515,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004792198978464692,
      "loss": 4.9469,
      "mean_token_accuracy": 0.20066129863262178,
      "num_tokens": 31955100.0,
      "step": 17370
    },
    {
      "entropy": 5.258881330490112,
      "epoch": 1.4513030404276646,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004792074003830705,
      "loss": 5.0666,
      "mean_token_accuracy": 0.20106458216905593,
      "num_tokens": 31965096.0,
      "step": 17375
    },
    {
      "entropy": 5.213639497756958,
      "epoch": 1.4517206815903776,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004791948993447754,
      "loss": 4.9477,
      "mean_token_accuracy": 0.20864515155553817,
      "num_tokens": 31974680.0,
      "step": 17380
    },
    {
      "entropy": 5.063657903671265,
      "epoch": 1.4521383227530906,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004791823947318029,
      "loss": 4.8387,
      "mean_token_accuracy": 0.2115778535604477,
      "num_tokens": 31984294.0,
      "step": 17385
    },
    {
      "entropy": 5.107378768920898,
      "epoch": 1.4525559639158034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004791698865443717,
      "loss": 4.8311,
      "mean_token_accuracy": 0.20950253903865815,
      "num_tokens": 31993331.0,
      "step": 17390
    },
    {
      "entropy": 5.21185827255249,
      "epoch": 1.4529736050785165,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047915737478270093,
      "loss": 4.9239,
      "mean_token_accuracy": 0.20461954176425934,
      "num_tokens": 32003270.0,
      "step": 17395
    },
    {
      "entropy": 5.147885465621949,
      "epoch": 1.4533912462412295,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004791448594470095,
      "loss": 4.8327,
      "mean_token_accuracy": 0.210548035800457,
      "num_tokens": 32012051.0,
      "step": 17400
    },
    {
      "entropy": 5.028901815414429,
      "epoch": 1.4538088874039425,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004791323405375167,
      "loss": 4.8809,
      "mean_token_accuracy": 0.21449045091867447,
      "num_tokens": 32021139.0,
      "step": 17405
    },
    {
      "entropy": 5.227742671966553,
      "epoch": 1.4542265285666556,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004791198180544415,
      "loss": 4.9494,
      "mean_token_accuracy": 0.20439751595258712,
      "num_tokens": 32030439.0,
      "step": 17410
    },
    {
      "entropy": 5.1754539012908936,
      "epoch": 1.4546441697293686,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004791072919980032,
      "loss": 4.9075,
      "mean_token_accuracy": 0.21416800022125243,
      "num_tokens": 32039917.0,
      "step": 17415
    },
    {
      "entropy": 5.07571234703064,
      "epoch": 1.4550618108920816,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047909476236842105,
      "loss": 4.7865,
      "mean_token_accuracy": 0.221162748336792,
      "num_tokens": 32048143.0,
      "step": 17420
    },
    {
      "entropy": 5.125170516967773,
      "epoch": 1.4554794520547945,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004790822291659144,
      "loss": 4.9625,
      "mean_token_accuracy": 0.19899593591690062,
      "num_tokens": 32058317.0,
      "step": 17425
    },
    {
      "entropy": 5.19636287689209,
      "epoch": 1.4558970932175075,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047906969239070276,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2158623993396759,
      "num_tokens": 32067298.0,
      "step": 17430
    },
    {
      "entropy": 5.197236680984497,
      "epoch": 1.4563147343802205,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047905715204300543,
      "loss": 4.9043,
      "mean_token_accuracy": 0.20144923627376557,
      "num_tokens": 32077540.0,
      "step": 17435
    },
    {
      "entropy": 5.211188936233521,
      "epoch": 1.4567323755429336,
      "grad_norm": 1.34375,
      "learning_rate": 0.000479044608123042,
      "loss": 5.0019,
      "mean_token_accuracy": 0.20024482160806656,
      "num_tokens": 32087524.0,
      "step": 17440
    },
    {
      "entropy": 5.120899868011475,
      "epoch": 1.4571500167056466,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004790320606310322,
      "loss": 4.8744,
      "mean_token_accuracy": 0.20899944603443146,
      "num_tokens": 32097084.0,
      "step": 17445
    },
    {
      "entropy": 5.153277397155762,
      "epoch": 1.4575676578683594,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004790195095671956,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21080857366323472,
      "num_tokens": 32107120.0,
      "step": 17450
    },
    {
      "entropy": 5.110561561584473,
      "epoch": 1.4579852990310724,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004790069549317518,
      "loss": 4.8186,
      "mean_token_accuracy": 0.21690795570611954,
      "num_tokens": 32116783.0,
      "step": 17455
    },
    {
      "entropy": 5.045911169052124,
      "epoch": 1.4584029401937855,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004789943967249207,
      "loss": 4.8491,
      "mean_token_accuracy": 0.20887876451015472,
      "num_tokens": 32126467.0,
      "step": 17460
    },
    {
      "entropy": 5.156898069381714,
      "epoch": 1.4588205813564985,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004789818349469222,
      "loss": 4.9418,
      "mean_token_accuracy": 0.20160799026489257,
      "num_tokens": 32136074.0,
      "step": 17465
    },
    {
      "entropy": 5.148965787887573,
      "epoch": 1.4592382225192115,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047896926959797615,
      "loss": 4.9165,
      "mean_token_accuracy": 0.21252138912677765,
      "num_tokens": 32145216.0,
      "step": 17470
    },
    {
      "entropy": 5.079937219619751,
      "epoch": 1.4596558636819246,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004789567006783025,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20398555397987367,
      "num_tokens": 32154189.0,
      "step": 17475
    },
    {
      "entropy": 5.253566789627075,
      "epoch": 1.4600735048446376,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004789441281881213,
      "loss": 5.0728,
      "mean_token_accuracy": 0.19221516102552413,
      "num_tokens": 32164437.0,
      "step": 17480
    },
    {
      "entropy": 5.199404096603393,
      "epoch": 1.4604911460073504,
      "grad_norm": 1.25,
      "learning_rate": 0.0004789315521276526,
      "loss": 4.8907,
      "mean_token_accuracy": 0.20904563814401628,
      "num_tokens": 32174147.0,
      "step": 17485
    },
    {
      "entropy": 5.133506155014038,
      "epoch": 1.4609087871700634,
      "grad_norm": 1.25,
      "learning_rate": 0.0004789189724971167,
      "loss": 4.8611,
      "mean_token_accuracy": 0.21151877343654632,
      "num_tokens": 32183003.0,
      "step": 17490
    },
    {
      "entropy": 5.094928741455078,
      "epoch": 1.4613264283327765,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047890638929673366,
      "loss": 4.9483,
      "mean_token_accuracy": 0.19807758033275605,
      "num_tokens": 32193427.0,
      "step": 17495
    },
    {
      "entropy": 5.085470724105835,
      "epoch": 1.4617440694954895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004788938025267238,
      "loss": 4.863,
      "mean_token_accuracy": 0.21332986652851105,
      "num_tokens": 32202426.0,
      "step": 17500
    },
    {
      "entropy": 5.110441112518311,
      "epoch": 1.4621617106582026,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047888121218730756,
      "loss": 4.8672,
      "mean_token_accuracy": 0.21296552568674088,
      "num_tokens": 32211424.0,
      "step": 17505
    },
    {
      "entropy": 5.076054334640503,
      "epoch": 1.4625793518209154,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047886861827870533,
      "loss": 4.8125,
      "mean_token_accuracy": 0.2147373765707016,
      "num_tokens": 32220329.0,
      "step": 17510
    },
    {
      "entropy": 5.17243275642395,
      "epoch": 1.4629969929836284,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004788560208011375,
      "loss": 4.953,
      "mean_token_accuracy": 0.20892332643270492,
      "num_tokens": 32229968.0,
      "step": 17515
    },
    {
      "entropy": 5.206861591339111,
      "epoch": 1.4634146341463414,
      "grad_norm": 1.25,
      "learning_rate": 0.0004788434197548246,
      "loss": 4.9781,
      "mean_token_accuracy": 0.2148119643330574,
      "num_tokens": 32238451.0,
      "step": 17520
    },
    {
      "entropy": 5.178503942489624,
      "epoch": 1.4638322753090545,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004788308151399873,
      "loss": 4.8939,
      "mean_token_accuracy": 0.206138476729393,
      "num_tokens": 32247368.0,
      "step": 17525
    },
    {
      "entropy": 5.084959077835083,
      "epoch": 1.4642499164717675,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047881820695684624,
      "loss": 4.8812,
      "mean_token_accuracy": 0.21149296909570695,
      "num_tokens": 32256021.0,
      "step": 17530
    },
    {
      "entropy": 5.176804399490356,
      "epoch": 1.4646675576344805,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004788055952056221,
      "loss": 5.0058,
      "mean_token_accuracy": 0.19891735911369324,
      "num_tokens": 32265877.0,
      "step": 17535
    },
    {
      "entropy": 5.184988021850586,
      "epoch": 1.4650851987971936,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047879297988653576,
      "loss": 4.889,
      "mean_token_accuracy": 0.21129671037197112,
      "num_tokens": 32274774.0,
      "step": 17540
    },
    {
      "entropy": 5.160930442810058,
      "epoch": 1.4655028399599064,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047878036099980795,
      "loss": 4.9575,
      "mean_token_accuracy": 0.2069009691476822,
      "num_tokens": 32283460.0,
      "step": 17545
    },
    {
      "entropy": 5.084412431716919,
      "epoch": 1.4659204811226194,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047876773854565964,
      "loss": 4.8662,
      "mean_token_accuracy": 0.20677349865436553,
      "num_tokens": 32292235.0,
      "step": 17550
    },
    {
      "entropy": 5.152161693572998,
      "epoch": 1.4663381222853324,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004787551125243118,
      "loss": 4.9374,
      "mean_token_accuracy": 0.20822787433862686,
      "num_tokens": 32301058.0,
      "step": 17555
    },
    {
      "entropy": 5.143720531463623,
      "epoch": 1.4667557634480455,
      "grad_norm": 1.25,
      "learning_rate": 0.0004787424829359854,
      "loss": 4.9238,
      "mean_token_accuracy": 0.20691325068473815,
      "num_tokens": 32311130.0,
      "step": 17560
    },
    {
      "entropy": 5.149203634262085,
      "epoch": 1.4671734046107585,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004787298497809017,
      "loss": 4.9176,
      "mean_token_accuracy": 0.20701824128627777,
      "num_tokens": 32320576.0,
      "step": 17565
    },
    {
      "entropy": 5.165751791000366,
      "epoch": 1.4675910457734713,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004787172130592817,
      "loss": 4.9472,
      "mean_token_accuracy": 0.2093962699174881,
      "num_tokens": 32329847.0,
      "step": 17570
    },
    {
      "entropy": 5.170899486541748,
      "epoch": 1.4680086869361844,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047870457277134667,
      "loss": 4.8774,
      "mean_token_accuracy": 0.20515911877155305,
      "num_tokens": 32338691.0,
      "step": 17575
    },
    {
      "entropy": 5.155981206893921,
      "epoch": 1.4684263280988974,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004786919289173179,
      "loss": 4.8686,
      "mean_token_accuracy": 0.20683289617300032,
      "num_tokens": 32348765.0,
      "step": 17580
    },
    {
      "entropy": 5.159283542633057,
      "epoch": 1.4688439692616104,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047867928149741666,
      "loss": 4.9278,
      "mean_token_accuracy": 0.20389823317527772,
      "num_tokens": 32357952.0,
      "step": 17585
    },
    {
      "entropy": 5.182269763946533,
      "epoch": 1.4692616104243235,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047866663051186444,
      "loss": 4.8477,
      "mean_token_accuracy": 0.20957975834608078,
      "num_tokens": 32366511.0,
      "step": 17590
    },
    {
      "entropy": 5.149849128723145,
      "epoch": 1.4696792515870365,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004786539759608828,
      "loss": 4.8986,
      "mean_token_accuracy": 0.21521106660366057,
      "num_tokens": 32375613.0,
      "step": 17595
    },
    {
      "entropy": 5.117518424987793,
      "epoch": 1.4700968927497495,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047864131784469314,
      "loss": 4.8359,
      "mean_token_accuracy": 0.21686352789402008,
      "num_tokens": 32383818.0,
      "step": 17600
    },
    {
      "entropy": 5.08196873664856,
      "epoch": 1.4705145339124623,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000478628656163517,
      "loss": 4.8058,
      "mean_token_accuracy": 0.2233527272939682,
      "num_tokens": 32391747.0,
      "step": 17605
    },
    {
      "entropy": 5.128751945495606,
      "epoch": 1.4709321750751754,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004786159909175762,
      "loss": 4.8967,
      "mean_token_accuracy": 0.21383422166109084,
      "num_tokens": 32401421.0,
      "step": 17610
    },
    {
      "entropy": 5.136932706832885,
      "epoch": 1.4713498162378884,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004786033221070924,
      "loss": 4.8314,
      "mean_token_accuracy": 0.21082740277051926,
      "num_tokens": 32410317.0,
      "step": 17615
    },
    {
      "entropy": 5.102098178863526,
      "epoch": 1.4717674574006014,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004785906497322873,
      "loss": 4.8639,
      "mean_token_accuracy": 0.21153685301542283,
      "num_tokens": 32419269.0,
      "step": 17620
    },
    {
      "entropy": 5.099305772781372,
      "epoch": 1.4721850985633145,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047857797379338284,
      "loss": 4.8888,
      "mean_token_accuracy": 0.2070673942565918,
      "num_tokens": 32427691.0,
      "step": 17625
    },
    {
      "entropy": 5.297449588775635,
      "epoch": 1.4726027397260273,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047856529429060097,
      "loss": 5.1027,
      "mean_token_accuracy": 0.19474996924400328,
      "num_tokens": 32437011.0,
      "step": 17630
    },
    {
      "entropy": 5.21338586807251,
      "epoch": 1.4730203808887403,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047855261122416355,
      "loss": 4.8951,
      "mean_token_accuracy": 0.21687048822641372,
      "num_tokens": 32445000.0,
      "step": 17635
    },
    {
      "entropy": 5.052731990814209,
      "epoch": 1.4734380220514534,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004785399245942926,
      "loss": 4.8197,
      "mean_token_accuracy": 0.20866369754076003,
      "num_tokens": 32454991.0,
      "step": 17640
    },
    {
      "entropy": 5.15196533203125,
      "epoch": 1.4738556632141664,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004785272344012104,
      "loss": 4.9124,
      "mean_token_accuracy": 0.20905728489160538,
      "num_tokens": 32464870.0,
      "step": 17645
    },
    {
      "entropy": 5.068506908416748,
      "epoch": 1.4742733043768794,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004785145406451389,
      "loss": 4.8791,
      "mean_token_accuracy": 0.2096198245882988,
      "num_tokens": 32474497.0,
      "step": 17650
    },
    {
      "entropy": 5.086861276626587,
      "epoch": 1.4746909455395925,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004785018433263004,
      "loss": 4.8775,
      "mean_token_accuracy": 0.20945051312446594,
      "num_tokens": 32483715.0,
      "step": 17655
    },
    {
      "entropy": 5.146970319747925,
      "epoch": 1.4751085867023055,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047848914244491715,
      "loss": 4.8196,
      "mean_token_accuracy": 0.21919173300266265,
      "num_tokens": 32493212.0,
      "step": 17660
    },
    {
      "entropy": 5.137461853027344,
      "epoch": 1.4755262278650183,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004784764380012115,
      "loss": 4.8612,
      "mean_token_accuracy": 0.2076539844274521,
      "num_tokens": 32502486.0,
      "step": 17665
    },
    {
      "entropy": 5.128058528900146,
      "epoch": 1.4759438690277313,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047846372999540596,
      "loss": 4.8954,
      "mean_token_accuracy": 0.21358271539211274,
      "num_tokens": 32510866.0,
      "step": 17670
    },
    {
      "entropy": 5.226132726669311,
      "epoch": 1.4763615101904444,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047845101842772285,
      "loss": 5.0106,
      "mean_token_accuracy": 0.20021334290504456,
      "num_tokens": 32520234.0,
      "step": 17675
    },
    {
      "entropy": 5.290080308914185,
      "epoch": 1.4767791513531574,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047843830329838475,
      "loss": 5.0404,
      "mean_token_accuracy": 0.1911394253373146,
      "num_tokens": 32529775.0,
      "step": 17680
    },
    {
      "entropy": 5.0563451766967775,
      "epoch": 1.4771967925158704,
      "grad_norm": 1.375,
      "learning_rate": 0.00047842558460761437,
      "loss": 4.8305,
      "mean_token_accuracy": 0.22151874303817748,
      "num_tokens": 32538234.0,
      "step": 17685
    },
    {
      "entropy": 5.155958366394043,
      "epoch": 1.4776144336785833,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047841286235563417,
      "loss": 4.9057,
      "mean_token_accuracy": 0.20814600437879563,
      "num_tokens": 32547959.0,
      "step": 17690
    },
    {
      "entropy": 5.042892646789551,
      "epoch": 1.4780320748412963,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047840013654266707,
      "loss": 4.7381,
      "mean_token_accuracy": 0.22521034479141236,
      "num_tokens": 32556095.0,
      "step": 17695
    },
    {
      "entropy": 5.1732114315032955,
      "epoch": 1.4784497160040093,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004783874071689357,
      "loss": 4.9125,
      "mean_token_accuracy": 0.208835469186306,
      "num_tokens": 32566315.0,
      "step": 17700
    },
    {
      "entropy": 5.049576044082642,
      "epoch": 1.4788673571667224,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047837467423466284,
      "loss": 4.7547,
      "mean_token_accuracy": 0.22323331981897354,
      "num_tokens": 32574325.0,
      "step": 17705
    },
    {
      "entropy": 5.171627616882324,
      "epoch": 1.4792849983294354,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047836193774007164,
      "loss": 4.9527,
      "mean_token_accuracy": 0.209492090344429,
      "num_tokens": 32583714.0,
      "step": 17710
    },
    {
      "entropy": 5.098697996139526,
      "epoch": 1.4797026394921484,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004783491976853849,
      "loss": 4.824,
      "mean_token_accuracy": 0.21209313422441484,
      "num_tokens": 32592060.0,
      "step": 17715
    },
    {
      "entropy": 5.240378332138062,
      "epoch": 1.4801202806548615,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047833645407082564,
      "loss": 4.9995,
      "mean_token_accuracy": 0.19604078233242034,
      "num_tokens": 32602349.0,
      "step": 17720
    },
    {
      "entropy": 5.10972490310669,
      "epoch": 1.4805379218175743,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004783237068966171,
      "loss": 4.8636,
      "mean_token_accuracy": 0.21666874289512633,
      "num_tokens": 32610417.0,
      "step": 17725
    },
    {
      "entropy": 5.1866868019104,
      "epoch": 1.4809555629802873,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047831095616298225,
      "loss": 4.9626,
      "mean_token_accuracy": 0.2072961449623108,
      "num_tokens": 32620829.0,
      "step": 17730
    },
    {
      "entropy": 5.0983504295349125,
      "epoch": 1.4813732041430003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004782982018701444,
      "loss": 4.801,
      "mean_token_accuracy": 0.21860390454530715,
      "num_tokens": 32629764.0,
      "step": 17735
    },
    {
      "entropy": 5.143632555007935,
      "epoch": 1.4817908453057134,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004782854440183268,
      "loss": 4.9797,
      "mean_token_accuracy": 0.2135188952088356,
      "num_tokens": 32639381.0,
      "step": 17740
    },
    {
      "entropy": 5.132673454284668,
      "epoch": 1.4822084864684264,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004782726826077528,
      "loss": 4.8924,
      "mean_token_accuracy": 0.2070842742919922,
      "num_tokens": 32649334.0,
      "step": 17745
    },
    {
      "entropy": 5.117616033554077,
      "epoch": 1.4826261276311392,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047825991763864587,
      "loss": 4.9218,
      "mean_token_accuracy": 0.21280403435230255,
      "num_tokens": 32657879.0,
      "step": 17750
    },
    {
      "entropy": 5.039640283584594,
      "epoch": 1.4830437687938522,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047824714911122934,
      "loss": 4.7712,
      "mean_token_accuracy": 0.21460940539836884,
      "num_tokens": 32666268.0,
      "step": 17755
    },
    {
      "entropy": 5.145097541809082,
      "epoch": 1.4834614099565653,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047823437702572684,
      "loss": 4.9088,
      "mean_token_accuracy": 0.2033887654542923,
      "num_tokens": 32675162.0,
      "step": 17760
    },
    {
      "entropy": 5.148525905609131,
      "epoch": 1.4838790511192783,
      "grad_norm": 1.296875,
      "learning_rate": 0.000478221601382362,
      "loss": 4.8613,
      "mean_token_accuracy": 0.21012044847011566,
      "num_tokens": 32683876.0,
      "step": 17765
    },
    {
      "entropy": 5.120468997955323,
      "epoch": 1.4842966922819913,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047820882218135835,
      "loss": 4.8555,
      "mean_token_accuracy": 0.20736627727746965,
      "num_tokens": 32693014.0,
      "step": 17770
    },
    {
      "entropy": 5.154133558273315,
      "epoch": 1.4847143334447044,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047819603942293964,
      "loss": 4.8881,
      "mean_token_accuracy": 0.20710244178771972,
      "num_tokens": 32702611.0,
      "step": 17775
    },
    {
      "entropy": 5.058179187774658,
      "epoch": 1.4851319746074174,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004781832531073297,
      "loss": 4.8317,
      "mean_token_accuracy": 0.2136715054512024,
      "num_tokens": 32712273.0,
      "step": 17780
    },
    {
      "entropy": 5.120554399490357,
      "epoch": 1.4855496157701302,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047817046323475246,
      "loss": 4.8771,
      "mean_token_accuracy": 0.19862136989831924,
      "num_tokens": 32722100.0,
      "step": 17785
    },
    {
      "entropy": 5.18491530418396,
      "epoch": 1.4859672569328433,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004781576698054316,
      "loss": 4.8953,
      "mean_token_accuracy": 0.21126817166805267,
      "num_tokens": 32730833.0,
      "step": 17790
    },
    {
      "entropy": 5.194043970108032,
      "epoch": 1.4863848980955563,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004781448728195912,
      "loss": 4.9263,
      "mean_token_accuracy": 0.20318205505609513,
      "num_tokens": 32740671.0,
      "step": 17795
    },
    {
      "entropy": 5.192591524124145,
      "epoch": 1.4868025392582693,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047813207227745527,
      "loss": 4.863,
      "mean_token_accuracy": 0.20503560900688172,
      "num_tokens": 32749992.0,
      "step": 17800
    },
    {
      "entropy": 5.089015817642212,
      "epoch": 1.4872201804209824,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047811926817924785,
      "loss": 4.8731,
      "mean_token_accuracy": 0.209914530813694,
      "num_tokens": 32758591.0,
      "step": 17805
    },
    {
      "entropy": 5.150824642181396,
      "epoch": 1.4876378215836952,
      "grad_norm": 1.25,
      "learning_rate": 0.00047810646052519324,
      "loss": 4.9695,
      "mean_token_accuracy": 0.2089395508170128,
      "num_tokens": 32767338.0,
      "step": 17810
    },
    {
      "entropy": 5.185086488723755,
      "epoch": 1.4880554627464082,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004780936493155155,
      "loss": 4.9579,
      "mean_token_accuracy": 0.20173976719379424,
      "num_tokens": 32777093.0,
      "step": 17815
    },
    {
      "entropy": 5.201314496994018,
      "epoch": 1.4884731039091212,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047808083455043893,
      "loss": 4.9555,
      "mean_token_accuracy": 0.2024687871336937,
      "num_tokens": 32786586.0,
      "step": 17820
    },
    {
      "entropy": 5.15090970993042,
      "epoch": 1.4888907450718343,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047806801623018804,
      "loss": 5.0065,
      "mean_token_accuracy": 0.20247256010770798,
      "num_tokens": 32795166.0,
      "step": 17825
    },
    {
      "entropy": 5.14714298248291,
      "epoch": 1.4893083862345473,
      "grad_norm": 1.25,
      "learning_rate": 0.000478055194354987,
      "loss": 4.8889,
      "mean_token_accuracy": 0.20583096742630005,
      "num_tokens": 32803608.0,
      "step": 17830
    },
    {
      "entropy": 5.195831823348999,
      "epoch": 1.4897260273972603,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047804236892506034,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2096435621380806,
      "num_tokens": 32811812.0,
      "step": 17835
    },
    {
      "entropy": 5.090518283843994,
      "epoch": 1.4901436685599734,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047802953994063266,
      "loss": 4.9194,
      "mean_token_accuracy": 0.21239205151796342,
      "num_tokens": 32821235.0,
      "step": 17840
    },
    {
      "entropy": 5.1778381824493405,
      "epoch": 1.4905613097226862,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047801670740192843,
      "loss": 4.8687,
      "mean_token_accuracy": 0.21184130609035492,
      "num_tokens": 32830947.0,
      "step": 17845
    },
    {
      "entropy": 5.2437234878540036,
      "epoch": 1.4909789508853992,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047800387130917245,
      "loss": 4.9822,
      "mean_token_accuracy": 0.20337675213813783,
      "num_tokens": 32840436.0,
      "step": 17850
    },
    {
      "entropy": 5.226590490341186,
      "epoch": 1.4913965920481123,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047799103166258926,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21140623092651367,
      "num_tokens": 32849360.0,
      "step": 17855
    },
    {
      "entropy": 5.103819417953491,
      "epoch": 1.4918142332108253,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047797818846240383,
      "loss": 4.8754,
      "mean_token_accuracy": 0.21068423092365265,
      "num_tokens": 32857535.0,
      "step": 17860
    },
    {
      "entropy": 5.035498905181885,
      "epoch": 1.4922318743735383,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004779653417088408,
      "loss": 4.7802,
      "mean_token_accuracy": 0.21418958753347397,
      "num_tokens": 32867650.0,
      "step": 17865
    },
    {
      "entropy": 5.073046112060547,
      "epoch": 1.4926495155362511,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047795249140212523,
      "loss": 4.8726,
      "mean_token_accuracy": 0.21388939619064332,
      "num_tokens": 32876431.0,
      "step": 17870
    },
    {
      "entropy": 5.217727613449097,
      "epoch": 1.4930671566989642,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047793963754248195,
      "loss": 4.9513,
      "mean_token_accuracy": 0.20150362700223923,
      "num_tokens": 32886095.0,
      "step": 17875
    },
    {
      "entropy": 5.215565633773804,
      "epoch": 1.4934847978616772,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000477926780130136,
      "loss": 4.9706,
      "mean_token_accuracy": 0.20836709588766097,
      "num_tokens": 32895507.0,
      "step": 17880
    },
    {
      "entropy": 5.061017942428589,
      "epoch": 1.4939024390243902,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047791391916531255,
      "loss": 4.8026,
      "mean_token_accuracy": 0.21098943948745727,
      "num_tokens": 32904920.0,
      "step": 17885
    },
    {
      "entropy": 5.112650966644287,
      "epoch": 1.4943200801871033,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004779010546482367,
      "loss": 4.9355,
      "mean_token_accuracy": 0.21356968730688095,
      "num_tokens": 32914059.0,
      "step": 17890
    },
    {
      "entropy": 5.195305585861206,
      "epoch": 1.4947377213498163,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004778881865791337,
      "loss": 4.9404,
      "mean_token_accuracy": 0.20442514270544052,
      "num_tokens": 32923490.0,
      "step": 17895
    },
    {
      "entropy": 5.216765642166138,
      "epoch": 1.4951553625125293,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047787531495822866,
      "loss": 4.9664,
      "mean_token_accuracy": 0.19608409702777863,
      "num_tokens": 32932653.0,
      "step": 17900
    },
    {
      "entropy": 5.182994174957275,
      "epoch": 1.4955730036752422,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047786243978574706,
      "loss": 4.9158,
      "mean_token_accuracy": 0.20714071542024612,
      "num_tokens": 32942000.0,
      "step": 17905
    },
    {
      "entropy": 5.167428779602051,
      "epoch": 1.4959906448379552,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047784956106191435,
      "loss": 4.8649,
      "mean_token_accuracy": 0.20937742441892623,
      "num_tokens": 32951762.0,
      "step": 17910
    },
    {
      "entropy": 5.184979677200317,
      "epoch": 1.4964082860006682,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004778366787869559,
      "loss": 4.983,
      "mean_token_accuracy": 0.20604320615530014,
      "num_tokens": 32961406.0,
      "step": 17915
    },
    {
      "entropy": 5.131099033355713,
      "epoch": 1.4968259271633813,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004778237929610971,
      "loss": 4.9143,
      "mean_token_accuracy": 0.20471430867910384,
      "num_tokens": 32970086.0,
      "step": 17920
    },
    {
      "entropy": 5.121146154403687,
      "epoch": 1.4972435683260943,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004778109035845638,
      "loss": 4.9224,
      "mean_token_accuracy": 0.2046671152114868,
      "num_tokens": 32979576.0,
      "step": 17925
    },
    {
      "entropy": 5.220463132858276,
      "epoch": 1.497661209488807,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047779801065758146,
      "loss": 4.8699,
      "mean_token_accuracy": 0.21901459395885467,
      "num_tokens": 32987906.0,
      "step": 17930
    },
    {
      "entropy": 5.287953186035156,
      "epoch": 1.4980788506515201,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004777851141803758,
      "loss": 4.965,
      "mean_token_accuracy": 0.20420041382312776,
      "num_tokens": 32996983.0,
      "step": 17935
    },
    {
      "entropy": 5.243406629562378,
      "epoch": 1.4984964918142332,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004777722141531727,
      "loss": 5.034,
      "mean_token_accuracy": 0.1992589220404625,
      "num_tokens": 33006134.0,
      "step": 17940
    },
    {
      "entropy": 5.087921905517578,
      "epoch": 1.4989141329769462,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047775931057619786,
      "loss": 4.8171,
      "mean_token_accuracy": 0.21399959325790405,
      "num_tokens": 33014987.0,
      "step": 17945
    },
    {
      "entropy": 5.185365104675293,
      "epoch": 1.4993317741396592,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004777464034496773,
      "loss": 4.925,
      "mean_token_accuracy": 0.206088587641716,
      "num_tokens": 33025347.0,
      "step": 17950
    },
    {
      "entropy": 5.139272308349609,
      "epoch": 1.4997494153023723,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004777334927738368,
      "loss": 4.8508,
      "mean_token_accuracy": 0.20478181689977645,
      "num_tokens": 33035033.0,
      "step": 17955
    },
    {
      "entropy": 5.1712305545806885,
      "epoch": 1.5001670564650853,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047772057854890247,
      "loss": 4.9628,
      "mean_token_accuracy": 0.2077002078294754,
      "num_tokens": 33045135.0,
      "step": 17960
    },
    {
      "entropy": 5.200890827178955,
      "epoch": 1.5005846976277981,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047770766077510045,
      "loss": 4.9101,
      "mean_token_accuracy": 0.20108472853899,
      "num_tokens": 33054124.0,
      "step": 17965
    },
    {
      "entropy": 5.176437139511108,
      "epoch": 1.5010023387905111,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047769473945265685,
      "loss": 4.9485,
      "mean_token_accuracy": 0.20176150351762773,
      "num_tokens": 33062735.0,
      "step": 17970
    },
    {
      "entropy": 5.1219236850738525,
      "epoch": 1.5014199799532242,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004776818145817979,
      "loss": 4.8652,
      "mean_token_accuracy": 0.21171530336141586,
      "num_tokens": 33072569.0,
      "step": 17975
    },
    {
      "entropy": 5.159666633605957,
      "epoch": 1.5018376211159372,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047766888616274973,
      "loss": 4.8769,
      "mean_token_accuracy": 0.21219520419836044,
      "num_tokens": 33081887.0,
      "step": 17980
    },
    {
      "entropy": 5.237988090515136,
      "epoch": 1.5022552622786502,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047765595419573876,
      "loss": 4.9888,
      "mean_token_accuracy": 0.205902399122715,
      "num_tokens": 33091016.0,
      "step": 17985
    },
    {
      "entropy": 5.159904193878174,
      "epoch": 1.502672903441363,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047764301868099136,
      "loss": 4.846,
      "mean_token_accuracy": 0.21081001609563826,
      "num_tokens": 33100318.0,
      "step": 17990
    },
    {
      "entropy": 5.18365969657898,
      "epoch": 1.5030905446040763,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047763007961873407,
      "loss": 5.0306,
      "mean_token_accuracy": 0.2022309795022011,
      "num_tokens": 33108950.0,
      "step": 17995
    },
    {
      "entropy": 5.05850772857666,
      "epoch": 1.5035081857667891,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004776171370091933,
      "loss": 4.8436,
      "mean_token_accuracy": 0.2149359792470932,
      "num_tokens": 33117983.0,
      "step": 18000
    },
    {
      "epoch": 1.5035081857667891,
      "eval_entropy": 4.876133741046685,
      "eval_loss": 4.921685695648193,
      "eval_mean_token_accuracy": 0.21852036524193946,
      "eval_num_tokens": 33117983.0,
      "eval_runtime": 27.712,
      "eval_samples_per_second": 1362.694,
      "eval_steps_per_second": 170.359,
      "step": 18000
    },
    {
      "entropy": 5.123652601242066,
      "epoch": 1.5039258269295022,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004776041908525956,
      "loss": 4.8726,
      "mean_token_accuracy": 0.20989387333393097,
      "num_tokens": 33127725.0,
      "step": 18005
    },
    {
      "entropy": 5.0876648902893065,
      "epoch": 1.5043434680922152,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004775912411491677,
      "loss": 4.8484,
      "mean_token_accuracy": 0.21491675227880477,
      "num_tokens": 33136921.0,
      "step": 18010
    },
    {
      "entropy": 5.173797225952148,
      "epoch": 1.504761109254928,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004775782878991363,
      "loss": 4.9271,
      "mean_token_accuracy": 0.21060003191232682,
      "num_tokens": 33146242.0,
      "step": 18015
    },
    {
      "entropy": 5.141316223144531,
      "epoch": 1.5051787504176413,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047756533110272813,
      "loss": 4.842,
      "mean_token_accuracy": 0.2129431739449501,
      "num_tokens": 33154555.0,
      "step": 18020
    },
    {
      "entropy": 5.165144348144532,
      "epoch": 1.505596391580354,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047755237076016997,
      "loss": 4.9017,
      "mean_token_accuracy": 0.20972767174243928,
      "num_tokens": 33163832.0,
      "step": 18025
    },
    {
      "entropy": 5.210700941085816,
      "epoch": 1.506014032743067,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047753940687168876,
      "loss": 4.9827,
      "mean_token_accuracy": 0.20356750935316087,
      "num_tokens": 33173015.0,
      "step": 18030
    },
    {
      "entropy": 5.080803537368775,
      "epoch": 1.5064316739057801,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004775264394375114,
      "loss": 4.7724,
      "mean_token_accuracy": 0.22232043594121934,
      "num_tokens": 33181643.0,
      "step": 18035
    },
    {
      "entropy": 5.0786958694458,
      "epoch": 1.5068493150684932,
      "grad_norm": 1.1875,
      "learning_rate": 0.000477513468457865,
      "loss": 4.8591,
      "mean_token_accuracy": 0.21013163179159164,
      "num_tokens": 33190750.0,
      "step": 18040
    },
    {
      "entropy": 5.121452140808105,
      "epoch": 1.5072669562312062,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047750049393297656,
      "loss": 4.9417,
      "mean_token_accuracy": 0.20667119175195695,
      "num_tokens": 33200118.0,
      "step": 18045
    },
    {
      "entropy": 5.201645088195801,
      "epoch": 1.507684597393919,
      "grad_norm": 1.375,
      "learning_rate": 0.00047748751586307314,
      "loss": 4.9522,
      "mean_token_accuracy": 0.20299015194177628,
      "num_tokens": 33210157.0,
      "step": 18050
    },
    {
      "entropy": 5.257601118087768,
      "epoch": 1.5081022385566323,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047747453424838216,
      "loss": 4.9931,
      "mean_token_accuracy": 0.2005255028605461,
      "num_tokens": 33218850.0,
      "step": 18055
    },
    {
      "entropy": 5.122921419143677,
      "epoch": 1.508519879719345,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047746154908913065,
      "loss": 4.8281,
      "mean_token_accuracy": 0.21400179862976074,
      "num_tokens": 33228179.0,
      "step": 18060
    },
    {
      "entropy": 5.085597848892212,
      "epoch": 1.5089375208820581,
      "grad_norm": 1.15625,
      "learning_rate": 0.000477448560385546,
      "loss": 4.9141,
      "mean_token_accuracy": 0.2060859829187393,
      "num_tokens": 33237259.0,
      "step": 18065
    },
    {
      "entropy": 5.159873199462891,
      "epoch": 1.5093551620447712,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004774355681378557,
      "loss": 4.9702,
      "mean_token_accuracy": 0.20541775971651077,
      "num_tokens": 33246560.0,
      "step": 18070
    },
    {
      "entropy": 5.194597816467285,
      "epoch": 1.509772803207484,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047742257234628703,
      "loss": 4.942,
      "mean_token_accuracy": 0.20528552830219268,
      "num_tokens": 33257069.0,
      "step": 18075
    },
    {
      "entropy": 5.085633230209351,
      "epoch": 1.5101904443701972,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047740957301106756,
      "loss": 4.6963,
      "mean_token_accuracy": 0.22347544580698014,
      "num_tokens": 33266104.0,
      "step": 18080
    },
    {
      "entropy": 5.096164131164551,
      "epoch": 1.51060808553291,
      "grad_norm": 1.28125,
      "learning_rate": 0.000477396570132425,
      "loss": 4.9054,
      "mean_token_accuracy": 0.20995844006538392,
      "num_tokens": 33276567.0,
      "step": 18085
    },
    {
      "entropy": 5.104204559326172,
      "epoch": 1.511025726695623,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047738356371058673,
      "loss": 4.8504,
      "mean_token_accuracy": 0.21342946141958236,
      "num_tokens": 33285329.0,
      "step": 18090
    },
    {
      "entropy": 5.076615333557129,
      "epoch": 1.511443367858336,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004773705537457806,
      "loss": 4.745,
      "mean_token_accuracy": 0.22495145797729493,
      "num_tokens": 33294213.0,
      "step": 18095
    },
    {
      "entropy": 5.086140203475952,
      "epoch": 1.5118610090210491,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047735754023823433,
      "loss": 4.785,
      "mean_token_accuracy": 0.21261763274669648,
      "num_tokens": 33303816.0,
      "step": 18100
    },
    {
      "entropy": 5.258515310287476,
      "epoch": 1.5122786501837622,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004773445231881757,
      "loss": 5.0227,
      "mean_token_accuracy": 0.2011199489235878,
      "num_tokens": 33313060.0,
      "step": 18105
    },
    {
      "entropy": 5.144974279403686,
      "epoch": 1.512696291346475,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047733150259583274,
      "loss": 4.9176,
      "mean_token_accuracy": 0.21380036771297456,
      "num_tokens": 33322586.0,
      "step": 18110
    },
    {
      "entropy": 5.09441909790039,
      "epoch": 1.5131139325091882,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047731847846143325,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2121329203248024,
      "num_tokens": 33331684.0,
      "step": 18115
    },
    {
      "entropy": 5.139348125457763,
      "epoch": 1.513531573671901,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004773054507852052,
      "loss": 4.8565,
      "mean_token_accuracy": 0.21042075604200364,
      "num_tokens": 33340980.0,
      "step": 18120
    },
    {
      "entropy": 5.226842737197876,
      "epoch": 1.513949214834614,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004772924195673768,
      "loss": 5.0059,
      "mean_token_accuracy": 0.20774481445550919,
      "num_tokens": 33349486.0,
      "step": 18125
    },
    {
      "entropy": 5.066594505310059,
      "epoch": 1.5143668559973271,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047727938480817604,
      "loss": 4.7976,
      "mean_token_accuracy": 0.217968125641346,
      "num_tokens": 33359128.0,
      "step": 18130
    },
    {
      "entropy": 5.229175853729248,
      "epoch": 1.51478449716004,
      "grad_norm": 1.375,
      "learning_rate": 0.00047726634650783117,
      "loss": 4.9871,
      "mean_token_accuracy": 0.19911148846149446,
      "num_tokens": 33368059.0,
      "step": 18135
    },
    {
      "entropy": 5.157021141052246,
      "epoch": 1.5152021383227532,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004772533046665705,
      "loss": 4.8783,
      "mean_token_accuracy": 0.20979826301336288,
      "num_tokens": 33376643.0,
      "step": 18140
    },
    {
      "entropy": 5.154545545578003,
      "epoch": 1.515619779485466,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004772402592846222,
      "loss": 4.9156,
      "mean_token_accuracy": 0.20331271290779113,
      "num_tokens": 33386423.0,
      "step": 18145
    },
    {
      "entropy": 5.0831952571868895,
      "epoch": 1.516037420648179,
      "grad_norm": 1.25,
      "learning_rate": 0.00047722721036221467,
      "loss": 4.8133,
      "mean_token_accuracy": 0.21464014202356338,
      "num_tokens": 33395049.0,
      "step": 18150
    },
    {
      "entropy": 5.100665187835693,
      "epoch": 1.516455061810892,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004772141578995765,
      "loss": 4.9415,
      "mean_token_accuracy": 0.20402145236730576,
      "num_tokens": 33403572.0,
      "step": 18155
    },
    {
      "entropy": 5.1020105361938475,
      "epoch": 1.516872702973605,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047720110189693605,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21117090433835983,
      "num_tokens": 33412263.0,
      "step": 18160
    },
    {
      "entropy": 5.098607683181763,
      "epoch": 1.5172903441363181,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047718804235452193,
      "loss": 4.9055,
      "mean_token_accuracy": 0.2169795587658882,
      "num_tokens": 33421041.0,
      "step": 18165
    },
    {
      "entropy": 5.262117195129394,
      "epoch": 1.517707985299031,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004771749792725628,
      "loss": 5.0831,
      "mean_token_accuracy": 0.1988689512014389,
      "num_tokens": 33431222.0,
      "step": 18170
    },
    {
      "entropy": 5.123106575012207,
      "epoch": 1.5181256264617442,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004771619126512873,
      "loss": 4.8224,
      "mean_token_accuracy": 0.2151470512151718,
      "num_tokens": 33440600.0,
      "step": 18175
    },
    {
      "entropy": 5.119487714767456,
      "epoch": 1.518543267624457,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047714884249092413,
      "loss": 5.0092,
      "mean_token_accuracy": 0.20621704757213594,
      "num_tokens": 33449593.0,
      "step": 18180
    },
    {
      "entropy": 5.117685127258301,
      "epoch": 1.51896090878717,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004771357687917021,
      "loss": 4.8492,
      "mean_token_accuracy": 0.21077531427145005,
      "num_tokens": 33459568.0,
      "step": 18185
    },
    {
      "entropy": 5.178852128982544,
      "epoch": 1.519378549949883,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047712269155385017,
      "loss": 4.8753,
      "mean_token_accuracy": 0.21052484810352326,
      "num_tokens": 33468243.0,
      "step": 18190
    },
    {
      "entropy": 5.210101222991943,
      "epoch": 1.519796191112596,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047710961077759736,
      "loss": 4.948,
      "mean_token_accuracy": 0.19654192477464677,
      "num_tokens": 33477496.0,
      "step": 18195
    },
    {
      "entropy": 5.266223287582397,
      "epoch": 1.5202138322753092,
      "grad_norm": 1.25,
      "learning_rate": 0.00047709652646317236,
      "loss": 4.9875,
      "mean_token_accuracy": 0.19953552037477493,
      "num_tokens": 33486993.0,
      "step": 18200
    },
    {
      "entropy": 5.231377792358399,
      "epoch": 1.520631473438022,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047708343861080447,
      "loss": 4.9648,
      "mean_token_accuracy": 0.2008322671055794,
      "num_tokens": 33497258.0,
      "step": 18205
    },
    {
      "entropy": 5.171537160873413,
      "epoch": 1.521049114600735,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004770703472207227,
      "loss": 4.9619,
      "mean_token_accuracy": 0.21039622873067856,
      "num_tokens": 33506328.0,
      "step": 18210
    },
    {
      "entropy": 5.183963060379028,
      "epoch": 1.521466755763448,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004770572522931563,
      "loss": 4.9272,
      "mean_token_accuracy": 0.20919496715068817,
      "num_tokens": 33515667.0,
      "step": 18215
    },
    {
      "entropy": 5.168969011306762,
      "epoch": 1.521884396926161,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004770441538283345,
      "loss": 4.8347,
      "mean_token_accuracy": 0.21511904150247574,
      "num_tokens": 33523641.0,
      "step": 18220
    },
    {
      "entropy": 5.174999856948853,
      "epoch": 1.522302038088874,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047703105182648656,
      "loss": 4.926,
      "mean_token_accuracy": 0.20309316515922546,
      "num_tokens": 33532989.0,
      "step": 18225
    },
    {
      "entropy": 5.1665277004241945,
      "epoch": 1.522719679251587,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004770179462878419,
      "loss": 4.8607,
      "mean_token_accuracy": 0.21650907546281814,
      "num_tokens": 33541810.0,
      "step": 18230
    },
    {
      "entropy": 5.125249290466309,
      "epoch": 1.5231373204143002,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004770048372126299,
      "loss": 4.9354,
      "mean_token_accuracy": 0.20918734222650529,
      "num_tokens": 33550260.0,
      "step": 18235
    },
    {
      "entropy": 5.120838832855225,
      "epoch": 1.523554961577013,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004769917246010801,
      "loss": 4.8708,
      "mean_token_accuracy": 0.21348764896392822,
      "num_tokens": 33559595.0,
      "step": 18240
    },
    {
      "entropy": 5.17507266998291,
      "epoch": 1.523972602739726,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004769786084534219,
      "loss": 4.9269,
      "mean_token_accuracy": 0.20710486620664598,
      "num_tokens": 33568780.0,
      "step": 18245
    },
    {
      "entropy": 5.126837158203125,
      "epoch": 1.524390243902439,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047696548876988517,
      "loss": 4.8656,
      "mean_token_accuracy": 0.21037127077579498,
      "num_tokens": 33578060.0,
      "step": 18250
    },
    {
      "entropy": 5.12902045249939,
      "epoch": 1.5248078850651519,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004769523655506994,
      "loss": 4.9111,
      "mean_token_accuracy": 0.20773013830184936,
      "num_tokens": 33586866.0,
      "step": 18255
    },
    {
      "entropy": 5.145849704742432,
      "epoch": 1.5252255262278651,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004769392387960944,
      "loss": 4.8835,
      "mean_token_accuracy": 0.21614997386932372,
      "num_tokens": 33595522.0,
      "step": 18260
    },
    {
      "entropy": 5.181083106994629,
      "epoch": 1.525643167390578,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004769261085063,
      "loss": 4.9335,
      "mean_token_accuracy": 0.20714088827371596,
      "num_tokens": 33604691.0,
      "step": 18265
    },
    {
      "entropy": 5.1911506175994875,
      "epoch": 1.526060808553291,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047691297468154594,
      "loss": 4.9199,
      "mean_token_accuracy": 0.2064562886953354,
      "num_tokens": 33613401.0,
      "step": 18270
    },
    {
      "entropy": 5.1310715675354,
      "epoch": 1.526478449716004,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004768998373220622,
      "loss": 4.9084,
      "mean_token_accuracy": 0.20441832393407822,
      "num_tokens": 33623871.0,
      "step": 18275
    },
    {
      "entropy": 5.154557514190674,
      "epoch": 1.526896090878717,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004768866964280789,
      "loss": 4.906,
      "mean_token_accuracy": 0.21222960203886032,
      "num_tokens": 33634327.0,
      "step": 18280
    },
    {
      "entropy": 5.080580472946167,
      "epoch": 1.52731373204143,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047687355199982584,
      "loss": 4.8434,
      "mean_token_accuracy": 0.21411128342151642,
      "num_tokens": 33644139.0,
      "step": 18285
    },
    {
      "entropy": 5.220663452148438,
      "epoch": 1.5277313732041429,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004768604040375333,
      "loss": 5.024,
      "mean_token_accuracy": 0.1963203728199005,
      "num_tokens": 33654228.0,
      "step": 18290
    },
    {
      "entropy": 5.213138580322266,
      "epoch": 1.5281490143668561,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047684725254143137,
      "loss": 4.9506,
      "mean_token_accuracy": 0.21795656234025956,
      "num_tokens": 33663511.0,
      "step": 18295
    },
    {
      "entropy": 5.1594212532043455,
      "epoch": 1.528566655529569,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004768340975117504,
      "loss": 4.9277,
      "mean_token_accuracy": 0.207415434718132,
      "num_tokens": 33672814.0,
      "step": 18300
    },
    {
      "entropy": 5.135950374603271,
      "epoch": 1.528984296692282,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004768209389487205,
      "loss": 4.8541,
      "mean_token_accuracy": 0.2158544823527336,
      "num_tokens": 33681527.0,
      "step": 18305
    },
    {
      "entropy": 5.043429374694824,
      "epoch": 1.529401937854995,
      "grad_norm": 1.25,
      "learning_rate": 0.0004768077768525722,
      "loss": 4.8441,
      "mean_token_accuracy": 0.20596176236867905,
      "num_tokens": 33690034.0,
      "step": 18310
    },
    {
      "entropy": 5.145067262649536,
      "epoch": 1.5298195790177078,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004767946112235359,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2120111420750618,
      "num_tokens": 33698748.0,
      "step": 18315
    },
    {
      "entropy": 5.135795736312867,
      "epoch": 1.530237220180421,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000476781442061842,
      "loss": 4.8642,
      "mean_token_accuracy": 0.20505469292402267,
      "num_tokens": 33708118.0,
      "step": 18320
    },
    {
      "entropy": 5.105589771270752,
      "epoch": 1.5306548613431339,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047676826936772097,
      "loss": 4.9787,
      "mean_token_accuracy": 0.2003694012761116,
      "num_tokens": 33718221.0,
      "step": 18325
    },
    {
      "entropy": 5.1768735408782955,
      "epoch": 1.531072502505847,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047675509314140365,
      "loss": 4.883,
      "mean_token_accuracy": 0.2122856318950653,
      "num_tokens": 33727438.0,
      "step": 18330
    },
    {
      "entropy": 5.148052597045899,
      "epoch": 1.53149014366856,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004767419133831206,
      "loss": 4.9088,
      "mean_token_accuracy": 0.21316836923360824,
      "num_tokens": 33735732.0,
      "step": 18335
    },
    {
      "entropy": 5.2245728969573975,
      "epoch": 1.531907784831273,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047672873009310234,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20585545897483826,
      "num_tokens": 33745198.0,
      "step": 18340
    },
    {
      "entropy": 5.136308860778809,
      "epoch": 1.532325425993986,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047671554327157996,
      "loss": 4.7694,
      "mean_token_accuracy": 0.2244168996810913,
      "num_tokens": 33753661.0,
      "step": 18345
    },
    {
      "entropy": 5.167673206329345,
      "epoch": 1.5327430671566988,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047670235291878417,
      "loss": 4.9216,
      "mean_token_accuracy": 0.20892246514558793,
      "num_tokens": 33762467.0,
      "step": 18350
    },
    {
      "entropy": 5.101335859298706,
      "epoch": 1.533160708319412,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004766891590349459,
      "loss": 4.9085,
      "mean_token_accuracy": 0.2102018803358078,
      "num_tokens": 33771309.0,
      "step": 18355
    },
    {
      "entropy": 5.208294153213501,
      "epoch": 1.533578349482125,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004766759616202961,
      "loss": 4.9181,
      "mean_token_accuracy": 0.21423488408327102,
      "num_tokens": 33779690.0,
      "step": 18360
    },
    {
      "entropy": 5.221598291397095,
      "epoch": 1.533995990644838,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004766627606750659,
      "loss": 4.9862,
      "mean_token_accuracy": 0.20309192687273026,
      "num_tokens": 33788625.0,
      "step": 18365
    },
    {
      "entropy": 5.163074636459351,
      "epoch": 1.534413631807551,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004766495561994863,
      "loss": 4.864,
      "mean_token_accuracy": 0.20646654367446898,
      "num_tokens": 33798686.0,
      "step": 18370
    },
    {
      "entropy": 5.142886114120484,
      "epoch": 1.534831272970264,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004766363481937885,
      "loss": 4.8693,
      "mean_token_accuracy": 0.20917627215385437,
      "num_tokens": 33807948.0,
      "step": 18375
    },
    {
      "entropy": 5.100201177597046,
      "epoch": 1.535248914132977,
      "grad_norm": 1.125,
      "learning_rate": 0.00047662313665820365,
      "loss": 4.9222,
      "mean_token_accuracy": 0.21223629117012024,
      "num_tokens": 33817219.0,
      "step": 18380
    },
    {
      "entropy": 5.140187931060791,
      "epoch": 1.5356665552956899,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004766099215929632,
      "loss": 4.8299,
      "mean_token_accuracy": 0.21961036920547486,
      "num_tokens": 33825428.0,
      "step": 18385
    },
    {
      "entropy": 5.17720398902893,
      "epoch": 1.536084196458403,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047659670299829833,
      "loss": 4.8942,
      "mean_token_accuracy": 0.2095017671585083,
      "num_tokens": 33834593.0,
      "step": 18390
    },
    {
      "entropy": 5.076858949661255,
      "epoch": 1.536501837621116,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004765834808744405,
      "loss": 4.7737,
      "mean_token_accuracy": 0.21905491799116134,
      "num_tokens": 33842673.0,
      "step": 18395
    },
    {
      "entropy": 5.028508138656616,
      "epoch": 1.536919478783829,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047657025522162123,
      "loss": 4.8695,
      "mean_token_accuracy": 0.22216283679008483,
      "num_tokens": 33852387.0,
      "step": 18400
    },
    {
      "entropy": 5.167828369140625,
      "epoch": 1.537337119946542,
      "grad_norm": 1.234375,
      "learning_rate": 0.000476557026040072,
      "loss": 4.8693,
      "mean_token_accuracy": 0.21851834505796433,
      "num_tokens": 33861264.0,
      "step": 18405
    },
    {
      "entropy": 5.26661319732666,
      "epoch": 1.5377547611092548,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047654379333002446,
      "loss": 4.9932,
      "mean_token_accuracy": 0.20695339590311052,
      "num_tokens": 33870084.0,
      "step": 18410
    },
    {
      "entropy": 5.112091779708862,
      "epoch": 1.538172402271968,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047653055709171013,
      "loss": 4.792,
      "mean_token_accuracy": 0.21716798841953278,
      "num_tokens": 33878806.0,
      "step": 18415
    },
    {
      "entropy": 5.069058990478515,
      "epoch": 1.5385900434346809,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047651731732536084,
      "loss": 4.8953,
      "mean_token_accuracy": 0.2141016826033592,
      "num_tokens": 33888277.0,
      "step": 18420
    },
    {
      "entropy": 5.096891975402832,
      "epoch": 1.539007684597394,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047650407403120845,
      "loss": 4.8804,
      "mean_token_accuracy": 0.21218026131391526,
      "num_tokens": 33897523.0,
      "step": 18425
    },
    {
      "entropy": 5.156210041046142,
      "epoch": 1.539425325760107,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047649082720948455,
      "loss": 4.9416,
      "mean_token_accuracy": 0.20002980381250382,
      "num_tokens": 33907199.0,
      "step": 18430
    },
    {
      "entropy": 5.212652683258057,
      "epoch": 1.53984296692282,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047647757686042125,
      "loss": 5.0393,
      "mean_token_accuracy": 0.19944989085197448,
      "num_tokens": 33917191.0,
      "step": 18435
    },
    {
      "entropy": 5.189802598953247,
      "epoch": 1.540260608085533,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047646432298425047,
      "loss": 4.8995,
      "mean_token_accuracy": 0.21066271662712097,
      "num_tokens": 33926595.0,
      "step": 18440
    },
    {
      "entropy": 5.188390302658081,
      "epoch": 1.5406782492482458,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047645106558120417,
      "loss": 4.9847,
      "mean_token_accuracy": 0.19993089735507966,
      "num_tokens": 33936787.0,
      "step": 18445
    },
    {
      "entropy": 5.178163337707519,
      "epoch": 1.541095890410959,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004764378046515145,
      "loss": 4.9053,
      "mean_token_accuracy": 0.21187727004289628,
      "num_tokens": 33946260.0,
      "step": 18450
    },
    {
      "entropy": 5.221795225143433,
      "epoch": 1.5415135315736719,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004764245401954136,
      "loss": 4.9587,
      "mean_token_accuracy": 0.2097621515393257,
      "num_tokens": 33955346.0,
      "step": 18455
    },
    {
      "entropy": 5.17407546043396,
      "epoch": 1.541931172736385,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004764112722131337,
      "loss": 4.8434,
      "mean_token_accuracy": 0.21724287420511246,
      "num_tokens": 33964580.0,
      "step": 18460
    },
    {
      "entropy": 5.138194608688354,
      "epoch": 1.542348813899098,
      "grad_norm": 1.25,
      "learning_rate": 0.00047639800070490695,
      "loss": 4.8972,
      "mean_token_accuracy": 0.21225025802850722,
      "num_tokens": 33973465.0,
      "step": 18465
    },
    {
      "entropy": 5.152113580703736,
      "epoch": 1.5427664550618108,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004763847256709658,
      "loss": 4.8617,
      "mean_token_accuracy": 0.21148843914270402,
      "num_tokens": 33982153.0,
      "step": 18470
    },
    {
      "entropy": 5.09111852645874,
      "epoch": 1.543184096224524,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004763714471115427,
      "loss": 4.8582,
      "mean_token_accuracy": 0.21751130223274232,
      "num_tokens": 33990936.0,
      "step": 18475
    },
    {
      "entropy": 5.080809545516968,
      "epoch": 1.5436017373872368,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004763581650268699,
      "loss": 4.8217,
      "mean_token_accuracy": 0.21302342116832734,
      "num_tokens": 34000704.0,
      "step": 18480
    },
    {
      "entropy": 5.128146362304688,
      "epoch": 1.5440193785499499,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047634487941718016,
      "loss": 4.8417,
      "mean_token_accuracy": 0.21262540221214293,
      "num_tokens": 34009629.0,
      "step": 18485
    },
    {
      "entropy": 5.211516618728638,
      "epoch": 1.544437019712663,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004763315902827059,
      "loss": 4.9487,
      "mean_token_accuracy": 0.19953279793262482,
      "num_tokens": 34019562.0,
      "step": 18490
    },
    {
      "entropy": 5.085388088226319,
      "epoch": 1.544854660875376,
      "grad_norm": 1.25,
      "learning_rate": 0.00047631829762367976,
      "loss": 4.8161,
      "mean_token_accuracy": 0.21158342510461808,
      "num_tokens": 34027674.0,
      "step": 18495
    },
    {
      "entropy": 5.19853630065918,
      "epoch": 1.545272302038089,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004763050014403346,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2127891331911087,
      "num_tokens": 34036874.0,
      "step": 18500
    },
    {
      "entropy": 5.262269020080566,
      "epoch": 1.5456899432008018,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047629170173290296,
      "loss": 5.0147,
      "mean_token_accuracy": 0.19773835837841033,
      "num_tokens": 34047184.0,
      "step": 18505
    },
    {
      "entropy": 5.186597204208374,
      "epoch": 1.546107584363515,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047627839850161784,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21016698181629181,
      "num_tokens": 34055925.0,
      "step": 18510
    },
    {
      "entropy": 5.157141590118409,
      "epoch": 1.5465252255262278,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047626509174671214,
      "loss": 4.9226,
      "mean_token_accuracy": 0.21020255237817764,
      "num_tokens": 34065401.0,
      "step": 18515
    },
    {
      "entropy": 5.018243074417114,
      "epoch": 1.5469428666889409,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004762517814684187,
      "loss": 4.8354,
      "mean_token_accuracy": 0.21726772785186768,
      "num_tokens": 34074103.0,
      "step": 18520
    },
    {
      "entropy": 5.103427457809448,
      "epoch": 1.547360507851654,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047623846766697056,
      "loss": 4.8941,
      "mean_token_accuracy": 0.21665998101234435,
      "num_tokens": 34082668.0,
      "step": 18525
    },
    {
      "entropy": 5.189871120452881,
      "epoch": 1.5477781490143667,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047622515034260077,
      "loss": 4.9212,
      "mean_token_accuracy": 0.20797317624092101,
      "num_tokens": 34091755.0,
      "step": 18530
    },
    {
      "entropy": 5.236175632476806,
      "epoch": 1.54819579017708,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004762118294955426,
      "loss": 5.0028,
      "mean_token_accuracy": 0.20663774609565735,
      "num_tokens": 34100760.0,
      "step": 18535
    },
    {
      "entropy": 5.131887483596802,
      "epoch": 1.5486134313397928,
      "grad_norm": 1.125,
      "learning_rate": 0.0004761985051260291,
      "loss": 4.8128,
      "mean_token_accuracy": 0.21222875267267227,
      "num_tokens": 34111280.0,
      "step": 18540
    },
    {
      "entropy": 5.142032146453857,
      "epoch": 1.5490310725025058,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047618517723429366,
      "loss": 4.9071,
      "mean_token_accuracy": 0.20662265866994858,
      "num_tokens": 34120545.0,
      "step": 18545
    },
    {
      "entropy": 5.145383358001709,
      "epoch": 1.5494487136652189,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047617184582056946,
      "loss": 4.8414,
      "mean_token_accuracy": 0.21343111246824265,
      "num_tokens": 34128631.0,
      "step": 18550
    },
    {
      "entropy": 5.182892990112305,
      "epoch": 1.549866354827932,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047615851088508995,
      "loss": 4.9181,
      "mean_token_accuracy": 0.20599696785211563,
      "num_tokens": 34139424.0,
      "step": 18555
    },
    {
      "entropy": 5.1824709415435795,
      "epoch": 1.550283995990645,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047614517242808866,
      "loss": 4.8729,
      "mean_token_accuracy": 0.2139705240726471,
      "num_tokens": 34148102.0,
      "step": 18560
    },
    {
      "entropy": 5.171601486206055,
      "epoch": 1.5507016371533577,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047613183044979885,
      "loss": 4.9631,
      "mean_token_accuracy": 0.20535570085048677,
      "num_tokens": 34156886.0,
      "step": 18565
    },
    {
      "entropy": 5.14026255607605,
      "epoch": 1.551119278316071,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047611848495045434,
      "loss": 4.8973,
      "mean_token_accuracy": 0.20715626925230027,
      "num_tokens": 34166710.0,
      "step": 18570
    },
    {
      "entropy": 5.195924854278564,
      "epoch": 1.5515369194787838,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047610513593028867,
      "loss": 4.999,
      "mean_token_accuracy": 0.19587974697351457,
      "num_tokens": 34176432.0,
      "step": 18575
    },
    {
      "entropy": 5.169472408294678,
      "epoch": 1.5519545606414968,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004760917833895355,
      "loss": 4.8113,
      "mean_token_accuracy": 0.2115744322538376,
      "num_tokens": 34185493.0,
      "step": 18580
    },
    {
      "entropy": 5.245399618148804,
      "epoch": 1.5523722018042099,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004760784273284286,
      "loss": 4.9818,
      "mean_token_accuracy": 0.1964244708418846,
      "num_tokens": 34195829.0,
      "step": 18585
    },
    {
      "entropy": 5.0305451393127445,
      "epoch": 1.5527898429669227,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004760650677472018,
      "loss": 4.7966,
      "mean_token_accuracy": 0.21557627022266387,
      "num_tokens": 34204543.0,
      "step": 18590
    },
    {
      "entropy": 5.154136896133423,
      "epoch": 1.553207484129636,
      "grad_norm": 1.25,
      "learning_rate": 0.000476051704646089,
      "loss": 4.8839,
      "mean_token_accuracy": 0.21157011091709138,
      "num_tokens": 34213620.0,
      "step": 18595
    },
    {
      "entropy": 5.208582210540771,
      "epoch": 1.5536251252923488,
      "grad_norm": 1.359375,
      "learning_rate": 0.000476038338025324,
      "loss": 4.9349,
      "mean_token_accuracy": 0.20439496785402297,
      "num_tokens": 34223292.0,
      "step": 18600
    },
    {
      "entropy": 5.188493156433106,
      "epoch": 1.5540427664550618,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047602496788514103,
      "loss": 4.9232,
      "mean_token_accuracy": 0.21165300607681276,
      "num_tokens": 34232753.0,
      "step": 18605
    },
    {
      "entropy": 5.178284072875977,
      "epoch": 1.5544604076177748,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004760115942257739,
      "loss": 4.9212,
      "mean_token_accuracy": 0.2052068218588829,
      "num_tokens": 34242101.0,
      "step": 18610
    },
    {
      "entropy": 5.210091829299927,
      "epoch": 1.5548780487804879,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047599821704745693,
      "loss": 4.9624,
      "mean_token_accuracy": 0.2108595609664917,
      "num_tokens": 34251401.0,
      "step": 18615
    },
    {
      "entropy": 5.1538914203643795,
      "epoch": 1.5552956899432009,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004759848363504243,
      "loss": 4.901,
      "mean_token_accuracy": 0.20908331722021103,
      "num_tokens": 34260508.0,
      "step": 18620
    },
    {
      "entropy": 5.128532266616821,
      "epoch": 1.5557133311059137,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004759714521349101,
      "loss": 4.8598,
      "mean_token_accuracy": 0.20463836193084717,
      "num_tokens": 34269391.0,
      "step": 18625
    },
    {
      "entropy": 5.112678050994873,
      "epoch": 1.556130972268627,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004759580644011487,
      "loss": 4.8106,
      "mean_token_accuracy": 0.21040337681770324,
      "num_tokens": 34278354.0,
      "step": 18630
    },
    {
      "entropy": 5.065941572189331,
      "epoch": 1.5565486134313398,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004759446731493745,
      "loss": 4.8186,
      "mean_token_accuracy": 0.22048127204179763,
      "num_tokens": 34287018.0,
      "step": 18635
    },
    {
      "entropy": 5.08178300857544,
      "epoch": 1.5569662545940528,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000475931278379822,
      "loss": 4.7827,
      "mean_token_accuracy": 0.2135027453303337,
      "num_tokens": 34296448.0,
      "step": 18640
    },
    {
      "entropy": 5.162529468536377,
      "epoch": 1.5573838957567658,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004759178800927255,
      "loss": 4.9466,
      "mean_token_accuracy": 0.21120925098657609,
      "num_tokens": 34305701.0,
      "step": 18645
    },
    {
      "entropy": 5.216934442520142,
      "epoch": 1.5578015369194786,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004759044782883198,
      "loss": 4.971,
      "mean_token_accuracy": 0.20713170766830444,
      "num_tokens": 34314694.0,
      "step": 18650
    },
    {
      "entropy": 5.159377384185791,
      "epoch": 1.558219178082192,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004758910729668392,
      "loss": 4.947,
      "mean_token_accuracy": 0.2051720604300499,
      "num_tokens": 34323758.0,
      "step": 18655
    },
    {
      "entropy": 5.134027433395386,
      "epoch": 1.5586368192449047,
      "grad_norm": 1.25,
      "learning_rate": 0.0004758776641285187,
      "loss": 4.8429,
      "mean_token_accuracy": 0.20727854669094087,
      "num_tokens": 34333019.0,
      "step": 18660
    },
    {
      "entropy": 5.174594449996948,
      "epoch": 1.5590544604076177,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004758642517735928,
      "loss": 4.8951,
      "mean_token_accuracy": 0.20851633399724961,
      "num_tokens": 34343220.0,
      "step": 18665
    },
    {
      "entropy": 5.1789812564849855,
      "epoch": 1.5594721015703308,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047585083590229643,
      "loss": 4.985,
      "mean_token_accuracy": 0.20107894986867905,
      "num_tokens": 34352993.0,
      "step": 18670
    },
    {
      "entropy": 5.158680057525634,
      "epoch": 1.5598897427330438,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047583741651486443,
      "loss": 4.9202,
      "mean_token_accuracy": 0.2142076686024666,
      "num_tokens": 34361541.0,
      "step": 18675
    },
    {
      "entropy": 5.162228631973266,
      "epoch": 1.5603073838957568,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004758239936115317,
      "loss": 4.9122,
      "mean_token_accuracy": 0.20743556171655655,
      "num_tokens": 34370951.0,
      "step": 18680
    },
    {
      "entropy": 5.131360483169556,
      "epoch": 1.5607250250584697,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047581056719253324,
      "loss": 4.9283,
      "mean_token_accuracy": 0.20784182846546173,
      "num_tokens": 34380232.0,
      "step": 18685
    },
    {
      "entropy": 5.2364397048950195,
      "epoch": 1.561142666221183,
      "grad_norm": 1.125,
      "learning_rate": 0.0004757971372581041,
      "loss": 5.0058,
      "mean_token_accuracy": 0.2045059993863106,
      "num_tokens": 34389124.0,
      "step": 18690
    },
    {
      "entropy": 5.142298793792724,
      "epoch": 1.5615603073838957,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047578370380847935,
      "loss": 4.8195,
      "mean_token_accuracy": 0.2066564753651619,
      "num_tokens": 34397428.0,
      "step": 18695
    },
    {
      "entropy": 5.148032712936401,
      "epoch": 1.5619779485466088,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047577026684389413,
      "loss": 4.8237,
      "mean_token_accuracy": 0.21577443331480026,
      "num_tokens": 34406059.0,
      "step": 18700
    },
    {
      "entropy": 5.1014564514160154,
      "epoch": 1.5623955897093218,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004757568263645838,
      "loss": 4.8558,
      "mean_token_accuracy": 0.21039193272590637,
      "num_tokens": 34415113.0,
      "step": 18705
    },
    {
      "entropy": 5.0988983631134035,
      "epoch": 1.5628132308720346,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004757433823707835,
      "loss": 4.8876,
      "mean_token_accuracy": 0.2108754724264145,
      "num_tokens": 34423966.0,
      "step": 18710
    },
    {
      "entropy": 5.017631769180298,
      "epoch": 1.5632308720347479,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004757299348627287,
      "loss": 4.8026,
      "mean_token_accuracy": 0.20653978437185289,
      "num_tokens": 34432690.0,
      "step": 18715
    },
    {
      "entropy": 5.19428448677063,
      "epoch": 1.5636485131974607,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047571648384065477,
      "loss": 4.8536,
      "mean_token_accuracy": 0.21167001873254776,
      "num_tokens": 34441536.0,
      "step": 18720
    },
    {
      "entropy": 5.160951423645019,
      "epoch": 1.5640661543601737,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004757030293047972,
      "loss": 4.8064,
      "mean_token_accuracy": 0.21561054438352584,
      "num_tokens": 34450121.0,
      "step": 18725
    },
    {
      "entropy": 5.166075563430786,
      "epoch": 1.5644837955228867,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004756895712553915,
      "loss": 4.9797,
      "mean_token_accuracy": 0.20361673533916474,
      "num_tokens": 34460332.0,
      "step": 18730
    },
    {
      "entropy": 5.172188091278076,
      "epoch": 1.5649014366855998,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047567610969267326,
      "loss": 4.9379,
      "mean_token_accuracy": 0.21106594204902648,
      "num_tokens": 34469424.0,
      "step": 18735
    },
    {
      "entropy": 5.227495527267456,
      "epoch": 1.5653190778483128,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004756626446168782,
      "loss": 4.9648,
      "mean_token_accuracy": 0.20884380340576172,
      "num_tokens": 34478663.0,
      "step": 18740
    },
    {
      "entropy": 5.088459873199463,
      "epoch": 1.5657367190110256,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047564917602824203,
      "loss": 4.7604,
      "mean_token_accuracy": 0.22568598240613938,
      "num_tokens": 34487358.0,
      "step": 18745
    },
    {
      "entropy": 5.149803256988525,
      "epoch": 1.5661543601737389,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004756357039270004,
      "loss": 4.9718,
      "mean_token_accuracy": 0.20665449500083924,
      "num_tokens": 34496039.0,
      "step": 18750
    },
    {
      "entropy": 5.165799331665039,
      "epoch": 1.5665720013364517,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047562222831338933,
      "loss": 4.9223,
      "mean_token_accuracy": 0.21239684671163558,
      "num_tokens": 34504810.0,
      "step": 18755
    },
    {
      "entropy": 5.153358793258667,
      "epoch": 1.5669896424991647,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047560874918764473,
      "loss": 4.9056,
      "mean_token_accuracy": 0.21075017601251603,
      "num_tokens": 34514039.0,
      "step": 18760
    },
    {
      "entropy": 5.167677164077759,
      "epoch": 1.5674072836618778,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047559526655000244,
      "loss": 4.8885,
      "mean_token_accuracy": 0.20465167760848998,
      "num_tokens": 34524134.0,
      "step": 18765
    },
    {
      "entropy": 5.182442998886108,
      "epoch": 1.5678249248245906,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004755817804006986,
      "loss": 4.9771,
      "mean_token_accuracy": 0.20576222687959672,
      "num_tokens": 34533777.0,
      "step": 18770
    },
    {
      "entropy": 5.230287885665893,
      "epoch": 1.5682425659873038,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004755682907399692,
      "loss": 4.9013,
      "mean_token_accuracy": 0.2109058305621147,
      "num_tokens": 34542875.0,
      "step": 18775
    },
    {
      "entropy": 5.203078174591065,
      "epoch": 1.5686602071500166,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047555479756805055,
      "loss": 4.987,
      "mean_token_accuracy": 0.20453484803438188,
      "num_tokens": 34552466.0,
      "step": 18780
    },
    {
      "entropy": 5.179382228851319,
      "epoch": 1.5690778483127297,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004755413008851786,
      "loss": 4.9299,
      "mean_token_accuracy": 0.20941899567842484,
      "num_tokens": 34562217.0,
      "step": 18785
    },
    {
      "entropy": 5.072654342651367,
      "epoch": 1.5694954894754427,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004755278006915899,
      "loss": 4.822,
      "mean_token_accuracy": 0.22220714390277863,
      "num_tokens": 34570407.0,
      "step": 18790
    },
    {
      "entropy": 5.0980956077575685,
      "epoch": 1.5699131306381557,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047551429698752064,
      "loss": 4.8562,
      "mean_token_accuracy": 0.20796814262866975,
      "num_tokens": 34579865.0,
      "step": 18795
    },
    {
      "entropy": 5.14849739074707,
      "epoch": 1.5703307718008688,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047550078977320724,
      "loss": 4.9343,
      "mean_token_accuracy": 0.21172842383384705,
      "num_tokens": 34588752.0,
      "step": 18800
    },
    {
      "entropy": 5.183244705200195,
      "epoch": 1.5707484129635816,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004754872790488862,
      "loss": 4.8457,
      "mean_token_accuracy": 0.21777550876140594,
      "num_tokens": 34597839.0,
      "step": 18805
    },
    {
      "entropy": 5.10694727897644,
      "epoch": 1.5711660541262948,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000475473764814794,
      "loss": 4.8254,
      "mean_token_accuracy": 0.21368143558502198,
      "num_tokens": 34607167.0,
      "step": 18810
    },
    {
      "entropy": 5.1761823177337645,
      "epoch": 1.5715836952890077,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047546024707116727,
      "loss": 4.923,
      "mean_token_accuracy": 0.19974349737167357,
      "num_tokens": 34616029.0,
      "step": 18815
    },
    {
      "entropy": 5.121559906005859,
      "epoch": 1.5720013364517207,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047544672581824256,
      "loss": 4.8736,
      "mean_token_accuracy": 0.21082795262336732,
      "num_tokens": 34625849.0,
      "step": 18820
    },
    {
      "entropy": 5.14666895866394,
      "epoch": 1.5724189776144337,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004754332010562567,
      "loss": 4.9312,
      "mean_token_accuracy": 0.20940391719341278,
      "num_tokens": 34635951.0,
      "step": 18825
    },
    {
      "entropy": 5.198271131515503,
      "epoch": 1.5728366187771465,
      "grad_norm": 1.25,
      "learning_rate": 0.0004754196727854463,
      "loss": 4.9658,
      "mean_token_accuracy": 0.203019180893898,
      "num_tokens": 34644799.0,
      "step": 18830
    },
    {
      "entropy": 5.178919410705566,
      "epoch": 1.5732542599398598,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047540614100604836,
      "loss": 4.9492,
      "mean_token_accuracy": 0.20755639523267747,
      "num_tokens": 34654377.0,
      "step": 18835
    },
    {
      "entropy": 5.239019966125488,
      "epoch": 1.5736719011025726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004753926057182997,
      "loss": 5.0238,
      "mean_token_accuracy": 0.21151710748672486,
      "num_tokens": 34663637.0,
      "step": 18840
    },
    {
      "entropy": 5.203082942962647,
      "epoch": 1.5740895422652856,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047537906692243716,
      "loss": 4.9593,
      "mean_token_accuracy": 0.2041359469294548,
      "num_tokens": 34673839.0,
      "step": 18845
    },
    {
      "entropy": 5.166948366165161,
      "epoch": 1.5745071834279987,
      "grad_norm": 1.171875,
      "learning_rate": 0.000475365524618698,
      "loss": 4.9179,
      "mean_token_accuracy": 0.21220589876174928,
      "num_tokens": 34684299.0,
      "step": 18850
    },
    {
      "entropy": 5.156102323532105,
      "epoch": 1.5749248245907117,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004753519788073191,
      "loss": 4.8052,
      "mean_token_accuracy": 0.21360131949186326,
      "num_tokens": 34693674.0,
      "step": 18855
    },
    {
      "entropy": 5.09609375,
      "epoch": 1.5753424657534247,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004753384294885376,
      "loss": 4.7976,
      "mean_token_accuracy": 0.21612902283668517,
      "num_tokens": 34702279.0,
      "step": 18860
    },
    {
      "entropy": 5.156481981277466,
      "epoch": 1.5757601069161375,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004753248766625908,
      "loss": 5.0026,
      "mean_token_accuracy": 0.20572623312473298,
      "num_tokens": 34711598.0,
      "step": 18865
    },
    {
      "entropy": 5.272520589828491,
      "epoch": 1.5761777480788508,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004753113203297159,
      "loss": 5.0288,
      "mean_token_accuracy": 0.20321780294179917,
      "num_tokens": 34721356.0,
      "step": 18870
    },
    {
      "entropy": 5.1547869682312015,
      "epoch": 1.5765953892415636,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004752977604901502,
      "loss": 4.8621,
      "mean_token_accuracy": 0.21960049569606782,
      "num_tokens": 34729658.0,
      "step": 18875
    },
    {
      "entropy": 5.032293319702148,
      "epoch": 1.5770130304042767,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047528419714413107,
      "loss": 4.8046,
      "mean_token_accuracy": 0.2223307117819786,
      "num_tokens": 34738298.0,
      "step": 18880
    },
    {
      "entropy": 5.092279100418091,
      "epoch": 1.5774306715669897,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047527063029189606,
      "loss": 4.8321,
      "mean_token_accuracy": 0.2186212420463562,
      "num_tokens": 34747544.0,
      "step": 18885
    },
    {
      "entropy": 5.111620950698852,
      "epoch": 1.5778483127297025,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004752570599336825,
      "loss": 4.8584,
      "mean_token_accuracy": 0.21108374893665313,
      "num_tokens": 34756380.0,
      "step": 18890
    },
    {
      "entropy": 5.095017290115356,
      "epoch": 1.5782659538924158,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047524348606972817,
      "loss": 4.8852,
      "mean_token_accuracy": 0.21261546313762664,
      "num_tokens": 34766164.0,
      "step": 18895
    },
    {
      "entropy": 5.1827075481414795,
      "epoch": 1.5786835950551286,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047522990870027055,
      "loss": 4.9232,
      "mean_token_accuracy": 0.20578682869672776,
      "num_tokens": 34775708.0,
      "step": 18900
    },
    {
      "entropy": 5.175114583969116,
      "epoch": 1.5791012362178416,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004752163278255473,
      "loss": 4.8398,
      "mean_token_accuracy": 0.21915908753871918,
      "num_tokens": 34785227.0,
      "step": 18905
    },
    {
      "entropy": 5.116562032699585,
      "epoch": 1.5795188773805546,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004752027434457963,
      "loss": 4.86,
      "mean_token_accuracy": 0.20783945471048354,
      "num_tokens": 34794331.0,
      "step": 18910
    },
    {
      "entropy": 5.1361565589904785,
      "epoch": 1.5799365185432677,
      "grad_norm": 1.125,
      "learning_rate": 0.0004751891555612552,
      "loss": 4.7884,
      "mean_token_accuracy": 0.21601420491933823,
      "num_tokens": 34803264.0,
      "step": 18915
    },
    {
      "entropy": 5.0267644882202145,
      "epoch": 1.5803541597059807,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004751755641721621,
      "loss": 4.8124,
      "mean_token_accuracy": 0.21459351032972335,
      "num_tokens": 34812161.0,
      "step": 18920
    },
    {
      "entropy": 5.102742624282837,
      "epoch": 1.5807718008686935,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047516196927875474,
      "loss": 4.8536,
      "mean_token_accuracy": 0.21126089245080948,
      "num_tokens": 34820987.0,
      "step": 18925
    },
    {
      "entropy": 5.1533080577850345,
      "epoch": 1.5811894420314068,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047514837088127117,
      "loss": 4.9306,
      "mean_token_accuracy": 0.21678370386362075,
      "num_tokens": 34830139.0,
      "step": 18930
    },
    {
      "entropy": 5.1701634407043455,
      "epoch": 1.5816070831941196,
      "grad_norm": 1.125,
      "learning_rate": 0.0004751347689799494,
      "loss": 4.8982,
      "mean_token_accuracy": 0.21232450157403945,
      "num_tokens": 34840178.0,
      "step": 18935
    },
    {
      "entropy": 5.2131823062896725,
      "epoch": 1.5820247243568326,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047512116357502766,
      "loss": 4.9597,
      "mean_token_accuracy": 0.20311215966939927,
      "num_tokens": 34849008.0,
      "step": 18940
    },
    {
      "entropy": 5.087475872039795,
      "epoch": 1.5824423655195456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047510755466674407,
      "loss": 4.834,
      "mean_token_accuracy": 0.21713546812534332,
      "num_tokens": 34858287.0,
      "step": 18945
    },
    {
      "entropy": 5.096488523483276,
      "epoch": 1.5828600066822585,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047509394225533685,
      "loss": 4.7761,
      "mean_token_accuracy": 0.21696523129940032,
      "num_tokens": 34867107.0,
      "step": 18950
    },
    {
      "entropy": 5.053239583969116,
      "epoch": 1.5832776478449717,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004750803263410443,
      "loss": 4.8415,
      "mean_token_accuracy": 0.22152031660079957,
      "num_tokens": 34876024.0,
      "step": 18955
    },
    {
      "entropy": 5.164964866638184,
      "epoch": 1.5836952890076845,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004750667069241048,
      "loss": 4.9463,
      "mean_token_accuracy": 0.20578233152627945,
      "num_tokens": 34885955.0,
      "step": 18960
    },
    {
      "entropy": 5.1431995868682865,
      "epoch": 1.5841129301703976,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004750530840047568,
      "loss": 4.9953,
      "mean_token_accuracy": 0.20624449402093886,
      "num_tokens": 34895041.0,
      "step": 18965
    },
    {
      "entropy": 5.169310998916626,
      "epoch": 1.5845305713331106,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047503945758323873,
      "loss": 4.7899,
      "mean_token_accuracy": 0.21874239146709443,
      "num_tokens": 34903575.0,
      "step": 18970
    },
    {
      "entropy": 5.10904221534729,
      "epoch": 1.5849482124958236,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004750258276597892,
      "loss": 4.8561,
      "mean_token_accuracy": 0.21443795412778854,
      "num_tokens": 34913326.0,
      "step": 18975
    },
    {
      "entropy": 5.13879017829895,
      "epoch": 1.5853658536585367,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047501219423464673,
      "loss": 4.8972,
      "mean_token_accuracy": 0.21358640342950821,
      "num_tokens": 34921652.0,
      "step": 18980
    },
    {
      "entropy": 5.169260931015015,
      "epoch": 1.5857834948212495,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004749985573080501,
      "loss": 4.9178,
      "mean_token_accuracy": 0.20456042140722275,
      "num_tokens": 34931885.0,
      "step": 18985
    },
    {
      "entropy": 5.155839681625366,
      "epoch": 1.5862011359839627,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000474984916880238,
      "loss": 4.9817,
      "mean_token_accuracy": 0.20217206180095673,
      "num_tokens": 34942105.0,
      "step": 18990
    },
    {
      "entropy": 5.116054201126099,
      "epoch": 1.5866187771466755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004749712729514491,
      "loss": 4.8445,
      "mean_token_accuracy": 0.21162777692079543,
      "num_tokens": 34950231.0,
      "step": 18995
    },
    {
      "entropy": 5.209437131881714,
      "epoch": 1.5870364183093886,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004749576255219225,
      "loss": 4.9856,
      "mean_token_accuracy": 0.20411159545183183,
      "num_tokens": 34959485.0,
      "step": 19000
    },
    {
      "entropy": 5.122876834869385,
      "epoch": 1.5874540594721016,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004749439745918969,
      "loss": 4.8123,
      "mean_token_accuracy": 0.2135621726512909,
      "num_tokens": 34968711.0,
      "step": 19005
    },
    {
      "entropy": 5.1361250400543215,
      "epoch": 1.5878717006348144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047493032016161134,
      "loss": 4.9308,
      "mean_token_accuracy": 0.20988865643739701,
      "num_tokens": 34978241.0,
      "step": 19010
    },
    {
      "entropy": 5.144276285171509,
      "epoch": 1.5882893417975277,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004749166622313049,
      "loss": 4.8068,
      "mean_token_accuracy": 0.21609339267015457,
      "num_tokens": 34987356.0,
      "step": 19015
    },
    {
      "entropy": 5.290378427505493,
      "epoch": 1.5887069829602405,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047490300080121667,
      "loss": 4.9571,
      "mean_token_accuracy": 0.20034946799278258,
      "num_tokens": 34996044.0,
      "step": 19020
    },
    {
      "entropy": 5.178801107406616,
      "epoch": 1.5891246241229535,
      "grad_norm": 1.375,
      "learning_rate": 0.00047488933587158585,
      "loss": 4.9033,
      "mean_token_accuracy": 0.20526841878890992,
      "num_tokens": 35005734.0,
      "step": 19025
    },
    {
      "entropy": 5.063468360900879,
      "epoch": 1.5895422652856666,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004748756674426514,
      "loss": 4.8026,
      "mean_token_accuracy": 0.21780952513217927,
      "num_tokens": 35014859.0,
      "step": 19030
    },
    {
      "entropy": 5.19588041305542,
      "epoch": 1.5899599064483796,
      "grad_norm": 1.421875,
      "learning_rate": 0.000474861995514653,
      "loss": 4.9762,
      "mean_token_accuracy": 0.1999057799577713,
      "num_tokens": 35023885.0,
      "step": 19035
    },
    {
      "entropy": 5.204515075683593,
      "epoch": 1.5903775476110926,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004748483200878297,
      "loss": 4.8437,
      "mean_token_accuracy": 0.219435678422451,
      "num_tokens": 35031901.0,
      "step": 19040
    },
    {
      "entropy": 5.205347061157227,
      "epoch": 1.5907951887738054,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000474834641162421,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20154365301132202,
      "num_tokens": 35041601.0,
      "step": 19045
    },
    {
      "entropy": 5.009580659866333,
      "epoch": 1.5912128299365187,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047482095873866635,
      "loss": 4.8575,
      "mean_token_accuracy": 0.2148676335811615,
      "num_tokens": 35050982.0,
      "step": 19050
    },
    {
      "entropy": 5.09777340888977,
      "epoch": 1.5916304710992315,
      "grad_norm": 1.25,
      "learning_rate": 0.0004748072728168053,
      "loss": 4.7821,
      "mean_token_accuracy": 0.2161940485239029,
      "num_tokens": 35060548.0,
      "step": 19055
    },
    {
      "entropy": 5.105748462677002,
      "epoch": 1.5920481122619445,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004747935833970775,
      "loss": 4.9002,
      "mean_token_accuracy": 0.21056068390607835,
      "num_tokens": 35069396.0,
      "step": 19060
    },
    {
      "entropy": 5.147435665130615,
      "epoch": 1.5924657534246576,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004747798904797225,
      "loss": 4.9014,
      "mean_token_accuracy": 0.20987477898597717,
      "num_tokens": 35078694.0,
      "step": 19065
    },
    {
      "entropy": 5.156363916397095,
      "epoch": 1.5928833945873704,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047476619406497996,
      "loss": 4.7805,
      "mean_token_accuracy": 0.2149868294596672,
      "num_tokens": 35087336.0,
      "step": 19070
    },
    {
      "entropy": 5.173246192932129,
      "epoch": 1.5933010357500836,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004747524941530898,
      "loss": 4.9381,
      "mean_token_accuracy": 0.20851770490407945,
      "num_tokens": 35095768.0,
      "step": 19075
    },
    {
      "entropy": 5.024761772155761,
      "epoch": 1.5937186769127965,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004747387907442918,
      "loss": 4.8033,
      "mean_token_accuracy": 0.20796211063861847,
      "num_tokens": 35104694.0,
      "step": 19080
    },
    {
      "entropy": 5.167335939407349,
      "epoch": 1.5941363180755095,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047472508383882586,
      "loss": 4.9042,
      "mean_token_accuracy": 0.2030287817120552,
      "num_tokens": 35114159.0,
      "step": 19085
    },
    {
      "entropy": 5.162821197509766,
      "epoch": 1.5945539592382225,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004747113734369319,
      "loss": 4.8862,
      "mean_token_accuracy": 0.20696432143449783,
      "num_tokens": 35122786.0,
      "step": 19090
    },
    {
      "entropy": 5.197068405151367,
      "epoch": 1.5949716004009356,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004746976595388499,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21744738072156905,
      "num_tokens": 35131369.0,
      "step": 19095
    },
    {
      "entropy": 5.136154699325561,
      "epoch": 1.5953892415636486,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047468394214482,
      "loss": 4.8926,
      "mean_token_accuracy": 0.21857378631830215,
      "num_tokens": 35140533.0,
      "step": 19100
    },
    {
      "entropy": 5.156340026855469,
      "epoch": 1.5958068827263614,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004746702212550824,
      "loss": 4.8744,
      "mean_token_accuracy": 0.2093157008290291,
      "num_tokens": 35149149.0,
      "step": 19105
    },
    {
      "entropy": 5.1571427345275875,
      "epoch": 1.5962245238890747,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047465649686987717,
      "loss": 4.9162,
      "mean_token_accuracy": 0.2107882857322693,
      "num_tokens": 35159143.0,
      "step": 19110
    },
    {
      "entropy": 5.111605882644653,
      "epoch": 1.5966421650517875,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004746427689894447,
      "loss": 4.8212,
      "mean_token_accuracy": 0.21379026472568513,
      "num_tokens": 35167135.0,
      "step": 19115
    },
    {
      "entropy": 5.17090072631836,
      "epoch": 1.5970598062145005,
      "grad_norm": 1.125,
      "learning_rate": 0.0004746290376140252,
      "loss": 4.9048,
      "mean_token_accuracy": 0.2105774849653244,
      "num_tokens": 35176724.0,
      "step": 19120
    },
    {
      "entropy": 5.1918927192687985,
      "epoch": 1.5974774473772135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004746153027438592,
      "loss": 4.8715,
      "mean_token_accuracy": 0.21290557086467743,
      "num_tokens": 35185884.0,
      "step": 19125
    },
    {
      "entropy": 5.202436399459839,
      "epoch": 1.5978950885399263,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004746015643791869,
      "loss": 4.9714,
      "mean_token_accuracy": 0.20289166420698165,
      "num_tokens": 35196562.0,
      "step": 19130
    },
    {
      "entropy": 5.125077629089356,
      "epoch": 1.5983127297026396,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000474587822520249,
      "loss": 4.8266,
      "mean_token_accuracy": 0.21624946296215058,
      "num_tokens": 35205351.0,
      "step": 19135
    },
    {
      "entropy": 5.105741834640503,
      "epoch": 1.5987303708653524,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004745740771672861,
      "loss": 4.85,
      "mean_token_accuracy": 0.21219571828842163,
      "num_tokens": 35214603.0,
      "step": 19140
    },
    {
      "entropy": 5.183910465240478,
      "epoch": 1.5991480120280654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004745603283205387,
      "loss": 4.9016,
      "mean_token_accuracy": 0.2096925273537636,
      "num_tokens": 35224175.0,
      "step": 19145
    },
    {
      "entropy": 5.119003868103027,
      "epoch": 1.5995656531907785,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047454657598024755,
      "loss": 4.9299,
      "mean_token_accuracy": 0.2108020856976509,
      "num_tokens": 35233002.0,
      "step": 19150
    },
    {
      "entropy": 5.18049726486206,
      "epoch": 1.5999832943534915,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047453282014665334,
      "loss": 4.7923,
      "mean_token_accuracy": 0.21773571521043777,
      "num_tokens": 35241281.0,
      "step": 19155
    },
    {
      "entropy": 5.107484006881714,
      "epoch": 1.6004009355162045,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047451906081999695,
      "loss": 4.8798,
      "mean_token_accuracy": 0.2169203445315361,
      "num_tokens": 35250289.0,
      "step": 19160
    },
    {
      "entropy": 5.057656335830688,
      "epoch": 1.6008185766789174,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004745052980005193,
      "loss": 4.8248,
      "mean_token_accuracy": 0.21678472608327864,
      "num_tokens": 35260221.0,
      "step": 19165
    },
    {
      "entropy": 5.183615779876709,
      "epoch": 1.6012362178416306,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004744915316884612,
      "loss": 4.8216,
      "mean_token_accuracy": 0.21570214480161667,
      "num_tokens": 35269409.0,
      "step": 19170
    },
    {
      "entropy": 5.126633214950561,
      "epoch": 1.6016538590043434,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004744777618840637,
      "loss": 4.8669,
      "mean_token_accuracy": 0.21251083463430404,
      "num_tokens": 35278241.0,
      "step": 19175
    },
    {
      "entropy": 5.06670503616333,
      "epoch": 1.6020715001670565,
      "grad_norm": 1.15625,
      "learning_rate": 0.000474463988587568,
      "loss": 4.8223,
      "mean_token_accuracy": 0.21326142996549607,
      "num_tokens": 35287124.0,
      "step": 19180
    },
    {
      "entropy": 5.165285968780518,
      "epoch": 1.6024891413297695,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047445021179921495,
      "loss": 4.8664,
      "mean_token_accuracy": 0.2119393065571785,
      "num_tokens": 35296021.0,
      "step": 19185
    },
    {
      "entropy": 5.204792118072509,
      "epoch": 1.6029067824924823,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004744364315192459,
      "loss": 4.8889,
      "mean_token_accuracy": 0.20935933589935302,
      "num_tokens": 35304866.0,
      "step": 19190
    },
    {
      "entropy": 5.141434621810913,
      "epoch": 1.6033244236551956,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047442264774790203,
      "loss": 4.9219,
      "mean_token_accuracy": 0.20917271673679352,
      "num_tokens": 35314165.0,
      "step": 19195
    },
    {
      "entropy": 5.09385986328125,
      "epoch": 1.6037420648179084,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004744088604854248,
      "loss": 4.7878,
      "mean_token_accuracy": 0.21447841078042984,
      "num_tokens": 35323166.0,
      "step": 19200
    },
    {
      "entropy": 5.180385255813599,
      "epoch": 1.6041597059806214,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047439506973205527,
      "loss": 4.9575,
      "mean_token_accuracy": 0.20647502243518828,
      "num_tokens": 35331797.0,
      "step": 19205
    },
    {
      "entropy": 5.061083745956421,
      "epoch": 1.6045773471433344,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004743812754880351,
      "loss": 4.7676,
      "mean_token_accuracy": 0.22539484202861787,
      "num_tokens": 35340914.0,
      "step": 19210
    },
    {
      "entropy": 5.142693901062012,
      "epoch": 1.6049949883060475,
      "grad_norm": 1.25,
      "learning_rate": 0.0004743674777536057,
      "loss": 4.9108,
      "mean_token_accuracy": 0.21076854020357133,
      "num_tokens": 35350142.0,
      "step": 19215
    },
    {
      "entropy": 5.252508068084717,
      "epoch": 1.6054126294687605,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004743536765290087,
      "loss": 5.0378,
      "mean_token_accuracy": 0.2037280172109604,
      "num_tokens": 35359673.0,
      "step": 19220
    },
    {
      "entropy": 5.14580888748169,
      "epoch": 1.6058302706314733,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047433987181448546,
      "loss": 4.8301,
      "mean_token_accuracy": 0.21587164998054503,
      "num_tokens": 35368922.0,
      "step": 19225
    },
    {
      "entropy": 5.098486375808716,
      "epoch": 1.6062479117941866,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004743260636102779,
      "loss": 4.8961,
      "mean_token_accuracy": 0.210992468893528,
      "num_tokens": 35378166.0,
      "step": 19230
    },
    {
      "entropy": 5.106770086288452,
      "epoch": 1.6066655529568994,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047431225191662776,
      "loss": 4.8001,
      "mean_token_accuracy": 0.21405679285526275,
      "num_tokens": 35387585.0,
      "step": 19235
    },
    {
      "entropy": 5.1707720279693605,
      "epoch": 1.6070831941196124,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047429843673377664,
      "loss": 4.939,
      "mean_token_accuracy": 0.20843535810709,
      "num_tokens": 35396344.0,
      "step": 19240
    },
    {
      "entropy": 5.2286255836486815,
      "epoch": 1.6075008352823255,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047428461806196646,
      "loss": 4.9997,
      "mean_token_accuracy": 0.19837345480918883,
      "num_tokens": 35405933.0,
      "step": 19245
    },
    {
      "entropy": 5.167977714538575,
      "epoch": 1.6079184764450383,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004742707959014393,
      "loss": 4.8792,
      "mean_token_accuracy": 0.2151471734046936,
      "num_tokens": 35414091.0,
      "step": 19250
    },
    {
      "entropy": 5.174715805053711,
      "epoch": 1.6083361176077515,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047425697025243693,
      "loss": 4.9487,
      "mean_token_accuracy": 0.20079672932624817,
      "num_tokens": 35423835.0,
      "step": 19255
    },
    {
      "entropy": 5.153983497619629,
      "epoch": 1.6087537587704643,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047424314111520134,
      "loss": 4.8845,
      "mean_token_accuracy": 0.2084815800189972,
      "num_tokens": 35433619.0,
      "step": 19260
    },
    {
      "entropy": 5.231896018981933,
      "epoch": 1.6091713999331774,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004742293084899749,
      "loss": 4.9748,
      "mean_token_accuracy": 0.20800044238567353,
      "num_tokens": 35443759.0,
      "step": 19265
    },
    {
      "entropy": 5.174439811706543,
      "epoch": 1.6095890410958904,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004742154723769995,
      "loss": 4.8053,
      "mean_token_accuracy": 0.2165703609585762,
      "num_tokens": 35452296.0,
      "step": 19270
    },
    {
      "entropy": 5.131257629394531,
      "epoch": 1.6100066822586034,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004742016327765175,
      "loss": 4.9122,
      "mean_token_accuracy": 0.20915295630693437,
      "num_tokens": 35462082.0,
      "step": 19275
    },
    {
      "entropy": 5.086362981796265,
      "epoch": 1.6104243234213165,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004741877896887711,
      "loss": 4.9038,
      "mean_token_accuracy": 0.2073094964027405,
      "num_tokens": 35470473.0,
      "step": 19280
    },
    {
      "entropy": 5.163616085052491,
      "epoch": 1.6108419645840293,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004741739431140027,
      "loss": 4.8174,
      "mean_token_accuracy": 0.21539407223463058,
      "num_tokens": 35479257.0,
      "step": 19285
    },
    {
      "entropy": 5.152368688583374,
      "epoch": 1.6112596057467425,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047416009305245474,
      "loss": 5.0107,
      "mean_token_accuracy": 0.20429852604866028,
      "num_tokens": 35489149.0,
      "step": 19290
    },
    {
      "entropy": 5.140770769119262,
      "epoch": 1.6116772469094554,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004741462395043695,
      "loss": 4.884,
      "mean_token_accuracy": 0.21324747055768967,
      "num_tokens": 35497939.0,
      "step": 19295
    },
    {
      "entropy": 5.1690428256988525,
      "epoch": 1.6120948880721684,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047413238246998975,
      "loss": 4.8347,
      "mean_token_accuracy": 0.2174767255783081,
      "num_tokens": 35506645.0,
      "step": 19300
    },
    {
      "entropy": 5.186855602264404,
      "epoch": 1.6125125292348814,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004741185219495579,
      "loss": 4.8483,
      "mean_token_accuracy": 0.21301568895578385,
      "num_tokens": 35514909.0,
      "step": 19305
    },
    {
      "entropy": 5.032540273666382,
      "epoch": 1.6129301703975942,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004741046579433166,
      "loss": 4.7906,
      "mean_token_accuracy": 0.22202490717172624,
      "num_tokens": 35523634.0,
      "step": 19310
    },
    {
      "entropy": 5.081476783752441,
      "epoch": 1.6133478115603075,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047409079045150865,
      "loss": 4.8679,
      "mean_token_accuracy": 0.21116019636392594,
      "num_tokens": 35533167.0,
      "step": 19315
    },
    {
      "entropy": 5.226933145523072,
      "epoch": 1.6137654527230203,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047407691947437677,
      "loss": 4.9593,
      "mean_token_accuracy": 0.2114596888422966,
      "num_tokens": 35542169.0,
      "step": 19320
    },
    {
      "entropy": 5.219224357604981,
      "epoch": 1.6141830938857333,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047406304501216375,
      "loss": 4.9295,
      "mean_token_accuracy": 0.21526082903146743,
      "num_tokens": 35550734.0,
      "step": 19325
    },
    {
      "entropy": 5.091986227035522,
      "epoch": 1.6146007350484464,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047404916706511255,
      "loss": 4.9139,
      "mean_token_accuracy": 0.20581378042697906,
      "num_tokens": 35559592.0,
      "step": 19330
    },
    {
      "entropy": 5.125500011444092,
      "epoch": 1.6150183762111594,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047403528563346603,
      "loss": 4.8413,
      "mean_token_accuracy": 0.2102677345275879,
      "num_tokens": 35568954.0,
      "step": 19335
    },
    {
      "entropy": 5.114569854736328,
      "epoch": 1.6154360173738724,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047402140071746743,
      "loss": 4.7477,
      "mean_token_accuracy": 0.2098894238471985,
      "num_tokens": 35578091.0,
      "step": 19340
    },
    {
      "entropy": 5.0753367900848385,
      "epoch": 1.6158536585365852,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047400751231735946,
      "loss": 4.869,
      "mean_token_accuracy": 0.21604908108711243,
      "num_tokens": 35586772.0,
      "step": 19345
    },
    {
      "entropy": 5.064572381973266,
      "epoch": 1.6162712996992985,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047399362043338547,
      "loss": 4.7573,
      "mean_token_accuracy": 0.216802416741848,
      "num_tokens": 35595866.0,
      "step": 19350
    },
    {
      "entropy": 5.122611665725708,
      "epoch": 1.6166889408620113,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004739797250657887,
      "loss": 4.8517,
      "mean_token_accuracy": 0.2118723139166832,
      "num_tokens": 35604721.0,
      "step": 19355
    },
    {
      "entropy": 5.137160682678223,
      "epoch": 1.6171065820247243,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047396582621481224,
      "loss": 4.906,
      "mean_token_accuracy": 0.2107534646987915,
      "num_tokens": 35614181.0,
      "step": 19360
    },
    {
      "entropy": 5.176036357879639,
      "epoch": 1.6175242231874374,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047395192388069955,
      "loss": 4.9152,
      "mean_token_accuracy": 0.20524170845746995,
      "num_tokens": 35622646.0,
      "step": 19365
    },
    {
      "entropy": 5.115852022171021,
      "epoch": 1.6179418643501502,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047393801806369396,
      "loss": 4.8745,
      "mean_token_accuracy": 0.20654046088457106,
      "num_tokens": 35631617.0,
      "step": 19370
    },
    {
      "entropy": 5.212823343276978,
      "epoch": 1.6183595055128634,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047392410876403887,
      "loss": 5.0,
      "mean_token_accuracy": 0.20753039717674254,
      "num_tokens": 35641000.0,
      "step": 19375
    },
    {
      "entropy": 5.126144695281982,
      "epoch": 1.6187771466755763,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047391019598197785,
      "loss": 4.9242,
      "mean_token_accuracy": 0.2038044586777687,
      "num_tokens": 35650756.0,
      "step": 19380
    },
    {
      "entropy": 5.142666816711426,
      "epoch": 1.6191947878382893,
      "grad_norm": 1.25,
      "learning_rate": 0.0004738962797177544,
      "loss": 4.8255,
      "mean_token_accuracy": 0.21107469201087953,
      "num_tokens": 35659625.0,
      "step": 19385
    },
    {
      "entropy": 5.141479349136352,
      "epoch": 1.6196124290010023,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004738823599716121,
      "loss": 4.7969,
      "mean_token_accuracy": 0.22033247351646423,
      "num_tokens": 35668451.0,
      "step": 19390
    },
    {
      "entropy": 5.008991479873657,
      "epoch": 1.6200300701637154,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047386843674379484,
      "loss": 4.756,
      "mean_token_accuracy": 0.21891753226518632,
      "num_tokens": 35677393.0,
      "step": 19395
    },
    {
      "entropy": 5.06404070854187,
      "epoch": 1.6204477113264284,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004738545100345461,
      "loss": 4.9204,
      "mean_token_accuracy": 0.20072436034679414,
      "num_tokens": 35687321.0,
      "step": 19400
    },
    {
      "entropy": 5.212156105041504,
      "epoch": 1.6208653524891412,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047384057984410985,
      "loss": 4.8992,
      "mean_token_accuracy": 0.2080441653728485,
      "num_tokens": 35697504.0,
      "step": 19405
    },
    {
      "entropy": 5.226457214355468,
      "epoch": 1.6212829936518545,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047382664617272985,
      "loss": 4.9512,
      "mean_token_accuracy": 0.20506729185581207,
      "num_tokens": 35707585.0,
      "step": 19410
    },
    {
      "entropy": 5.15043363571167,
      "epoch": 1.6217006348145673,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047381270902065013,
      "loss": 4.869,
      "mean_token_accuracy": 0.2115656018257141,
      "num_tokens": 35716722.0,
      "step": 19415
    },
    {
      "entropy": 5.207930040359497,
      "epoch": 1.6221182759772803,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047379876838811463,
      "loss": 4.9074,
      "mean_token_accuracy": 0.20265856385231018,
      "num_tokens": 35725865.0,
      "step": 19420
    },
    {
      "entropy": 5.102642679214478,
      "epoch": 1.6225359171399933,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004737848242753674,
      "loss": 4.8123,
      "mean_token_accuracy": 0.21778395920991897,
      "num_tokens": 35734813.0,
      "step": 19425
    },
    {
      "entropy": 5.119083499908447,
      "epoch": 1.6229535583027062,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004737708766826525,
      "loss": 4.9179,
      "mean_token_accuracy": 0.20339806079864503,
      "num_tokens": 35744311.0,
      "step": 19430
    },
    {
      "entropy": 5.123318862915039,
      "epoch": 1.6233711994654194,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004737569256102142,
      "loss": 4.7931,
      "mean_token_accuracy": 0.2117394581437111,
      "num_tokens": 35754239.0,
      "step": 19435
    },
    {
      "entropy": 5.146233987808228,
      "epoch": 1.6237888406281322,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047374297105829663,
      "loss": 4.8938,
      "mean_token_accuracy": 0.20056841671466827,
      "num_tokens": 35763241.0,
      "step": 19440
    },
    {
      "entropy": 5.186574029922485,
      "epoch": 1.6242064817908453,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047372901302714413,
      "loss": 4.8831,
      "mean_token_accuracy": 0.21573854088783265,
      "num_tokens": 35772388.0,
      "step": 19445
    },
    {
      "entropy": 5.0873500347137455,
      "epoch": 1.6246241229535583,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047371505151700105,
      "loss": 4.7827,
      "mean_token_accuracy": 0.214072647690773,
      "num_tokens": 35781949.0,
      "step": 19450
    },
    {
      "entropy": 5.106070327758789,
      "epoch": 1.6250417641162713,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047370108652811186,
      "loss": 4.877,
      "mean_token_accuracy": 0.20803916603326797,
      "num_tokens": 35791639.0,
      "step": 19455
    },
    {
      "entropy": 5.153497219085693,
      "epoch": 1.6254594052789844,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004736871180607209,
      "loss": 4.861,
      "mean_token_accuracy": 0.20990934520959853,
      "num_tokens": 35801392.0,
      "step": 19460
    },
    {
      "entropy": 5.165641975402832,
      "epoch": 1.6258770464416972,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047367314611507276,
      "loss": 4.9545,
      "mean_token_accuracy": 0.20704147070646287,
      "num_tokens": 35811016.0,
      "step": 19465
    },
    {
      "entropy": 5.1411652088165285,
      "epoch": 1.6262946876044104,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047365917069141215,
      "loss": 4.8732,
      "mean_token_accuracy": 0.20838245898485183,
      "num_tokens": 35819552.0,
      "step": 19470
    },
    {
      "entropy": 5.127146196365357,
      "epoch": 1.6267123287671232,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004736451917899835,
      "loss": 4.8363,
      "mean_token_accuracy": 0.21487400382757188,
      "num_tokens": 35828855.0,
      "step": 19475
    },
    {
      "entropy": 5.126818561553955,
      "epoch": 1.6271299699298363,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004736312094110317,
      "loss": 4.8014,
      "mean_token_accuracy": 0.223326575756073,
      "num_tokens": 35838140.0,
      "step": 19480
    },
    {
      "entropy": 5.160595035552978,
      "epoch": 1.6275476110925493,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004736172235548016,
      "loss": 4.9816,
      "mean_token_accuracy": 0.207153756916523,
      "num_tokens": 35847133.0,
      "step": 19485
    },
    {
      "entropy": 5.099270486831665,
      "epoch": 1.6279652522552621,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004736032342215378,
      "loss": 4.8941,
      "mean_token_accuracy": 0.21093732863664627,
      "num_tokens": 35856917.0,
      "step": 19490
    },
    {
      "entropy": 5.174907112121582,
      "epoch": 1.6283828934179754,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047358924141148543,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2181338980793953,
      "num_tokens": 35864739.0,
      "step": 19495
    },
    {
      "entropy": 5.071548891067505,
      "epoch": 1.6288005345806882,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047357524512488925,
      "loss": 4.7882,
      "mean_token_accuracy": 0.21329980194568635,
      "num_tokens": 35874105.0,
      "step": 19500
    },
    {
      "entropy": 5.088203096389771,
      "epoch": 1.6292181757434012,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004735612453619944,
      "loss": 4.7643,
      "mean_token_accuracy": 0.2168786808848381,
      "num_tokens": 35883492.0,
      "step": 19505
    },
    {
      "entropy": 5.057540798187256,
      "epoch": 1.6296358169061143,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000473547242123046,
      "loss": 4.8911,
      "mean_token_accuracy": 0.20946891456842423,
      "num_tokens": 35892902.0,
      "step": 19510
    },
    {
      "entropy": 5.24179310798645,
      "epoch": 1.6300534580688273,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004735332354082891,
      "loss": 5.0166,
      "mean_token_accuracy": 0.20590457171201706,
      "num_tokens": 35903130.0,
      "step": 19515
    },
    {
      "entropy": 5.275386953353882,
      "epoch": 1.6304710992315403,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047351922521796894,
      "loss": 5.0592,
      "mean_token_accuracy": 0.19180281609296798,
      "num_tokens": 35913462.0,
      "step": 19520
    },
    {
      "entropy": 5.11017336845398,
      "epoch": 1.6308887403942531,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004735052115523307,
      "loss": 4.8906,
      "mean_token_accuracy": 0.21035451143980027,
      "num_tokens": 35922340.0,
      "step": 19525
    },
    {
      "entropy": 5.233419513702392,
      "epoch": 1.6313063815569664,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004734911944116199,
      "loss": 4.9765,
      "mean_token_accuracy": 0.20198761224746703,
      "num_tokens": 35931558.0,
      "step": 19530
    },
    {
      "entropy": 5.114267206192016,
      "epoch": 1.6317240227196792,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047347717379608184,
      "loss": 4.895,
      "mean_token_accuracy": 0.20736209750175477,
      "num_tokens": 35941811.0,
      "step": 19535
    },
    {
      "entropy": 5.207329607009887,
      "epoch": 1.6321416638823922,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047346314970596185,
      "loss": 4.969,
      "mean_token_accuracy": 0.20575950890779496,
      "num_tokens": 35951134.0,
      "step": 19540
    },
    {
      "entropy": 5.262368392944336,
      "epoch": 1.6325593050451053,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047344912214150565,
      "loss": 4.9771,
      "mean_token_accuracy": 0.20156730562448502,
      "num_tokens": 35960716.0,
      "step": 19545
    },
    {
      "entropy": 5.050384902954102,
      "epoch": 1.632976946207818,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004734350911029586,
      "loss": 4.7419,
      "mean_token_accuracy": 0.21714129149913788,
      "num_tokens": 35968931.0,
      "step": 19550
    },
    {
      "entropy": 5.123586893081665,
      "epoch": 1.6333945873705313,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047342105659056645,
      "loss": 4.8618,
      "mean_token_accuracy": 0.2130320444703102,
      "num_tokens": 35978009.0,
      "step": 19555
    },
    {
      "entropy": 5.122021198272705,
      "epoch": 1.6338122285332441,
      "grad_norm": 1.125,
      "learning_rate": 0.0004734070186045749,
      "loss": 4.9219,
      "mean_token_accuracy": 0.20954838395118713,
      "num_tokens": 35988025.0,
      "step": 19560
    },
    {
      "entropy": 5.112794637680054,
      "epoch": 1.6342298696959572,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004733929771452296,
      "loss": 4.8713,
      "mean_token_accuracy": 0.20992948859930038,
      "num_tokens": 35997074.0,
      "step": 19565
    },
    {
      "entropy": 5.150512409210205,
      "epoch": 1.6346475108586702,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004733789322127764,
      "loss": 4.8884,
      "mean_token_accuracy": 0.21501000970602036,
      "num_tokens": 36006038.0,
      "step": 19570
    },
    {
      "entropy": 5.25858063697815,
      "epoch": 1.6350651520213833,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004733648838074613,
      "loss": 5.0092,
      "mean_token_accuracy": 0.20158184617757796,
      "num_tokens": 36014554.0,
      "step": 19575
    },
    {
      "entropy": 5.129584836959839,
      "epoch": 1.6354827931840963,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047335083192953,
      "loss": 4.9069,
      "mean_token_accuracy": 0.20901763439178467,
      "num_tokens": 36023538.0,
      "step": 19580
    },
    {
      "entropy": 5.1813645362854,
      "epoch": 1.635900434346809,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004733367765792288,
      "loss": 4.8911,
      "mean_token_accuracy": 0.20225934982299804,
      "num_tokens": 36033537.0,
      "step": 19585
    },
    {
      "entropy": 5.180348253250122,
      "epoch": 1.6363180755095224,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047332271775680346,
      "loss": 4.7844,
      "mean_token_accuracy": 0.21601081639528275,
      "num_tokens": 36042942.0,
      "step": 19590
    },
    {
      "entropy": 5.001672601699829,
      "epoch": 1.6367357166722352,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004733086554625004,
      "loss": 4.8282,
      "mean_token_accuracy": 0.2205368921160698,
      "num_tokens": 36052416.0,
      "step": 19595
    },
    {
      "entropy": 5.049933385848999,
      "epoch": 1.6371533578349482,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004732945896965654,
      "loss": 4.8465,
      "mean_token_accuracy": 0.2149621218442917,
      "num_tokens": 36061577.0,
      "step": 19600
    },
    {
      "entropy": 5.138888263702393,
      "epoch": 1.6375709989976612,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047328052045924506,
      "loss": 4.9589,
      "mean_token_accuracy": 0.2014012798666954,
      "num_tokens": 36071829.0,
      "step": 19605
    },
    {
      "entropy": 5.081200313568115,
      "epoch": 1.637988640160374,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004732664477507855,
      "loss": 4.7623,
      "mean_token_accuracy": 0.2165132537484169,
      "num_tokens": 36080378.0,
      "step": 19610
    },
    {
      "entropy": 5.137047147750854,
      "epoch": 1.6384062813230873,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047325237157143314,
      "loss": 4.8692,
      "mean_token_accuracy": 0.21063496619462968,
      "num_tokens": 36089010.0,
      "step": 19615
    },
    {
      "entropy": 5.075570583343506,
      "epoch": 1.6388239224858,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047323829192143436,
      "loss": 4.8886,
      "mean_token_accuracy": 0.21832526475191116,
      "num_tokens": 36097665.0,
      "step": 19620
    },
    {
      "entropy": 5.138750219345093,
      "epoch": 1.6392415636485131,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047322420880103566,
      "loss": 4.8076,
      "mean_token_accuracy": 0.21746742725372314,
      "num_tokens": 36107122.0,
      "step": 19625
    },
    {
      "entropy": 5.07195930480957,
      "epoch": 1.6396592048112262,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004732101222104836,
      "loss": 4.855,
      "mean_token_accuracy": 0.21523759961128236,
      "num_tokens": 36116549.0,
      "step": 19630
    },
    {
      "entropy": 5.070925664901734,
      "epoch": 1.6400768459739392,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004731960321500247,
      "loss": 4.7996,
      "mean_token_accuracy": 0.21466871798038484,
      "num_tokens": 36126506.0,
      "step": 19635
    },
    {
      "entropy": 5.150140190124512,
      "epoch": 1.6404944871366522,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004731819386199058,
      "loss": 4.9548,
      "mean_token_accuracy": 0.20906873196363449,
      "num_tokens": 36135638.0,
      "step": 19640
    },
    {
      "entropy": 5.162244367599487,
      "epoch": 1.640912128299365,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004731678416203734,
      "loss": 4.8508,
      "mean_token_accuracy": 0.21958714723587036,
      "num_tokens": 36143995.0,
      "step": 19645
    },
    {
      "entropy": 5.157527303695678,
      "epoch": 1.6413297694620783,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047315374115167446,
      "loss": 4.8638,
      "mean_token_accuracy": 0.21219314336776735,
      "num_tokens": 36152785.0,
      "step": 19650
    },
    {
      "entropy": 5.105141115188599,
      "epoch": 1.6417474106247911,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004731396372140558,
      "loss": 4.8815,
      "mean_token_accuracy": 0.20683688670396805,
      "num_tokens": 36162166.0,
      "step": 19655
    },
    {
      "entropy": 5.243606948852539,
      "epoch": 1.6421650517875042,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004731255298077642,
      "loss": 4.9777,
      "mean_token_accuracy": 0.19972773492336274,
      "num_tokens": 36172393.0,
      "step": 19660
    },
    {
      "entropy": 5.2621817111969,
      "epoch": 1.6425826929502172,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047311141893304675,
      "loss": 4.9498,
      "mean_token_accuracy": 0.21044939160346984,
      "num_tokens": 36182569.0,
      "step": 19665
    },
    {
      "entropy": 5.033015441894531,
      "epoch": 1.64300033411293,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004730973045901505,
      "loss": 4.7745,
      "mean_token_accuracy": 0.21883726567029954,
      "num_tokens": 36191516.0,
      "step": 19670
    },
    {
      "entropy": 5.054647636413574,
      "epoch": 1.6434179752756433,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047308318677932243,
      "loss": 4.8543,
      "mean_token_accuracy": 0.2120754227042198,
      "num_tokens": 36200964.0,
      "step": 19675
    },
    {
      "entropy": 5.0994524478912355,
      "epoch": 1.643835616438356,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004730690655008097,
      "loss": 4.857,
      "mean_token_accuracy": 0.2106857940554619,
      "num_tokens": 36210912.0,
      "step": 19680
    },
    {
      "entropy": 5.105824947357178,
      "epoch": 1.644253257601069,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004730549407548597,
      "loss": 4.8876,
      "mean_token_accuracy": 0.21016015112400055,
      "num_tokens": 36220197.0,
      "step": 19685
    },
    {
      "entropy": 5.171480417251587,
      "epoch": 1.6446708987637821,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047304081254171944,
      "loss": 4.8764,
      "mean_token_accuracy": 0.21089292466640472,
      "num_tokens": 36230191.0,
      "step": 19690
    },
    {
      "entropy": 5.097531127929687,
      "epoch": 1.6450885399264952,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004730266808616364,
      "loss": 4.8292,
      "mean_token_accuracy": 0.21005280166864396,
      "num_tokens": 36239247.0,
      "step": 19695
    },
    {
      "entropy": 5.162014532089233,
      "epoch": 1.6455061810892082,
      "grad_norm": 1.15625,
      "learning_rate": 0.000473012545714858,
      "loss": 4.9461,
      "mean_token_accuracy": 0.20870419591665268,
      "num_tokens": 36249374.0,
      "step": 19700
    },
    {
      "entropy": 5.216449022293091,
      "epoch": 1.645923822251921,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047299840710163157,
      "loss": 4.8885,
      "mean_token_accuracy": 0.21000660210847855,
      "num_tokens": 36259194.0,
      "step": 19705
    },
    {
      "entropy": 5.1514355659484865,
      "epoch": 1.6463414634146343,
      "grad_norm": 1.25,
      "learning_rate": 0.0004729842650222047,
      "loss": 4.8546,
      "mean_token_accuracy": 0.2162085235118866,
      "num_tokens": 36268388.0,
      "step": 19710
    },
    {
      "entropy": 5.10872712135315,
      "epoch": 1.646759104577347,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047297011947682494,
      "loss": 4.9321,
      "mean_token_accuracy": 0.20968770682811738,
      "num_tokens": 36277749.0,
      "step": 19715
    },
    {
      "entropy": 5.108411502838135,
      "epoch": 1.6471767457400601,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047295597046574,
      "loss": 4.8885,
      "mean_token_accuracy": 0.2134990155696869,
      "num_tokens": 36287471.0,
      "step": 19720
    },
    {
      "entropy": 5.156692314147949,
      "epoch": 1.6475943869027732,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004729418179891975,
      "loss": 4.8681,
      "mean_token_accuracy": 0.21176272630691528,
      "num_tokens": 36295860.0,
      "step": 19725
    },
    {
      "entropy": 5.159810543060303,
      "epoch": 1.6480120280654862,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004729276620474452,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2147675037384033,
      "num_tokens": 36304932.0,
      "step": 19730
    },
    {
      "entropy": 5.142549848556518,
      "epoch": 1.6484296692281992,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047291350264073095,
      "loss": 4.8875,
      "mean_token_accuracy": 0.21385958045721054,
      "num_tokens": 36314436.0,
      "step": 19735
    },
    {
      "entropy": 5.2572160243988035,
      "epoch": 1.648847310390912,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047289933976930263,
      "loss": 4.9544,
      "mean_token_accuracy": 0.20153694450855256,
      "num_tokens": 36323476.0,
      "step": 19740
    },
    {
      "entropy": 5.128851985931396,
      "epoch": 1.6492649515536253,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004728851734334081,
      "loss": 4.7988,
      "mean_token_accuracy": 0.21465957909822464,
      "num_tokens": 36332422.0,
      "step": 19745
    },
    {
      "entropy": 5.188562393188477,
      "epoch": 1.649682592716338,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004728710036332954,
      "loss": 4.9278,
      "mean_token_accuracy": 0.20843924582004547,
      "num_tokens": 36341945.0,
      "step": 19750
    },
    {
      "entropy": 5.171015024185181,
      "epoch": 1.6501002338790511,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047285683036921275,
      "loss": 4.9789,
      "mean_token_accuracy": 0.2098074212670326,
      "num_tokens": 36350587.0,
      "step": 19755
    },
    {
      "entropy": 5.148541641235352,
      "epoch": 1.6505178750417642,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047284265364140794,
      "loss": 4.917,
      "mean_token_accuracy": 0.21383593380451202,
      "num_tokens": 36359445.0,
      "step": 19760
    },
    {
      "entropy": 5.121838903427124,
      "epoch": 1.650935516204477,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004728284734501295,
      "loss": 4.8359,
      "mean_token_accuracy": 0.21586498767137527,
      "num_tokens": 36368338.0,
      "step": 19765
    },
    {
      "entropy": 5.183434343338012,
      "epoch": 1.6513531573671902,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004728142897956253,
      "loss": 4.8924,
      "mean_token_accuracy": 0.20827788710594178,
      "num_tokens": 36376995.0,
      "step": 19770
    },
    {
      "entropy": 5.078594636917114,
      "epoch": 1.651770798529903,
      "grad_norm": 1.140625,
      "learning_rate": 0.000472800102678144,
      "loss": 4.7263,
      "mean_token_accuracy": 0.21706456542015076,
      "num_tokens": 36386710.0,
      "step": 19775
    },
    {
      "entropy": 5.098106718063354,
      "epoch": 1.652188439692616,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047278591209793384,
      "loss": 4.9455,
      "mean_token_accuracy": 0.20136729627847672,
      "num_tokens": 36395634.0,
      "step": 19780
    },
    {
      "entropy": 5.130293130874634,
      "epoch": 1.6526060808553291,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004727717180552431,
      "loss": 4.9408,
      "mean_token_accuracy": 0.20811755806207657,
      "num_tokens": 36404918.0,
      "step": 19785
    },
    {
      "entropy": 5.2841236114501955,
      "epoch": 1.6530237220180422,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047275752055032043,
      "loss": 5.0052,
      "mean_token_accuracy": 0.20249519497156143,
      "num_tokens": 36414801.0,
      "step": 19790
    },
    {
      "entropy": 5.158917665481567,
      "epoch": 1.6534413631807552,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047274331958341434,
      "loss": 4.7974,
      "mean_token_accuracy": 0.21956181526184082,
      "num_tokens": 36422908.0,
      "step": 19795
    },
    {
      "entropy": 5.11718602180481,
      "epoch": 1.653859004343468,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004727291151547733,
      "loss": 4.9031,
      "mean_token_accuracy": 0.21400924921035766,
      "num_tokens": 36432137.0,
      "step": 19800
    },
    {
      "entropy": 5.155436563491821,
      "epoch": 1.6542766455061813,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004727149072646463,
      "loss": 4.8729,
      "mean_token_accuracy": 0.2138337880373001,
      "num_tokens": 36440296.0,
      "step": 19805
    },
    {
      "entropy": 5.11661376953125,
      "epoch": 1.654694286668894,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047270069591328175,
      "loss": 4.8501,
      "mean_token_accuracy": 0.21352879256010054,
      "num_tokens": 36449323.0,
      "step": 19810
    },
    {
      "entropy": 5.004731273651123,
      "epoch": 1.655111927831607,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047268648110092857,
      "loss": 4.821,
      "mean_token_accuracy": 0.21611949056386948,
      "num_tokens": 36458182.0,
      "step": 19815
    },
    {
      "entropy": 5.265736818313599,
      "epoch": 1.6555295689943201,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047267226282783555,
      "loss": 5.118,
      "mean_token_accuracy": 0.19208419770002366,
      "num_tokens": 36468276.0,
      "step": 19820
    },
    {
      "entropy": 5.152170133590698,
      "epoch": 1.655947210157033,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004726580410942516,
      "loss": 4.8465,
      "mean_token_accuracy": 0.20755318254232408,
      "num_tokens": 36477202.0,
      "step": 19825
    },
    {
      "entropy": 5.1689427375793455,
      "epoch": 1.6563648513197462,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047264381590042575,
      "loss": 5.0132,
      "mean_token_accuracy": 0.19938839823007584,
      "num_tokens": 36486374.0,
      "step": 19830
    },
    {
      "entropy": 5.196919298171997,
      "epoch": 1.656782492482459,
      "grad_norm": 1.296875,
      "learning_rate": 0.000472629587246607,
      "loss": 4.9091,
      "mean_token_accuracy": 0.20822001844644547,
      "num_tokens": 36496167.0,
      "step": 19835
    },
    {
      "entropy": 5.161489772796631,
      "epoch": 1.657200133645172,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047261535513304446,
      "loss": 4.9714,
      "mean_token_accuracy": 0.204331174492836,
      "num_tokens": 36506410.0,
      "step": 19840
    },
    {
      "entropy": 5.1877443313598635,
      "epoch": 1.657617774807885,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047260111955998725,
      "loss": 4.8419,
      "mean_token_accuracy": 0.21628785282373428,
      "num_tokens": 36515511.0,
      "step": 19845
    },
    {
      "entropy": 5.089254140853882,
      "epoch": 1.6580354159705981,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047258688052768456,
      "loss": 4.7918,
      "mean_token_accuracy": 0.21884079426527023,
      "num_tokens": 36524639.0,
      "step": 19850
    },
    {
      "entropy": 5.012351703643799,
      "epoch": 1.6584530571333111,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004725726380363857,
      "loss": 4.8052,
      "mean_token_accuracy": 0.21364715546369553,
      "num_tokens": 36534314.0,
      "step": 19855
    },
    {
      "entropy": 5.098216247558594,
      "epoch": 1.658870698296024,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047255839208634005,
      "loss": 4.8372,
      "mean_token_accuracy": 0.21783955693244933,
      "num_tokens": 36543469.0,
      "step": 19860
    },
    {
      "entropy": 5.234014987945557,
      "epoch": 1.6592883394587372,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047254414267779686,
      "loss": 5.0292,
      "mean_token_accuracy": 0.20542742013931276,
      "num_tokens": 36552690.0,
      "step": 19865
    },
    {
      "entropy": 5.200622701644898,
      "epoch": 1.65970598062145,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047252988981100576,
      "loss": 4.9828,
      "mean_token_accuracy": 0.1996054545044899,
      "num_tokens": 36562679.0,
      "step": 19870
    },
    {
      "entropy": 5.130401945114135,
      "epoch": 1.660123621784163,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004725156334862161,
      "loss": 4.8089,
      "mean_token_accuracy": 0.20997802466154097,
      "num_tokens": 36571564.0,
      "step": 19875
    },
    {
      "entropy": 5.203746032714844,
      "epoch": 1.660541262946876,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047250137370367755,
      "loss": 4.9342,
      "mean_token_accuracy": 0.20594387948513032,
      "num_tokens": 36581022.0,
      "step": 19880
    },
    {
      "entropy": 5.09020471572876,
      "epoch": 1.660958904109589,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047248711046363976,
      "loss": 4.8537,
      "mean_token_accuracy": 0.21990229487419127,
      "num_tokens": 36590068.0,
      "step": 19885
    },
    {
      "entropy": 5.140152788162231,
      "epoch": 1.6613765452723022,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004724728437663523,
      "loss": 4.8875,
      "mean_token_accuracy": 0.2103304773569107,
      "num_tokens": 36600019.0,
      "step": 19890
    },
    {
      "entropy": 5.193677997589111,
      "epoch": 1.661794186435015,
      "grad_norm": 1.25,
      "learning_rate": 0.00047245857361206506,
      "loss": 4.9381,
      "mean_token_accuracy": 0.20528475493192672,
      "num_tokens": 36608850.0,
      "step": 19895
    },
    {
      "entropy": 5.052873802185059,
      "epoch": 1.662211827597728,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047244430000102785,
      "loss": 4.7453,
      "mean_token_accuracy": 0.21719938218593599,
      "num_tokens": 36616847.0,
      "step": 19900
    },
    {
      "entropy": 5.084103298187256,
      "epoch": 1.662629468760441,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004724300229334905,
      "loss": 4.8223,
      "mean_token_accuracy": 0.214922758936882,
      "num_tokens": 36625717.0,
      "step": 19905
    },
    {
      "entropy": 5.124482583999634,
      "epoch": 1.663047109923154,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047241574240970284,
      "loss": 5.0069,
      "mean_token_accuracy": 0.19575542956590652,
      "num_tokens": 36636152.0,
      "step": 19910
    },
    {
      "entropy": 5.233613204956055,
      "epoch": 1.663464751085867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047240145842991513,
      "loss": 4.8766,
      "mean_token_accuracy": 0.20924119651317596,
      "num_tokens": 36645545.0,
      "step": 19915
    },
    {
      "entropy": 5.092205286026001,
      "epoch": 1.66388239224858,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047238717099437713,
      "loss": 4.8513,
      "mean_token_accuracy": 0.21021052002906798,
      "num_tokens": 36654707.0,
      "step": 19920
    },
    {
      "entropy": 5.183172512054443,
      "epoch": 1.6643000334112932,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004723728801033392,
      "loss": 4.9537,
      "mean_token_accuracy": 0.2042350172996521,
      "num_tokens": 36664005.0,
      "step": 19925
    },
    {
      "entropy": 5.104406785964966,
      "epoch": 1.664717674574006,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004723585857570514,
      "loss": 4.8607,
      "mean_token_accuracy": 0.21352100372314453,
      "num_tokens": 36674584.0,
      "step": 19930
    },
    {
      "entropy": 5.155652046203613,
      "epoch": 1.665135315736719,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000472344287955764,
      "loss": 4.9206,
      "mean_token_accuracy": 0.2139970615506172,
      "num_tokens": 36684832.0,
      "step": 19935
    },
    {
      "entropy": 5.107335090637207,
      "epoch": 1.665552956899432,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004723299866997273,
      "loss": 4.8287,
      "mean_token_accuracy": 0.21275645792484282,
      "num_tokens": 36694368.0,
      "step": 19940
    },
    {
      "entropy": 5.149318981170654,
      "epoch": 1.6659705980621449,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004723156819891917,
      "loss": 4.9326,
      "mean_token_accuracy": 0.20979706346988677,
      "num_tokens": 36704351.0,
      "step": 19945
    },
    {
      "entropy": 5.101915788650513,
      "epoch": 1.6663882392248581,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047230137382440747,
      "loss": 4.8293,
      "mean_token_accuracy": 0.20860149711370468,
      "num_tokens": 36714689.0,
      "step": 19950
    },
    {
      "entropy": 5.182568788528442,
      "epoch": 1.666805880387571,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004722870622056253,
      "loss": 4.8825,
      "mean_token_accuracy": 0.22084220945835115,
      "num_tokens": 36723020.0,
      "step": 19955
    },
    {
      "entropy": 5.152387046813965,
      "epoch": 1.667223521550284,
      "grad_norm": 1.125,
      "learning_rate": 0.0004722727471330956,
      "loss": 4.8764,
      "mean_token_accuracy": 0.21376501768827438,
      "num_tokens": 36732109.0,
      "step": 19960
    },
    {
      "entropy": 5.087404203414917,
      "epoch": 1.667641162712997,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000472258428607069,
      "loss": 4.924,
      "mean_token_accuracy": 0.20686960965394974,
      "num_tokens": 36742080.0,
      "step": 19965
    },
    {
      "entropy": 5.117727327346802,
      "epoch": 1.66805880387571,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004722441066277961,
      "loss": 4.9389,
      "mean_token_accuracy": 0.20222727358341216,
      "num_tokens": 36751288.0,
      "step": 19970
    },
    {
      "entropy": 5.166043663024903,
      "epoch": 1.668476445038423,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047222978119552776,
      "loss": 4.9104,
      "mean_token_accuracy": 0.2082187280058861,
      "num_tokens": 36761378.0,
      "step": 19975
    },
    {
      "entropy": 5.1745133876800535,
      "epoch": 1.6688940862011359,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047221545231051473,
      "loss": 4.928,
      "mean_token_accuracy": 0.20582748502492904,
      "num_tokens": 36770947.0,
      "step": 19980
    },
    {
      "entropy": 5.135239219665527,
      "epoch": 1.6693117273638491,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004722011199730078,
      "loss": 4.8289,
      "mean_token_accuracy": 0.21191948056221008,
      "num_tokens": 36780205.0,
      "step": 19985
    },
    {
      "entropy": 5.05585412979126,
      "epoch": 1.669729368526562,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004721867841832579,
      "loss": 4.7199,
      "mean_token_accuracy": 0.2227197378873825,
      "num_tokens": 36789167.0,
      "step": 19990
    },
    {
      "entropy": 5.009308815002441,
      "epoch": 1.670147009689275,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004721724449415159,
      "loss": 4.7391,
      "mean_token_accuracy": 0.2230490818619728,
      "num_tokens": 36798257.0,
      "step": 19995
    },
    {
      "entropy": 5.125503635406494,
      "epoch": 1.670564650851988,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047215810224803304,
      "loss": 4.8488,
      "mean_token_accuracy": 0.21522542238235473,
      "num_tokens": 36808746.0,
      "step": 20000
    },
    {
      "entropy": 5.163046598434448,
      "epoch": 1.6709822920147008,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047214375610306016,
      "loss": 4.8737,
      "mean_token_accuracy": 0.20951217114925386,
      "num_tokens": 36817572.0,
      "step": 20005
    },
    {
      "entropy": 5.101292705535888,
      "epoch": 1.671399933177414,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047212940650684853,
      "loss": 4.8958,
      "mean_token_accuracy": 0.2120031550526619,
      "num_tokens": 36826678.0,
      "step": 20010
    },
    {
      "entropy": 5.155959367752075,
      "epoch": 1.671817574340127,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004721150534596495,
      "loss": 4.8647,
      "mean_token_accuracy": 0.21460782140493392,
      "num_tokens": 36835862.0,
      "step": 20015
    },
    {
      "entropy": 5.071805095672607,
      "epoch": 1.67223521550284,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000472100696961714,
      "loss": 4.7714,
      "mean_token_accuracy": 0.2207110583782196,
      "num_tokens": 36845461.0,
      "step": 20020
    },
    {
      "entropy": 5.067549753189087,
      "epoch": 1.672652856665553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004720863370132937,
      "loss": 4.8564,
      "mean_token_accuracy": 0.20869653522968293,
      "num_tokens": 36855200.0,
      "step": 20025
    },
    {
      "entropy": 5.1826413631439205,
      "epoch": 1.673070497828266,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004720719736146398,
      "loss": 4.8708,
      "mean_token_accuracy": 0.2054842084646225,
      "num_tokens": 36864221.0,
      "step": 20030
    },
    {
      "entropy": 5.197817134857178,
      "epoch": 1.673488138990979,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047205760676600377,
      "loss": 4.9424,
      "mean_token_accuracy": 0.20943801999092101,
      "num_tokens": 36873698.0,
      "step": 20035
    },
    {
      "entropy": 5.136811923980713,
      "epoch": 1.6739057801536918,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004720432364676372,
      "loss": 4.8187,
      "mean_token_accuracy": 0.2193689987063408,
      "num_tokens": 36882026.0,
      "step": 20040
    },
    {
      "entropy": 5.120841455459595,
      "epoch": 1.674323421316405,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004720288627197915,
      "loss": 4.7294,
      "mean_token_accuracy": 0.2276133641600609,
      "num_tokens": 36890948.0,
      "step": 20045
    },
    {
      "entropy": 5.179144811630249,
      "epoch": 1.674741062479118,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004720144855227185,
      "loss": 4.9493,
      "mean_token_accuracy": 0.20247936397790908,
      "num_tokens": 36901635.0,
      "step": 20050
    },
    {
      "entropy": 5.111210536956787,
      "epoch": 1.675158703641831,
      "grad_norm": 1.25,
      "learning_rate": 0.0004720001048766696,
      "loss": 4.8245,
      "mean_token_accuracy": 0.21652900129556657,
      "num_tokens": 36910275.0,
      "step": 20055
    },
    {
      "entropy": 5.194299507141113,
      "epoch": 1.675576344804544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004719857207818969,
      "loss": 4.9613,
      "mean_token_accuracy": 0.20642045587301255,
      "num_tokens": 36920243.0,
      "step": 20060
    },
    {
      "entropy": 5.09620943069458,
      "epoch": 1.6759939859672568,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000471971333238652,
      "loss": 4.7991,
      "mean_token_accuracy": 0.21976744681596755,
      "num_tokens": 36931008.0,
      "step": 20065
    },
    {
      "entropy": 5.09450855255127,
      "epoch": 1.67641162712997,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047195694224718684,
      "loss": 4.8128,
      "mean_token_accuracy": 0.22534408420324326,
      "num_tokens": 36939936.0,
      "step": 20070
    },
    {
      "entropy": 5.1108006000518795,
      "epoch": 1.6768292682926829,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047194254780775325,
      "loss": 4.8531,
      "mean_token_accuracy": 0.20692527890205384,
      "num_tokens": 36948936.0,
      "step": 20075
    },
    {
      "entropy": 5.1115576267242435,
      "epoch": 1.677246909455396,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004719281499206034,
      "loss": 4.7853,
      "mean_token_accuracy": 0.2175554722547531,
      "num_tokens": 36958362.0,
      "step": 20080
    },
    {
      "entropy": 5.0141278266906735,
      "epoch": 1.677664550618109,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047191374858598917,
      "loss": 4.7789,
      "mean_token_accuracy": 0.22281930297613145,
      "num_tokens": 36967744.0,
      "step": 20085
    },
    {
      "entropy": 5.096109247207641,
      "epoch": 1.678082191780822,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004718993438041628,
      "loss": 4.9605,
      "mean_token_accuracy": 0.20877107679843904,
      "num_tokens": 36976808.0,
      "step": 20090
    },
    {
      "entropy": 5.137476778030395,
      "epoch": 1.678499832943535,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047188493557537634,
      "loss": 4.8117,
      "mean_token_accuracy": 0.22031086534261704,
      "num_tokens": 36985819.0,
      "step": 20095
    },
    {
      "entropy": 5.174354553222656,
      "epoch": 1.6789174741062478,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004718705238998822,
      "loss": 4.834,
      "mean_token_accuracy": 0.21654441356658935,
      "num_tokens": 36994044.0,
      "step": 20100
    },
    {
      "entropy": 5.051553010940552,
      "epoch": 1.679335115268961,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047185610877793245,
      "loss": 4.7845,
      "mean_token_accuracy": 0.21984178125858306,
      "num_tokens": 37002168.0,
      "step": 20105
    },
    {
      "entropy": 5.101382112503051,
      "epoch": 1.6797527564316739,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047184169020977965,
      "loss": 4.8958,
      "mean_token_accuracy": 0.21296921819448472,
      "num_tokens": 37011708.0,
      "step": 20110
    },
    {
      "entropy": 5.198537158966064,
      "epoch": 1.680170397594387,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047182726819567603,
      "loss": 4.929,
      "mean_token_accuracy": 0.2065126493573189,
      "num_tokens": 37020436.0,
      "step": 20115
    },
    {
      "entropy": 5.117106294631958,
      "epoch": 1.6805880387571,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047181284273587414,
      "loss": 4.8478,
      "mean_token_accuracy": 0.21432785391807557,
      "num_tokens": 37029885.0,
      "step": 20120
    },
    {
      "entropy": 5.0767748832702635,
      "epoch": 1.6810056799198128,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047179841383062667,
      "loss": 4.9047,
      "mean_token_accuracy": 0.21147147715091705,
      "num_tokens": 37038818.0,
      "step": 20125
    },
    {
      "entropy": 5.124965476989746,
      "epoch": 1.681423321082526,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047178398148018606,
      "loss": 4.8642,
      "mean_token_accuracy": 0.2126622274518013,
      "num_tokens": 37047382.0,
      "step": 20130
    },
    {
      "entropy": 5.156989526748657,
      "epoch": 1.6818409622452388,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004717695456848049,
      "loss": 4.8733,
      "mean_token_accuracy": 0.218117293715477,
      "num_tokens": 37056354.0,
      "step": 20135
    },
    {
      "entropy": 5.124832487106323,
      "epoch": 1.6822586034079519,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004717551064447361,
      "loss": 4.8853,
      "mean_token_accuracy": 0.21497889757156372,
      "num_tokens": 37064954.0,
      "step": 20140
    },
    {
      "entropy": 5.085470819473267,
      "epoch": 1.682676244570665,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004717406637602322,
      "loss": 4.8228,
      "mean_token_accuracy": 0.212307371199131,
      "num_tokens": 37074199.0,
      "step": 20145
    },
    {
      "entropy": 5.1593132495880125,
      "epoch": 1.683093885733378,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004717262176315462,
      "loss": 4.8888,
      "mean_token_accuracy": 0.21403444856405257,
      "num_tokens": 37083034.0,
      "step": 20150
    },
    {
      "entropy": 5.176192569732666,
      "epoch": 1.683511526896091,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000471711768058931,
      "loss": 4.8941,
      "mean_token_accuracy": 0.21241376250982286,
      "num_tokens": 37091725.0,
      "step": 20155
    },
    {
      "entropy": 5.13080849647522,
      "epoch": 1.6839291680588038,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004716973150426395,
      "loss": 4.856,
      "mean_token_accuracy": 0.21844005435705185,
      "num_tokens": 37100346.0,
      "step": 20160
    },
    {
      "entropy": 5.1210113048553465,
      "epoch": 1.684346809221517,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047168285858292477,
      "loss": 4.8781,
      "mean_token_accuracy": 0.2106308162212372,
      "num_tokens": 37108914.0,
      "step": 20165
    },
    {
      "entropy": 5.110776567459107,
      "epoch": 1.6847644503842298,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047166839868003984,
      "loss": 4.8208,
      "mean_token_accuracy": 0.21673003286123277,
      "num_tokens": 37118109.0,
      "step": 20170
    },
    {
      "entropy": 5.149813556671143,
      "epoch": 1.6851820915469429,
      "grad_norm": 1.125,
      "learning_rate": 0.00047165393533423784,
      "loss": 4.8929,
      "mean_token_accuracy": 0.2140464335680008,
      "num_tokens": 37128602.0,
      "step": 20175
    },
    {
      "entropy": 5.135681533813477,
      "epoch": 1.685599732709656,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047163946854577194,
      "loss": 4.9175,
      "mean_token_accuracy": 0.2092985451221466,
      "num_tokens": 37137843.0,
      "step": 20180
    },
    {
      "entropy": 5.1340337753295895,
      "epoch": 1.6860173738723687,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004716249983148955,
      "loss": 4.8437,
      "mean_token_accuracy": 0.21398521065711976,
      "num_tokens": 37147116.0,
      "step": 20185
    },
    {
      "entropy": 5.207225894927978,
      "epoch": 1.686435015035082,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004716105246418618,
      "loss": 5.0072,
      "mean_token_accuracy": 0.20219030231237411,
      "num_tokens": 37156333.0,
      "step": 20190
    },
    {
      "entropy": 5.042727708816528,
      "epoch": 1.6868526561977948,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004715960475269242,
      "loss": 4.7934,
      "mean_token_accuracy": 0.21005075573921203,
      "num_tokens": 37164636.0,
      "step": 20195
    },
    {
      "entropy": 5.192473840713501,
      "epoch": 1.6872702973605078,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004715815669703362,
      "loss": 5.0085,
      "mean_token_accuracy": 0.19980120956897734,
      "num_tokens": 37174877.0,
      "step": 20200
    },
    {
      "entropy": 5.230113554000854,
      "epoch": 1.6876879385232209,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047156708297235115,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20450180768966675,
      "num_tokens": 37184959.0,
      "step": 20205
    },
    {
      "entropy": 5.266556882858277,
      "epoch": 1.6881055796859339,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004715525955332227,
      "loss": 4.9778,
      "mean_token_accuracy": 0.20502413511276246,
      "num_tokens": 37194565.0,
      "step": 20210
    },
    {
      "entropy": 5.183176946640015,
      "epoch": 1.688523220848647,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004715381046532046,
      "loss": 4.8764,
      "mean_token_accuracy": 0.21087196916341783,
      "num_tokens": 37203302.0,
      "step": 20215
    },
    {
      "entropy": 5.05908784866333,
      "epoch": 1.6889408620113597,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004715236103325503,
      "loss": 4.7772,
      "mean_token_accuracy": 0.21132662296295165,
      "num_tokens": 37212808.0,
      "step": 20220
    },
    {
      "entropy": 5.069698190689087,
      "epoch": 1.689358503174073,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047150911257151364,
      "loss": 4.864,
      "mean_token_accuracy": 0.21755434572696686,
      "num_tokens": 37221944.0,
      "step": 20225
    },
    {
      "entropy": 5.17473292350769,
      "epoch": 1.6897761443367858,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004714946113703484,
      "loss": 4.8797,
      "mean_token_accuracy": 0.21529098004102706,
      "num_tokens": 37231147.0,
      "step": 20230
    },
    {
      "entropy": 5.229843091964722,
      "epoch": 1.6901937854994988,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004714801067293086,
      "loss": 5.0733,
      "mean_token_accuracy": 0.19731041193008422,
      "num_tokens": 37240951.0,
      "step": 20235
    },
    {
      "entropy": 5.120366334915161,
      "epoch": 1.6906114266622119,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047146559864864787,
      "loss": 4.8175,
      "mean_token_accuracy": 0.21417984515428543,
      "num_tokens": 37249333.0,
      "step": 20240
    },
    {
      "entropy": 5.141189956665039,
      "epoch": 1.6910290678249247,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047145108712862046,
      "loss": 4.897,
      "mean_token_accuracy": 0.21205742359161378,
      "num_tokens": 37259323.0,
      "step": 20245
    },
    {
      "entropy": 5.168588733673095,
      "epoch": 1.691446708987638,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004714365721694803,
      "loss": 4.8837,
      "mean_token_accuracy": 0.21272594928741456,
      "num_tokens": 37269434.0,
      "step": 20250
    },
    {
      "entropy": 5.156182813644409,
      "epoch": 1.6918643501503507,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047142205377148145,
      "loss": 4.9085,
      "mean_token_accuracy": 0.21393272131681443,
      "num_tokens": 37278721.0,
      "step": 20255
    },
    {
      "entropy": 5.1725856304168705,
      "epoch": 1.6922819913130638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004714075319348781,
      "loss": 4.8168,
      "mean_token_accuracy": 0.22423159182071686,
      "num_tokens": 37287519.0,
      "step": 20260
    },
    {
      "entropy": 5.070461082458496,
      "epoch": 1.6926996324757768,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004713930066599246,
      "loss": 4.8615,
      "mean_token_accuracy": 0.2085599571466446,
      "num_tokens": 37295896.0,
      "step": 20265
    },
    {
      "entropy": 5.133475065231323,
      "epoch": 1.6931172736384899,
      "grad_norm": 1.171875,
      "learning_rate": 0.000471378477946875,
      "loss": 4.863,
      "mean_token_accuracy": 0.2051178589463234,
      "num_tokens": 37307255.0,
      "step": 20270
    },
    {
      "entropy": 5.101776361465454,
      "epoch": 1.6935349148012029,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004713639457959838,
      "loss": 4.8172,
      "mean_token_accuracy": 0.217009474337101,
      "num_tokens": 37316765.0,
      "step": 20275
    },
    {
      "entropy": 5.088513612747192,
      "epoch": 1.6939525559639157,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004713494102075053,
      "loss": 4.8773,
      "mean_token_accuracy": 0.20300128161907197,
      "num_tokens": 37326094.0,
      "step": 20280
    },
    {
      "entropy": 5.191268634796143,
      "epoch": 1.694370197126629,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004713348711816942,
      "loss": 5.0177,
      "mean_token_accuracy": 0.19609731137752534,
      "num_tokens": 37336639.0,
      "step": 20285
    },
    {
      "entropy": 5.072662830352783,
      "epoch": 1.6947878382893418,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004713203287188047,
      "loss": 4.747,
      "mean_token_accuracy": 0.22517246305942534,
      "num_tokens": 37345772.0,
      "step": 20290
    },
    {
      "entropy": 5.256340169906617,
      "epoch": 1.6952054794520548,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004713057828190916,
      "loss": 5.0791,
      "mean_token_accuracy": 0.19290219694375993,
      "num_tokens": 37356708.0,
      "step": 20295
    },
    {
      "entropy": 5.123903894424439,
      "epoch": 1.6956231206147678,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004712912334828094,
      "loss": 4.8469,
      "mean_token_accuracy": 0.20537903904914856,
      "num_tokens": 37365391.0,
      "step": 20300
    },
    {
      "entropy": 5.103719186782837,
      "epoch": 1.6960407617774806,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047127668071021293,
      "loss": 4.8826,
      "mean_token_accuracy": 0.21028376817703248,
      "num_tokens": 37374038.0,
      "step": 20305
    },
    {
      "entropy": 5.271544933319092,
      "epoch": 1.696458402940194,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000471262124501557,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20253566205501555,
      "num_tokens": 37382765.0,
      "step": 20310
    },
    {
      "entropy": 5.13502459526062,
      "epoch": 1.6968760441029067,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004712475648570962,
      "loss": 4.7035,
      "mean_token_accuracy": 0.2148857057094574,
      "num_tokens": 37390870.0,
      "step": 20315
    },
    {
      "entropy": 5.086621475219727,
      "epoch": 1.6972936852656197,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047123300177708563,
      "loss": 4.8862,
      "mean_token_accuracy": 0.21263102144002916,
      "num_tokens": 37399422.0,
      "step": 20320
    },
    {
      "entropy": 5.035601949691772,
      "epoch": 1.6977113264283328,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004712184352617801,
      "loss": 4.854,
      "mean_token_accuracy": 0.2173978567123413,
      "num_tokens": 37408500.0,
      "step": 20325
    },
    {
      "entropy": 5.130883884429932,
      "epoch": 1.6981289675910458,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047120386531143476,
      "loss": 4.8746,
      "mean_token_accuracy": 0.21200544834136964,
      "num_tokens": 37417387.0,
      "step": 20330
    },
    {
      "entropy": 5.153274774551392,
      "epoch": 1.6985466087537588,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047118929192630453,
      "loss": 4.8099,
      "mean_token_accuracy": 0.21337425857782363,
      "num_tokens": 37425793.0,
      "step": 20335
    },
    {
      "entropy": 5.112049341201782,
      "epoch": 1.6989642499164717,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004711747151066446,
      "loss": 4.7221,
      "mean_token_accuracy": 0.22352167069911957,
      "num_tokens": 37434563.0,
      "step": 20340
    },
    {
      "entropy": 5.124646139144898,
      "epoch": 1.699381891079185,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047116013485271014,
      "loss": 4.9532,
      "mean_token_accuracy": 0.21163807660341263,
      "num_tokens": 37444265.0,
      "step": 20345
    },
    {
      "entropy": 5.134666109085083,
      "epoch": 1.6997995322418977,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004711455511647564,
      "loss": 4.9735,
      "mean_token_accuracy": 0.20662278234958648,
      "num_tokens": 37453727.0,
      "step": 20350
    },
    {
      "entropy": 5.140126037597656,
      "epoch": 1.7002171734046108,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047113096404303875,
      "loss": 4.891,
      "mean_token_accuracy": 0.20629487782716752,
      "num_tokens": 37463383.0,
      "step": 20355
    },
    {
      "entropy": 5.007933521270752,
      "epoch": 1.7006348145673238,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004711163734878124,
      "loss": 4.7152,
      "mean_token_accuracy": 0.22443329989910127,
      "num_tokens": 37472499.0,
      "step": 20360
    },
    {
      "entropy": 5.155351161956787,
      "epoch": 1.7010524557300366,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004711017794993329,
      "loss": 4.9477,
      "mean_token_accuracy": 0.20116439312696457,
      "num_tokens": 37482108.0,
      "step": 20365
    },
    {
      "entropy": 5.075755977630616,
      "epoch": 1.7014700968927499,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047108718207785575,
      "loss": 4.7613,
      "mean_token_accuracy": 0.21915675401687623,
      "num_tokens": 37491879.0,
      "step": 20370
    },
    {
      "entropy": 5.154397964477539,
      "epoch": 1.7018877380554627,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004710725812236364,
      "loss": 4.8636,
      "mean_token_accuracy": 0.2049018159508705,
      "num_tokens": 37501207.0,
      "step": 20375
    },
    {
      "entropy": 5.1455216884613035,
      "epoch": 1.7023053792181757,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004710579769369305,
      "loss": 4.8898,
      "mean_token_accuracy": 0.2106180816888809,
      "num_tokens": 37509988.0,
      "step": 20380
    },
    {
      "entropy": 5.163533496856689,
      "epoch": 1.7027230203808887,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004710433692179937,
      "loss": 4.9033,
      "mean_token_accuracy": 0.20523811876773834,
      "num_tokens": 37519013.0,
      "step": 20385
    },
    {
      "entropy": 5.061489534378052,
      "epoch": 1.7031406615436018,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004710287580670818,
      "loss": 4.7573,
      "mean_token_accuracy": 0.21430078148841858,
      "num_tokens": 37527874.0,
      "step": 20390
    },
    {
      "entropy": 5.04426760673523,
      "epoch": 1.7035583027063148,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047101414348445046,
      "loss": 4.8709,
      "mean_token_accuracy": 0.2173241049051285,
      "num_tokens": 37535825.0,
      "step": 20395
    },
    {
      "entropy": 5.18304181098938,
      "epoch": 1.7039759438690276,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047099952547035564,
      "loss": 4.86,
      "mean_token_accuracy": 0.2095210924744606,
      "num_tokens": 37544971.0,
      "step": 20400
    },
    {
      "entropy": 5.121310758590698,
      "epoch": 1.7043935850317409,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004709849040250532,
      "loss": 4.8997,
      "mean_token_accuracy": 0.20893124490976334,
      "num_tokens": 37554088.0,
      "step": 20405
    },
    {
      "entropy": 5.103823184967041,
      "epoch": 1.7048112261944537,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004709702791487991,
      "loss": 4.8899,
      "mean_token_accuracy": 0.21183300018310547,
      "num_tokens": 37563117.0,
      "step": 20410
    },
    {
      "entropy": 5.144977855682373,
      "epoch": 1.7052288673571667,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004709556508418494,
      "loss": 4.9129,
      "mean_token_accuracy": 0.208910471200943,
      "num_tokens": 37571729.0,
      "step": 20415
    },
    {
      "entropy": 5.180977392196655,
      "epoch": 1.7056465085198798,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004709410191044601,
      "loss": 4.8782,
      "mean_token_accuracy": 0.21576017588377,
      "num_tokens": 37580685.0,
      "step": 20420
    },
    {
      "entropy": 5.122219753265381,
      "epoch": 1.7060641496825926,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004709263839368875,
      "loss": 4.7778,
      "mean_token_accuracy": 0.2148606687784195,
      "num_tokens": 37590961.0,
      "step": 20425
    },
    {
      "entropy": 5.1323834419250485,
      "epoch": 1.7064817908453058,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004709117453393876,
      "loss": 4.8847,
      "mean_token_accuracy": 0.21641409993171692,
      "num_tokens": 37600997.0,
      "step": 20430
    },
    {
      "entropy": 5.032791709899902,
      "epoch": 1.7068994320080186,
      "grad_norm": 1.125,
      "learning_rate": 0.00047089710331221687,
      "loss": 4.8759,
      "mean_token_accuracy": 0.20457789152860642,
      "num_tokens": 37612003.0,
      "step": 20435
    },
    {
      "entropy": 5.198729228973389,
      "epoch": 1.7073170731707317,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004708824578556315,
      "loss": 4.8732,
      "mean_token_accuracy": 0.2095998466014862,
      "num_tokens": 37621735.0,
      "step": 20440
    },
    {
      "entropy": 5.153516960144043,
      "epoch": 1.7077347143334447,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047086780896988796,
      "loss": 4.8792,
      "mean_token_accuracy": 0.21560786813497543,
      "num_tokens": 37630325.0,
      "step": 20445
    },
    {
      "entropy": 5.063109636306763,
      "epoch": 1.7081523554961577,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047085315665524257,
      "loss": 4.7228,
      "mean_token_accuracy": 0.22819358557462693,
      "num_tokens": 37638618.0,
      "step": 20450
    },
    {
      "entropy": 5.017463541030883,
      "epoch": 1.7085699966588708,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000470838500911952,
      "loss": 4.8197,
      "mean_token_accuracy": 0.22387966215610505,
      "num_tokens": 37646691.0,
      "step": 20455
    },
    {
      "entropy": 5.178821229934693,
      "epoch": 1.7089876378215836,
      "grad_norm": 1.25,
      "learning_rate": 0.0004708238417402728,
      "loss": 4.8587,
      "mean_token_accuracy": 0.210019613802433,
      "num_tokens": 37656506.0,
      "step": 20460
    },
    {
      "entropy": 5.161155462265015,
      "epoch": 1.7094052789842968,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047080917914046146,
      "loss": 4.8384,
      "mean_token_accuracy": 0.2128883793950081,
      "num_tokens": 37665857.0,
      "step": 20465
    },
    {
      "entropy": 4.997651290893555,
      "epoch": 1.7098229201470097,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047079451311277473,
      "loss": 4.7242,
      "mean_token_accuracy": 0.22350741922855377,
      "num_tokens": 37675185.0,
      "step": 20470
    },
    {
      "entropy": 5.079292297363281,
      "epoch": 1.7102405613097227,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004707798436574694,
      "loss": 4.8856,
      "mean_token_accuracy": 0.21825852990150452,
      "num_tokens": 37685049.0,
      "step": 20475
    },
    {
      "entropy": 5.2474658489227295,
      "epoch": 1.7106582024724357,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047076517077480223,
      "loss": 4.9919,
      "mean_token_accuracy": 0.20002604126930237,
      "num_tokens": 37694919.0,
      "step": 20480
    },
    {
      "entropy": 5.190857458114624,
      "epoch": 1.7110758436351485,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004707504944650301,
      "loss": 4.9348,
      "mean_token_accuracy": 0.20437809079885483,
      "num_tokens": 37705445.0,
      "step": 20485
    },
    {
      "entropy": 5.132230472564697,
      "epoch": 1.7114934847978618,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047073581472840997,
      "loss": 4.7667,
      "mean_token_accuracy": 0.22259796261787415,
      "num_tokens": 37713878.0,
      "step": 20490
    },
    {
      "entropy": 5.121597719192505,
      "epoch": 1.7119111259605746,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004707211315651988,
      "loss": 4.8505,
      "mean_token_accuracy": 0.21490235924720763,
      "num_tokens": 37722166.0,
      "step": 20495
    },
    {
      "entropy": 5.138812637329101,
      "epoch": 1.7123287671232876,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004707064449756536,
      "loss": 4.8898,
      "mean_token_accuracy": 0.2147018551826477,
      "num_tokens": 37731620.0,
      "step": 20500
    },
    {
      "entropy": 5.075153779983521,
      "epoch": 1.7127464082860007,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047069175496003147,
      "loss": 4.8427,
      "mean_token_accuracy": 0.20657982379198075,
      "num_tokens": 37741384.0,
      "step": 20505
    },
    {
      "entropy": 5.119697856903076,
      "epoch": 1.7131640494487137,
      "grad_norm": 1.125,
      "learning_rate": 0.00047067706151858967,
      "loss": 4.8156,
      "mean_token_accuracy": 0.21558548808097838,
      "num_tokens": 37750488.0,
      "step": 20510
    },
    {
      "entropy": 5.1967723846435545,
      "epoch": 1.7135816906114267,
      "grad_norm": 1.25,
      "learning_rate": 0.0004706623646515854,
      "loss": 4.9106,
      "mean_token_accuracy": 0.20519382357597352,
      "num_tokens": 37759645.0,
      "step": 20515
    },
    {
      "entropy": 5.078259563446045,
      "epoch": 1.7139993317741395,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047064766435927585,
      "loss": 4.8112,
      "mean_token_accuracy": 0.21743467152118684,
      "num_tokens": 37768515.0,
      "step": 20520
    },
    {
      "entropy": 5.085023593902588,
      "epoch": 1.7144169729368528,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047063296064191846,
      "loss": 4.7634,
      "mean_token_accuracy": 0.21654411554336547,
      "num_tokens": 37777187.0,
      "step": 20525
    },
    {
      "entropy": 5.0983844757080075,
      "epoch": 1.7148346140995656,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047061825349977067,
      "loss": 4.8127,
      "mean_token_accuracy": 0.22046992927789688,
      "num_tokens": 37785552.0,
      "step": 20530
    },
    {
      "entropy": 5.113344001770019,
      "epoch": 1.7152522552622786,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047060354293308976,
      "loss": 4.865,
      "mean_token_accuracy": 0.20681077092885972,
      "num_tokens": 37795026.0,
      "step": 20535
    },
    {
      "entropy": 5.17663803100586,
      "epoch": 1.7156698964249917,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004705888289421335,
      "loss": 4.8904,
      "mean_token_accuracy": 0.20512870997190474,
      "num_tokens": 37804750.0,
      "step": 20540
    },
    {
      "entropy": 5.040407991409301,
      "epoch": 1.7160875375877045,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004705741115271593,
      "loss": 4.709,
      "mean_token_accuracy": 0.22259356826543808,
      "num_tokens": 37812941.0,
      "step": 20545
    },
    {
      "entropy": 5.069407892227173,
      "epoch": 1.7165051787504177,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004705593906884249,
      "loss": 4.8971,
      "mean_token_accuracy": 0.21662195175886154,
      "num_tokens": 37821911.0,
      "step": 20550
    },
    {
      "entropy": 5.1317604064941404,
      "epoch": 1.7169228199131306,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047054466642618793,
      "loss": 4.8926,
      "mean_token_accuracy": 0.20476185381412507,
      "num_tokens": 37832012.0,
      "step": 20555
    },
    {
      "entropy": 5.104404926300049,
      "epoch": 1.7173404610758436,
      "grad_norm": 1.25,
      "learning_rate": 0.00047052993874070623,
      "loss": 4.8083,
      "mean_token_accuracy": 0.21397096812725067,
      "num_tokens": 37841566.0,
      "step": 20560
    },
    {
      "entropy": 5.0066736221313475,
      "epoch": 1.7177581022385566,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047051520763223754,
      "loss": 4.7067,
      "mean_token_accuracy": 0.22203484773635865,
      "num_tokens": 37849565.0,
      "step": 20565
    },
    {
      "entropy": 5.01431679725647,
      "epoch": 1.7181757434012697,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047050047310103984,
      "loss": 4.7973,
      "mean_token_accuracy": 0.21687465012073517,
      "num_tokens": 37858409.0,
      "step": 20570
    },
    {
      "entropy": 5.124958848953247,
      "epoch": 1.7185933845639827,
      "grad_norm": 1.25,
      "learning_rate": 0.00047048573514737107,
      "loss": 4.9171,
      "mean_token_accuracy": 0.21245170384645462,
      "num_tokens": 37866608.0,
      "step": 20575
    },
    {
      "entropy": 5.191235685348511,
      "epoch": 1.7190110257266955,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047047099377148916,
      "loss": 4.8634,
      "mean_token_accuracy": 0.2130924478173256,
      "num_tokens": 37875371.0,
      "step": 20580
    },
    {
      "entropy": 5.072074699401855,
      "epoch": 1.7194286668894088,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004704562489736522,
      "loss": 4.8334,
      "mean_token_accuracy": 0.2183643698692322,
      "num_tokens": 37884366.0,
      "step": 20585
    },
    {
      "entropy": 5.1223616123199465,
      "epoch": 1.7198463080521216,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004704415007541184,
      "loss": 4.8162,
      "mean_token_accuracy": 0.2155591294169426,
      "num_tokens": 37893328.0,
      "step": 20590
    },
    {
      "entropy": 5.086316347122192,
      "epoch": 1.7202639492148346,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004704267491131458,
      "loss": 4.8221,
      "mean_token_accuracy": 0.21865233778953552,
      "num_tokens": 37902703.0,
      "step": 20595
    },
    {
      "entropy": 5.096631145477295,
      "epoch": 1.7206815903775476,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047041199405099273,
      "loss": 4.8517,
      "mean_token_accuracy": 0.21077930331230163,
      "num_tokens": 37911555.0,
      "step": 20600
    },
    {
      "entropy": 5.066667366027832,
      "epoch": 1.7210992315402605,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047039723556791744,
      "loss": 4.7821,
      "mean_token_accuracy": 0.2144336074590683,
      "num_tokens": 37921208.0,
      "step": 20605
    },
    {
      "entropy": 5.09377613067627,
      "epoch": 1.7215168727029737,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004703824736641785,
      "loss": 4.8254,
      "mean_token_accuracy": 0.21420473158359526,
      "num_tokens": 37931256.0,
      "step": 20610
    },
    {
      "entropy": 5.071801424026489,
      "epoch": 1.7219345138656865,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047036770834003405,
      "loss": 4.8363,
      "mean_token_accuracy": 0.2077750340104103,
      "num_tokens": 37939717.0,
      "step": 20615
    },
    {
      "entropy": 5.075103712081909,
      "epoch": 1.7223521550283996,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047035293959574273,
      "loss": 4.7538,
      "mean_token_accuracy": 0.22066730260849,
      "num_tokens": 37948723.0,
      "step": 20620
    },
    {
      "entropy": 5.161825132369995,
      "epoch": 1.7227697961911126,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000470338167431563,
      "loss": 4.8441,
      "mean_token_accuracy": 0.20946189165115356,
      "num_tokens": 37957666.0,
      "step": 20625
    },
    {
      "entropy": 5.131910276412964,
      "epoch": 1.7231874373538256,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047032339184775366,
      "loss": 4.8795,
      "mean_token_accuracy": 0.21333373486995696,
      "num_tokens": 37966231.0,
      "step": 20630
    },
    {
      "entropy": 5.154679918289185,
      "epoch": 1.7236050785165387,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004703086128445731,
      "loss": 4.9155,
      "mean_token_accuracy": 0.20287350416183472,
      "num_tokens": 37975684.0,
      "step": 20635
    },
    {
      "entropy": 5.248047018051148,
      "epoch": 1.7240227196792515,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004702938304222803,
      "loss": 4.9264,
      "mean_token_accuracy": 0.2030525028705597,
      "num_tokens": 37984874.0,
      "step": 20640
    },
    {
      "entropy": 5.112672567367554,
      "epoch": 1.7244403608419647,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047027904458113386,
      "loss": 4.8587,
      "mean_token_accuracy": 0.2153026983141899,
      "num_tokens": 37994407.0,
      "step": 20645
    },
    {
      "entropy": 5.134621572494507,
      "epoch": 1.7248580020046775,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004702642553213927,
      "loss": 5.0002,
      "mean_token_accuracy": 0.2079465225338936,
      "num_tokens": 38003588.0,
      "step": 20650
    },
    {
      "entropy": 5.252465295791626,
      "epoch": 1.7252756431673906,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047024946264331564,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20491937249898912,
      "num_tokens": 38012761.0,
      "step": 20655
    },
    {
      "entropy": 5.184746170043946,
      "epoch": 1.7256932843301036,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004702346665471618,
      "loss": 4.9137,
      "mean_token_accuracy": 0.21189108788967131,
      "num_tokens": 38022159.0,
      "step": 20660
    },
    {
      "entropy": 5.2094292640686035,
      "epoch": 1.7261109254928164,
      "grad_norm": 1.125,
      "learning_rate": 0.0004702198670331901,
      "loss": 4.7942,
      "mean_token_accuracy": 0.21205705404281616,
      "num_tokens": 38031095.0,
      "step": 20665
    },
    {
      "entropy": 5.2023382663726805,
      "epoch": 1.7265285666555297,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047020506410165965,
      "loss": 4.9866,
      "mean_token_accuracy": 0.19818095564842225,
      "num_tokens": 38040764.0,
      "step": 20670
    },
    {
      "entropy": 5.141391801834106,
      "epoch": 1.7269462078182425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047019025775282955,
      "loss": 4.8335,
      "mean_token_accuracy": 0.21111426651477813,
      "num_tokens": 38050212.0,
      "step": 20675
    },
    {
      "entropy": 5.071388578414917,
      "epoch": 1.7273638489809555,
      "grad_norm": 1.046875,
      "learning_rate": 0.000470175447986959,
      "loss": 4.8351,
      "mean_token_accuracy": 0.20746724754571916,
      "num_tokens": 38059980.0,
      "step": 20680
    },
    {
      "entropy": 5.105818271636963,
      "epoch": 1.7277814901436686,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004701606348043074,
      "loss": 4.819,
      "mean_token_accuracy": 0.21833140552043914,
      "num_tokens": 38068894.0,
      "step": 20685
    },
    {
      "entropy": 5.06648850440979,
      "epoch": 1.7281991313063816,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004701458182051339,
      "loss": 4.8039,
      "mean_token_accuracy": 0.21860468685626983,
      "num_tokens": 38078277.0,
      "step": 20690
    },
    {
      "entropy": 5.133325862884521,
      "epoch": 1.7286167724690946,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047013099818969797,
      "loss": 4.8021,
      "mean_token_accuracy": 0.21737958937883378,
      "num_tokens": 38087059.0,
      "step": 20695
    },
    {
      "entropy": 5.106836891174316,
      "epoch": 1.7290344136318074,
      "grad_norm": 1.21875,
      "learning_rate": 0.000470116174758259,
      "loss": 4.8461,
      "mean_token_accuracy": 0.21619001924991607,
      "num_tokens": 38097103.0,
      "step": 20700
    },
    {
      "entropy": 5.079181432723999,
      "epoch": 1.7294520547945207,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047010134791107663,
      "loss": 4.892,
      "mean_token_accuracy": 0.20823642760515212,
      "num_tokens": 38106173.0,
      "step": 20705
    },
    {
      "entropy": 5.103617000579834,
      "epoch": 1.7298696959572335,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047008651764841034,
      "loss": 4.7883,
      "mean_token_accuracy": 0.21604625582695008,
      "num_tokens": 38115996.0,
      "step": 20710
    },
    {
      "entropy": 5.125721263885498,
      "epoch": 1.7302873371199465,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047007168397051956,
      "loss": 4.836,
      "mean_token_accuracy": 0.21385694146156312,
      "num_tokens": 38125286.0,
      "step": 20715
    },
    {
      "entropy": 5.111641216278076,
      "epoch": 1.7307049782826596,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047005684687766424,
      "loss": 4.8436,
      "mean_token_accuracy": 0.2107926696538925,
      "num_tokens": 38133866.0,
      "step": 20720
    },
    {
      "entropy": 5.133185529708863,
      "epoch": 1.7311226194453724,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004700420063701041,
      "loss": 4.9127,
      "mean_token_accuracy": 0.2044650912284851,
      "num_tokens": 38142408.0,
      "step": 20725
    },
    {
      "entropy": 5.084585857391358,
      "epoch": 1.7315402606080856,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004700271624480988,
      "loss": 4.6671,
      "mean_token_accuracy": 0.23332337588071822,
      "num_tokens": 38150800.0,
      "step": 20730
    },
    {
      "entropy": 5.0128593921661375,
      "epoch": 1.7319579017707984,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047001231511190826,
      "loss": 4.8033,
      "mean_token_accuracy": 0.21848888248205184,
      "num_tokens": 38159885.0,
      "step": 20735
    },
    {
      "entropy": 5.134888696670532,
      "epoch": 1.7323755429335115,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004699974643617924,
      "loss": 4.8806,
      "mean_token_accuracy": 0.21069645881652832,
      "num_tokens": 38168663.0,
      "step": 20740
    },
    {
      "entropy": 5.113056468963623,
      "epoch": 1.7327931840962245,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004699826101980112,
      "loss": 4.7911,
      "mean_token_accuracy": 0.21926398426294327,
      "num_tokens": 38177326.0,
      "step": 20745
    },
    {
      "entropy": 5.159714889526367,
      "epoch": 1.7332108252589375,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004699677526208248,
      "loss": 4.8654,
      "mean_token_accuracy": 0.20961318910121918,
      "num_tokens": 38186333.0,
      "step": 20750
    },
    {
      "entropy": 5.10797963142395,
      "epoch": 1.7336284664216506,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046995289163049317,
      "loss": 4.8189,
      "mean_token_accuracy": 0.22035433053970338,
      "num_tokens": 38195479.0,
      "step": 20755
    },
    {
      "entropy": 5.079159212112427,
      "epoch": 1.7340461075843634,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046993802722727646,
      "loss": 4.8133,
      "mean_token_accuracy": 0.21587395071983337,
      "num_tokens": 38204450.0,
      "step": 20760
    },
    {
      "entropy": 5.049137830734253,
      "epoch": 1.7344637487470767,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004699231594114349,
      "loss": 4.7926,
      "mean_token_accuracy": 0.21366769075393677,
      "num_tokens": 38212907.0,
      "step": 20765
    },
    {
      "entropy": 5.0911651134490965,
      "epoch": 1.7348813899097895,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004699082881832289,
      "loss": 4.7751,
      "mean_token_accuracy": 0.2166141539812088,
      "num_tokens": 38222081.0,
      "step": 20770
    },
    {
      "entropy": 5.115407943725586,
      "epoch": 1.7352990310725025,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046989341354291864,
      "loss": 4.8646,
      "mean_token_accuracy": 0.20728254765272142,
      "num_tokens": 38232142.0,
      "step": 20775
    },
    {
      "entropy": 5.073178386688232,
      "epoch": 1.7357166722352155,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004698785354907646,
      "loss": 4.7609,
      "mean_token_accuracy": 0.22778155356645585,
      "num_tokens": 38240973.0,
      "step": 20780
    },
    {
      "entropy": 5.144483661651611,
      "epoch": 1.7361343133979283,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004698636540270272,
      "loss": 4.9241,
      "mean_token_accuracy": 0.21013398319482804,
      "num_tokens": 38250793.0,
      "step": 20785
    },
    {
      "entropy": 5.093199634552002,
      "epoch": 1.7365519545606416,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004698487691519671,
      "loss": 4.8884,
      "mean_token_accuracy": 0.2092108279466629,
      "num_tokens": 38260186.0,
      "step": 20790
    },
    {
      "entropy": 5.1239861965179445,
      "epoch": 1.7369695957233544,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046983388086584463,
      "loss": 4.8789,
      "mean_token_accuracy": 0.21470686048269272,
      "num_tokens": 38269385.0,
      "step": 20795
    },
    {
      "entropy": 5.203597497940064,
      "epoch": 1.7373872368860674,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046981898916892054,
      "loss": 4.91,
      "mean_token_accuracy": 0.205025377869606,
      "num_tokens": 38278746.0,
      "step": 20800
    },
    {
      "entropy": 5.163279294967651,
      "epoch": 1.7378048780487805,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046980409406145555,
      "loss": 4.8888,
      "mean_token_accuracy": 0.2097990781068802,
      "num_tokens": 38289111.0,
      "step": 20805
    },
    {
      "entropy": 5.180445766448974,
      "epoch": 1.7382225192114935,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004697891955437105,
      "loss": 4.9209,
      "mean_token_accuracy": 0.21061075627803802,
      "num_tokens": 38298250.0,
      "step": 20810
    },
    {
      "entropy": 5.113266324996948,
      "epoch": 1.7386401603742065,
      "grad_norm": 1.125,
      "learning_rate": 0.00046977429361594596,
      "loss": 4.8709,
      "mean_token_accuracy": 0.20932674407958984,
      "num_tokens": 38306804.0,
      "step": 20815
    },
    {
      "entropy": 5.115199565887451,
      "epoch": 1.7390578015369194,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004697593882784231,
      "loss": 4.8793,
      "mean_token_accuracy": 0.2136407122015953,
      "num_tokens": 38316183.0,
      "step": 20820
    },
    {
      "entropy": 5.115050983428955,
      "epoch": 1.7394754426996326,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004697444795314027,
      "loss": 4.7887,
      "mean_token_accuracy": 0.21747759282588958,
      "num_tokens": 38325624.0,
      "step": 20825
    },
    {
      "entropy": 5.166948318481445,
      "epoch": 1.7398930838623454,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046972956737514573,
      "loss": 4.9203,
      "mean_token_accuracy": 0.21865905374288558,
      "num_tokens": 38334858.0,
      "step": 20830
    },
    {
      "entropy": 5.156792831420899,
      "epoch": 1.7403107250250585,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004697146518099133,
      "loss": 4.826,
      "mean_token_accuracy": 0.2131825566291809,
      "num_tokens": 38343862.0,
      "step": 20835
    },
    {
      "entropy": 5.097906541824341,
      "epoch": 1.7407283661877715,
      "grad_norm": 1.25,
      "learning_rate": 0.00046969973283596657,
      "loss": 4.8725,
      "mean_token_accuracy": 0.22269484400749207,
      "num_tokens": 38352664.0,
      "step": 20840
    },
    {
      "entropy": 5.080182218551636,
      "epoch": 1.7411460073504843,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046968481045356656,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21471934020519257,
      "num_tokens": 38361806.0,
      "step": 20845
    },
    {
      "entropy": 5.135369729995728,
      "epoch": 1.7415636485131976,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046966988466297464,
      "loss": 4.8852,
      "mean_token_accuracy": 0.20965492874383926,
      "num_tokens": 38372132.0,
      "step": 20850
    },
    {
      "entropy": 5.074549579620362,
      "epoch": 1.7419812896759104,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046965495546445204,
      "loss": 4.7947,
      "mean_token_accuracy": 0.21658263057470323,
      "num_tokens": 38381300.0,
      "step": 20855
    },
    {
      "entropy": 5.117762899398803,
      "epoch": 1.7423989308386234,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046964002285826016,
      "loss": 4.8938,
      "mean_token_accuracy": 0.21389057189226152,
      "num_tokens": 38389887.0,
      "step": 20860
    },
    {
      "entropy": 5.067674398422241,
      "epoch": 1.7428165720013364,
      "grad_norm": 1.25,
      "learning_rate": 0.0004696250868446604,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2095635250210762,
      "num_tokens": 38398953.0,
      "step": 20865
    },
    {
      "entropy": 5.099326515197754,
      "epoch": 1.7432342131640495,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046961014742391425,
      "loss": 4.9198,
      "mean_token_accuracy": 0.21587122082710267,
      "num_tokens": 38408176.0,
      "step": 20870
    },
    {
      "entropy": 5.13854398727417,
      "epoch": 1.7436518543267625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046959520459628317,
      "loss": 5.038,
      "mean_token_accuracy": 0.20911342650651932,
      "num_tokens": 38417711.0,
      "step": 20875
    },
    {
      "entropy": 5.251838684082031,
      "epoch": 1.7440694954894753,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046958025836202886,
      "loss": 4.869,
      "mean_token_accuracy": 0.21142177134752274,
      "num_tokens": 38426524.0,
      "step": 20880
    },
    {
      "entropy": 5.093953084945679,
      "epoch": 1.7444871366521886,
      "grad_norm": 1.25,
      "learning_rate": 0.0004695653087214128,
      "loss": 4.8564,
      "mean_token_accuracy": 0.21611799895763398,
      "num_tokens": 38435584.0,
      "step": 20885
    },
    {
      "entropy": 5.046050977706909,
      "epoch": 1.7449047778149014,
      "grad_norm": 1.125,
      "learning_rate": 0.00046955035567469695,
      "loss": 4.7923,
      "mean_token_accuracy": 0.21280910819768906,
      "num_tokens": 38444958.0,
      "step": 20890
    },
    {
      "entropy": 5.142476654052734,
      "epoch": 1.7453224189776144,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046953539922214285,
      "loss": 5.0012,
      "mean_token_accuracy": 0.1957467943429947,
      "num_tokens": 38454781.0,
      "step": 20895
    },
    {
      "entropy": 4.994678211212158,
      "epoch": 1.7457400601403275,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004695204393640125,
      "loss": 4.6483,
      "mean_token_accuracy": 0.22326820939779282,
      "num_tokens": 38463430.0,
      "step": 20900
    },
    {
      "entropy": 5.1028724193573,
      "epoch": 1.7461577013030403,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046950547610056766,
      "loss": 4.9023,
      "mean_token_accuracy": 0.20673508048057557,
      "num_tokens": 38472649.0,
      "step": 20905
    },
    {
      "entropy": 5.129825735092163,
      "epoch": 1.7465753424657535,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004694905094320703,
      "loss": 4.7821,
      "mean_token_accuracy": 0.2204578325152397,
      "num_tokens": 38482090.0,
      "step": 20910
    },
    {
      "entropy": 5.207033920288086,
      "epoch": 1.7469929836284663,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046947553935878253,
      "loss": 4.9561,
      "mean_token_accuracy": 0.1975427582859993,
      "num_tokens": 38491521.0,
      "step": 20915
    },
    {
      "entropy": 5.171026182174683,
      "epoch": 1.7474106247911794,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004694605658809663,
      "loss": 4.909,
      "mean_token_accuracy": 0.21064421981573106,
      "num_tokens": 38500289.0,
      "step": 20920
    },
    {
      "entropy": 5.034094190597534,
      "epoch": 1.7478282659538924,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004694455889988838,
      "loss": 4.7122,
      "mean_token_accuracy": 0.2189274698495865,
      "num_tokens": 38509518.0,
      "step": 20925
    },
    {
      "entropy": 5.125507831573486,
      "epoch": 1.7482459071166054,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046943060871279725,
      "loss": 4.9552,
      "mean_token_accuracy": 0.20224931538105012,
      "num_tokens": 38518207.0,
      "step": 20930
    },
    {
      "entropy": 5.280204486846924,
      "epoch": 1.7486635482793185,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046941562502296886,
      "loss": 5.089,
      "mean_token_accuracy": 0.2004919856786728,
      "num_tokens": 38528080.0,
      "step": 20935
    },
    {
      "entropy": 5.168444204330444,
      "epoch": 1.7490811894420313,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004694006379296609,
      "loss": 4.8058,
      "mean_token_accuracy": 0.21868816614151002,
      "num_tokens": 38537051.0,
      "step": 20940
    },
    {
      "entropy": 5.033253574371338,
      "epoch": 1.7494988306047445,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046938564743313576,
      "loss": 4.8331,
      "mean_token_accuracy": 0.20862553268671036,
      "num_tokens": 38546303.0,
      "step": 20945
    },
    {
      "entropy": 5.102825689315796,
      "epoch": 1.7499164717674573,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046937065353365594,
      "loss": 4.8679,
      "mean_token_accuracy": 0.20955113768577577,
      "num_tokens": 38555504.0,
      "step": 20950
    },
    {
      "entropy": 5.116598892211914,
      "epoch": 1.7503341129301704,
      "grad_norm": 1.125,
      "learning_rate": 0.00046935565623148383,
      "loss": 4.8003,
      "mean_token_accuracy": 0.21623941361904145,
      "num_tokens": 38564194.0,
      "step": 20955
    },
    {
      "entropy": 5.139825677871704,
      "epoch": 1.7507517540928834,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046934065552688196,
      "loss": 4.9185,
      "mean_token_accuracy": 0.21231213957071304,
      "num_tokens": 38573332.0,
      "step": 20960
    },
    {
      "entropy": 5.093293857574463,
      "epoch": 1.7511693952555962,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004693256514201131,
      "loss": 4.8473,
      "mean_token_accuracy": 0.21435588896274566,
      "num_tokens": 38581737.0,
      "step": 20965
    },
    {
      "entropy": 5.126632165908814,
      "epoch": 1.7515870364183095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046931064391143976,
      "loss": 4.8555,
      "mean_token_accuracy": 0.21377064734697343,
      "num_tokens": 38590703.0,
      "step": 20970
    },
    {
      "entropy": 5.1166627407073975,
      "epoch": 1.7520046775810223,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046929563300112466,
      "loss": 4.8931,
      "mean_token_accuracy": 0.2062434062361717,
      "num_tokens": 38600516.0,
      "step": 20975
    },
    {
      "entropy": 5.101352119445801,
      "epoch": 1.7524223187437353,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046928061868943067,
      "loss": 4.8198,
      "mean_token_accuracy": 0.21373034566640853,
      "num_tokens": 38609447.0,
      "step": 20980
    },
    {
      "entropy": 5.0761181831359865,
      "epoch": 1.7528399599064484,
      "grad_norm": 1.125,
      "learning_rate": 0.0004692656009766206,
      "loss": 4.7918,
      "mean_token_accuracy": 0.21717526018619537,
      "num_tokens": 38619091.0,
      "step": 20985
    },
    {
      "entropy": 5.0608478546142575,
      "epoch": 1.7532576010691614,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046925057986295734,
      "loss": 4.8959,
      "mean_token_accuracy": 0.20733658075332642,
      "num_tokens": 38628012.0,
      "step": 20990
    },
    {
      "entropy": 5.1397590160369875,
      "epoch": 1.7536752422318744,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046923555534870385,
      "loss": 4.7641,
      "mean_token_accuracy": 0.217573444545269,
      "num_tokens": 38636822.0,
      "step": 20995
    },
    {
      "entropy": 5.017159366607666,
      "epoch": 1.7540928833945872,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004692205274341232,
      "loss": 4.7921,
      "mean_token_accuracy": 0.21931743174791335,
      "num_tokens": 38645127.0,
      "step": 21000
    },
    {
      "epoch": 1.7540928833945872,
      "eval_entropy": 4.815267143398997,
      "eval_loss": 4.8730926513671875,
      "eval_mean_token_accuracy": 0.22291728947765366,
      "eval_num_tokens": 38645127.0,
      "eval_runtime": 27.4046,
      "eval_samples_per_second": 1377.982,
      "eval_steps_per_second": 172.271,
      "step": 21000
    },
    {
      "entropy": 5.018155813217163,
      "epoch": 1.7545105245573005,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046920549611947834,
      "loss": 4.7343,
      "mean_token_accuracy": 0.21977752298116685,
      "num_tokens": 38653534.0,
      "step": 21005
    },
    {
      "entropy": 5.168481302261353,
      "epoch": 1.7549281657200133,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004691904614050326,
      "loss": 4.9247,
      "mean_token_accuracy": 0.2135736882686615,
      "num_tokens": 38663756.0,
      "step": 21010
    },
    {
      "entropy": 5.1746419906616214,
      "epoch": 1.7553458068827263,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046917542329104903,
      "loss": 4.9176,
      "mean_token_accuracy": 0.20757098644971847,
      "num_tokens": 38673548.0,
      "step": 21015
    },
    {
      "entropy": 5.1707347393035885,
      "epoch": 1.7557634480454394,
      "grad_norm": 1.125,
      "learning_rate": 0.0004691603817777909,
      "loss": 4.8894,
      "mean_token_accuracy": 0.21412585973739623,
      "num_tokens": 38683177.0,
      "step": 21020
    },
    {
      "entropy": 5.0406769752502445,
      "epoch": 1.7561810892081522,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046914533686552166,
      "loss": 4.772,
      "mean_token_accuracy": 0.22015785425901413,
      "num_tokens": 38691920.0,
      "step": 21025
    },
    {
      "entropy": 5.119328451156616,
      "epoch": 1.7565987303708654,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004691302885545046,
      "loss": 4.8708,
      "mean_token_accuracy": 0.20943653732538223,
      "num_tokens": 38701463.0,
      "step": 21030
    },
    {
      "entropy": 5.0336754322052,
      "epoch": 1.7570163715335783,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004691152368450032,
      "loss": 4.7319,
      "mean_token_accuracy": 0.22406265288591384,
      "num_tokens": 38710719.0,
      "step": 21035
    },
    {
      "entropy": 5.152604103088379,
      "epoch": 1.7574340126962913,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046910018173728084,
      "loss": 4.8247,
      "mean_token_accuracy": 0.2161321982741356,
      "num_tokens": 38719427.0,
      "step": 21040
    },
    {
      "entropy": 5.055587244033814,
      "epoch": 1.7578516538590043,
      "grad_norm": 1.125,
      "learning_rate": 0.0004690851232316012,
      "loss": 4.8616,
      "mean_token_accuracy": 0.218145714700222,
      "num_tokens": 38728588.0,
      "step": 21045
    },
    {
      "entropy": 5.06435751914978,
      "epoch": 1.7582692950217174,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046907006132822785,
      "loss": 4.8398,
      "mean_token_accuracy": 0.21528709828853607,
      "num_tokens": 38737245.0,
      "step": 21050
    },
    {
      "entropy": 5.137093639373779,
      "epoch": 1.7586869361844304,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046905499602742444,
      "loss": 4.8038,
      "mean_token_accuracy": 0.21213243007659913,
      "num_tokens": 38747302.0,
      "step": 21055
    },
    {
      "entropy": 5.078737831115722,
      "epoch": 1.7591045773471432,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046903992732945485,
      "loss": 4.7648,
      "mean_token_accuracy": 0.22276817113161088,
      "num_tokens": 38755593.0,
      "step": 21060
    },
    {
      "entropy": 5.1194525241851805,
      "epoch": 1.7595222185098565,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046902485523458266,
      "loss": 4.908,
      "mean_token_accuracy": 0.2068672686815262,
      "num_tokens": 38765266.0,
      "step": 21065
    },
    {
      "entropy": 5.0652428150177,
      "epoch": 1.7599398596725693,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046900977974307186,
      "loss": 4.8599,
      "mean_token_accuracy": 0.21267974823713304,
      "num_tokens": 38775515.0,
      "step": 21070
    },
    {
      "entropy": 5.166687679290772,
      "epoch": 1.7603575008352823,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004689947008551863,
      "loss": 4.9007,
      "mean_token_accuracy": 0.21323261559009551,
      "num_tokens": 38785492.0,
      "step": 21075
    },
    {
      "entropy": 5.059709358215332,
      "epoch": 1.7607751419979953,
      "grad_norm": 1.125,
      "learning_rate": 0.0004689796185711901,
      "loss": 4.8492,
      "mean_token_accuracy": 0.21411344558000564,
      "num_tokens": 38795006.0,
      "step": 21080
    },
    {
      "entropy": 5.062517833709717,
      "epoch": 1.7611927831607082,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046896453289134703,
      "loss": 4.7804,
      "mean_token_accuracy": 0.2127622976899147,
      "num_tokens": 38803480.0,
      "step": 21085
    },
    {
      "entropy": 5.215548467636109,
      "epoch": 1.7616104243234214,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046894944381592143,
      "loss": 4.9943,
      "mean_token_accuracy": 0.20018794685602187,
      "num_tokens": 38813869.0,
      "step": 21090
    },
    {
      "entropy": 5.197828102111816,
      "epoch": 1.7620280654861342,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046893435134517735,
      "loss": 4.922,
      "mean_token_accuracy": 0.2086518943309784,
      "num_tokens": 38823964.0,
      "step": 21095
    },
    {
      "entropy": 5.1074189186096195,
      "epoch": 1.7624457066488475,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046891925547937907,
      "loss": 4.8685,
      "mean_token_accuracy": 0.21772514581680297,
      "num_tokens": 38832917.0,
      "step": 21100
    },
    {
      "entropy": 5.053288841247559,
      "epoch": 1.7628633478115603,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046890415621879075,
      "loss": 4.8136,
      "mean_token_accuracy": 0.20966860502958298,
      "num_tokens": 38841611.0,
      "step": 21105
    },
    {
      "entropy": 5.103509855270386,
      "epoch": 1.7632809889742733,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004688890535636767,
      "loss": 4.7829,
      "mean_token_accuracy": 0.22039124965667725,
      "num_tokens": 38850632.0,
      "step": 21110
    },
    {
      "entropy": 5.08984637260437,
      "epoch": 1.7636986301369864,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004688739475143015,
      "loss": 4.8375,
      "mean_token_accuracy": 0.2181856483221054,
      "num_tokens": 38859910.0,
      "step": 21115
    },
    {
      "entropy": 5.115202379226685,
      "epoch": 1.7641162712996992,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046885883807092944,
      "loss": 4.8821,
      "mean_token_accuracy": 0.210603004693985,
      "num_tokens": 38868750.0,
      "step": 21120
    },
    {
      "entropy": 5.179526948928833,
      "epoch": 1.7645339124624124,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046884372523382507,
      "loss": 4.8843,
      "mean_token_accuracy": 0.21014532148838044,
      "num_tokens": 38877661.0,
      "step": 21125
    },
    {
      "entropy": 5.131812524795532,
      "epoch": 1.7649515536251252,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046882860900325296,
      "loss": 4.8923,
      "mean_token_accuracy": 0.2160434231162071,
      "num_tokens": 38886969.0,
      "step": 21130
    },
    {
      "entropy": 5.067768239974976,
      "epoch": 1.7653691947878383,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004688134893794777,
      "loss": 4.8335,
      "mean_token_accuracy": 0.2099620744585991,
      "num_tokens": 38896741.0,
      "step": 21135
    },
    {
      "entropy": 5.092642307281494,
      "epoch": 1.7657868359505513,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046879836636276405,
      "loss": 4.8905,
      "mean_token_accuracy": 0.2083564206957817,
      "num_tokens": 38906174.0,
      "step": 21140
    },
    {
      "entropy": 5.124623346328735,
      "epoch": 1.7662044771132643,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004687832399533767,
      "loss": 4.8514,
      "mean_token_accuracy": 0.2110178455710411,
      "num_tokens": 38914623.0,
      "step": 21145
    },
    {
      "entropy": 5.089209699630738,
      "epoch": 1.7666221182759774,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004687681101515805,
      "loss": 4.8161,
      "mean_token_accuracy": 0.21250998824834824,
      "num_tokens": 38923458.0,
      "step": 21150
    },
    {
      "entropy": 5.189557313919067,
      "epoch": 1.7670397594386902,
      "grad_norm": 1.125,
      "learning_rate": 0.0004687529769576403,
      "loss": 4.933,
      "mean_token_accuracy": 0.21280450969934464,
      "num_tokens": 38932763.0,
      "step": 21155
    },
    {
      "entropy": 5.129326725006104,
      "epoch": 1.7674574006014034,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046873784037182106,
      "loss": 4.8173,
      "mean_token_accuracy": 0.20679975152015687,
      "num_tokens": 38941936.0,
      "step": 21160
    },
    {
      "entropy": 5.146593999862671,
      "epoch": 1.7678750417641163,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046872270039438764,
      "loss": 4.9067,
      "mean_token_accuracy": 0.20826407223939897,
      "num_tokens": 38951662.0,
      "step": 21165
    },
    {
      "entropy": 5.270387315750122,
      "epoch": 1.7682926829268293,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004687075570256052,
      "loss": 5.0031,
      "mean_token_accuracy": 0.20513854324817657,
      "num_tokens": 38961903.0,
      "step": 21170
    },
    {
      "entropy": 5.123607206344604,
      "epoch": 1.7687103240895423,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046869241026573885,
      "loss": 4.7985,
      "mean_token_accuracy": 0.21245625913143157,
      "num_tokens": 38970598.0,
      "step": 21175
    },
    {
      "entropy": 5.082724285125733,
      "epoch": 1.7691279652522551,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046867726011505365,
      "loss": 4.8304,
      "mean_token_accuracy": 0.217740797996521,
      "num_tokens": 38979132.0,
      "step": 21180
    },
    {
      "entropy": 5.094335126876831,
      "epoch": 1.7695456064149684,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004686621065738149,
      "loss": 4.8771,
      "mean_token_accuracy": 0.21637383997440338,
      "num_tokens": 38987887.0,
      "step": 21185
    },
    {
      "entropy": 5.051493310928345,
      "epoch": 1.7699632475776812,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046864694964228785,
      "loss": 4.6948,
      "mean_token_accuracy": 0.22017006129026412,
      "num_tokens": 38995928.0,
      "step": 21190
    },
    {
      "entropy": 5.058885669708252,
      "epoch": 1.7703808887403942,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004686317893207378,
      "loss": 4.8365,
      "mean_token_accuracy": 0.22066232115030288,
      "num_tokens": 39004958.0,
      "step": 21195
    },
    {
      "entropy": 5.181520318984985,
      "epoch": 1.7707985299031073,
      "grad_norm": 1.125,
      "learning_rate": 0.0004686166256094303,
      "loss": 4.9739,
      "mean_token_accuracy": 0.20551970601081848,
      "num_tokens": 39015463.0,
      "step": 21200
    },
    {
      "entropy": 5.180568265914917,
      "epoch": 1.7712161710658203,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004686014585086307,
      "loss": 4.8601,
      "mean_token_accuracy": 0.20712645053863527,
      "num_tokens": 39024575.0,
      "step": 21205
    },
    {
      "entropy": 5.180229187011719,
      "epoch": 1.7716338122285333,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004685862880186045,
      "loss": 4.9601,
      "mean_token_accuracy": 0.2008558303117752,
      "num_tokens": 39033944.0,
      "step": 21210
    },
    {
      "entropy": 5.0521808624267575,
      "epoch": 1.7720514533912461,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004685711141396173,
      "loss": 4.8611,
      "mean_token_accuracy": 0.21161395460367202,
      "num_tokens": 39043191.0,
      "step": 21215
    },
    {
      "entropy": 5.092403364181519,
      "epoch": 1.7724690945539594,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046855593687193474,
      "loss": 4.8454,
      "mean_token_accuracy": 0.2111523777246475,
      "num_tokens": 39052238.0,
      "step": 21220
    },
    {
      "entropy": 5.062857389450073,
      "epoch": 1.7728867357166722,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004685407562158226,
      "loss": 4.7662,
      "mean_token_accuracy": 0.2210694283246994,
      "num_tokens": 39061222.0,
      "step": 21225
    },
    {
      "entropy": 4.98867449760437,
      "epoch": 1.7733043768793852,
      "grad_norm": 1.125,
      "learning_rate": 0.0004685255721715465,
      "loss": 4.7071,
      "mean_token_accuracy": 0.2195749834179878,
      "num_tokens": 39069585.0,
      "step": 21230
    },
    {
      "entropy": 5.177379035949707,
      "epoch": 1.7737220180420983,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004685103847393723,
      "loss": 4.8914,
      "mean_token_accuracy": 0.2123609721660614,
      "num_tokens": 39079418.0,
      "step": 21235
    },
    {
      "entropy": 5.0587756633758545,
      "epoch": 1.774139659204811,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046849519391956593,
      "loss": 4.8296,
      "mean_token_accuracy": 0.2180938810110092,
      "num_tokens": 39088666.0,
      "step": 21240
    },
    {
      "entropy": 5.146849822998047,
      "epoch": 1.7745573003675243,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004684799997123932,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21043998152017593,
      "num_tokens": 39097876.0,
      "step": 21245
    },
    {
      "entropy": 5.185274219512939,
      "epoch": 1.7749749415302372,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004684648021181203,
      "loss": 4.8601,
      "mean_token_accuracy": 0.21111945658922196,
      "num_tokens": 39107459.0,
      "step": 21250
    },
    {
      "entropy": 5.117038631439209,
      "epoch": 1.7753925826929502,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046844960113701307,
      "loss": 4.802,
      "mean_token_accuracy": 0.21577999889850616,
      "num_tokens": 39116098.0,
      "step": 21255
    },
    {
      "entropy": 5.119956064224243,
      "epoch": 1.7758102238556632,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046843439676933776,
      "loss": 4.9378,
      "mean_token_accuracy": 0.20842031091451646,
      "num_tokens": 39125461.0,
      "step": 21260
    },
    {
      "entropy": 5.156502771377563,
      "epoch": 1.7762278650183763,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046841918901536047,
      "loss": 4.869,
      "mean_token_accuracy": 0.2036302551627159,
      "num_tokens": 39135957.0,
      "step": 21265
    },
    {
      "entropy": 5.162035655975342,
      "epoch": 1.7766455061810893,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004684039778753474,
      "loss": 4.904,
      "mean_token_accuracy": 0.20520631819963456,
      "num_tokens": 39145187.0,
      "step": 21270
    },
    {
      "entropy": 5.014419460296631,
      "epoch": 1.777063147343802,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046838876334956507,
      "loss": 4.7829,
      "mean_token_accuracy": 0.22064700722694397,
      "num_tokens": 39154836.0,
      "step": 21275
    },
    {
      "entropy": 5.060709381103516,
      "epoch": 1.7774807885065154,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004683735454382795,
      "loss": 4.7785,
      "mean_token_accuracy": 0.22161255925893783,
      "num_tokens": 39164189.0,
      "step": 21280
    },
    {
      "entropy": 4.9935235500335695,
      "epoch": 1.7778984296692282,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004683583241417573,
      "loss": 4.6997,
      "mean_token_accuracy": 0.22527484446763993,
      "num_tokens": 39173055.0,
      "step": 21285
    },
    {
      "entropy": 5.048897171020508,
      "epoch": 1.7783160708319412,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004683430994602649,
      "loss": 4.7989,
      "mean_token_accuracy": 0.2178294077515602,
      "num_tokens": 39181452.0,
      "step": 21290
    },
    {
      "entropy": 5.136565256118774,
      "epoch": 1.7787337119946542,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046832787139406887,
      "loss": 4.8565,
      "mean_token_accuracy": 0.21149153709411622,
      "num_tokens": 39190513.0,
      "step": 21295
    },
    {
      "entropy": 5.122166919708252,
      "epoch": 1.779151353157367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004683126399434357,
      "loss": 4.88,
      "mean_token_accuracy": 0.2085953399538994,
      "num_tokens": 39200293.0,
      "step": 21300
    },
    {
      "entropy": 5.1651664733886715,
      "epoch": 1.7795689943200803,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004682974051086321,
      "loss": 4.9582,
      "mean_token_accuracy": 0.20334525108337403,
      "num_tokens": 39209922.0,
      "step": 21305
    },
    {
      "entropy": 5.257776165008545,
      "epoch": 1.7799866354827931,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046828216688992466,
      "loss": 4.974,
      "mean_token_accuracy": 0.19791336506605148,
      "num_tokens": 39219082.0,
      "step": 21310
    },
    {
      "entropy": 5.070527124404907,
      "epoch": 1.7804042766455062,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004682669252875803,
      "loss": 4.7648,
      "mean_token_accuracy": 0.21933268308639525,
      "num_tokens": 39228639.0,
      "step": 21315
    },
    {
      "entropy": 4.991222286224366,
      "epoch": 1.7808219178082192,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046825168030186577,
      "loss": 4.7733,
      "mean_token_accuracy": 0.21763911843299866,
      "num_tokens": 39237756.0,
      "step": 21320
    },
    {
      "entropy": 5.172712755203247,
      "epoch": 1.7812395589709322,
      "grad_norm": 1.109375,
      "learning_rate": 0.000468236431933048,
      "loss": 4.8335,
      "mean_token_accuracy": 0.21785761564970016,
      "num_tokens": 39246974.0,
      "step": 21325
    },
    {
      "entropy": 5.125245571136475,
      "epoch": 1.7816572001336453,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004682211801813938,
      "loss": 4.7996,
      "mean_token_accuracy": 0.21806920319795609,
      "num_tokens": 39256203.0,
      "step": 21330
    },
    {
      "entropy": 4.976913499832153,
      "epoch": 1.782074841296358,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004682059250471703,
      "loss": 4.7581,
      "mean_token_accuracy": 0.2280579313635826,
      "num_tokens": 39265249.0,
      "step": 21335
    },
    {
      "entropy": 5.069141149520874,
      "epoch": 1.7824924824590713,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046819066653064454,
      "loss": 4.8172,
      "mean_token_accuracy": 0.222671240568161,
      "num_tokens": 39274222.0,
      "step": 21340
    },
    {
      "entropy": 5.15507321357727,
      "epoch": 1.7829101236217841,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046817540463208357,
      "loss": 4.9013,
      "mean_token_accuracy": 0.21006147116422652,
      "num_tokens": 39284095.0,
      "step": 21345
    },
    {
      "entropy": 5.0308387756347654,
      "epoch": 1.7833277647844972,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046816013935175465,
      "loss": 4.7912,
      "mean_token_accuracy": 0.22703619301319122,
      "num_tokens": 39293253.0,
      "step": 21350
    },
    {
      "entropy": 5.051210737228393,
      "epoch": 1.7837454059472102,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000468144870689925,
      "loss": 4.7683,
      "mean_token_accuracy": 0.21812895983457564,
      "num_tokens": 39302817.0,
      "step": 21355
    },
    {
      "entropy": 5.073461198806763,
      "epoch": 1.784163047109923,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004681295986468619,
      "loss": 4.852,
      "mean_token_accuracy": 0.2147398293018341,
      "num_tokens": 39312436.0,
      "step": 21360
    },
    {
      "entropy": 5.126283264160156,
      "epoch": 1.7845806882726363,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004681143232228328,
      "loss": 4.8939,
      "mean_token_accuracy": 0.21209747791290284,
      "num_tokens": 39321649.0,
      "step": 21365
    },
    {
      "entropy": 5.167861747741699,
      "epoch": 1.784998329435349,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046809904441810486,
      "loss": 4.7942,
      "mean_token_accuracy": 0.21326768398284912,
      "num_tokens": 39331040.0,
      "step": 21370
    },
    {
      "entropy": 5.042833137512207,
      "epoch": 1.7854159705980621,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004680837622329457,
      "loss": 4.7762,
      "mean_token_accuracy": 0.21516204625368118,
      "num_tokens": 39340352.0,
      "step": 21375
    },
    {
      "entropy": 5.0827325820922855,
      "epoch": 1.7858336117607752,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046806847666762305,
      "loss": 4.8271,
      "mean_token_accuracy": 0.21598578095436097,
      "num_tokens": 39349072.0,
      "step": 21380
    },
    {
      "entropy": 5.04100775718689,
      "epoch": 1.7862512529234882,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004680531877224042,
      "loss": 4.7436,
      "mean_token_accuracy": 0.2165000930428505,
      "num_tokens": 39357751.0,
      "step": 21385
    },
    {
      "entropy": 5.061936759948731,
      "epoch": 1.7866688940862012,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046803789539755686,
      "loss": 4.8206,
      "mean_token_accuracy": 0.2107674792408943,
      "num_tokens": 39366831.0,
      "step": 21390
    },
    {
      "entropy": 5.07839732170105,
      "epoch": 1.787086535248914,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046802259969334887,
      "loss": 4.7824,
      "mean_token_accuracy": 0.2153575375676155,
      "num_tokens": 39376110.0,
      "step": 21395
    },
    {
      "entropy": 5.172158288955688,
      "epoch": 1.7875041764116273,
      "grad_norm": 1.125,
      "learning_rate": 0.0004680073006100479,
      "loss": 4.9943,
      "mean_token_accuracy": 0.19898739606142044,
      "num_tokens": 39385871.0,
      "step": 21400
    },
    {
      "entropy": 5.156144380569458,
      "epoch": 1.78792181757434,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046799199814792186,
      "loss": 4.8739,
      "mean_token_accuracy": 0.21694606244564058,
      "num_tokens": 39395674.0,
      "step": 21405
    },
    {
      "entropy": 5.223722076416015,
      "epoch": 1.7883394587370531,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004679766923072386,
      "loss": 4.9702,
      "mean_token_accuracy": 0.20459388494491576,
      "num_tokens": 39405264.0,
      "step": 21410
    },
    {
      "entropy": 5.227148246765137,
      "epoch": 1.7887570998997662,
      "grad_norm": 1.171875,
      "learning_rate": 0.000467961383088266,
      "loss": 4.8579,
      "mean_token_accuracy": 0.21104829162359237,
      "num_tokens": 39414172.0,
      "step": 21415
    },
    {
      "entropy": 5.086734294891357,
      "epoch": 1.789174741062479,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046794607049127217,
      "loss": 4.7906,
      "mean_token_accuracy": 0.2219889283180237,
      "num_tokens": 39422944.0,
      "step": 21420
    },
    {
      "entropy": 5.109917211532593,
      "epoch": 1.7895923822251922,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046793075451652513,
      "loss": 4.9232,
      "mean_token_accuracy": 0.20981525927782058,
      "num_tokens": 39433468.0,
      "step": 21425
    },
    {
      "entropy": 5.076291227340699,
      "epoch": 1.790010023387905,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046791543516429295,
      "loss": 4.8423,
      "mean_token_accuracy": 0.20858058631420134,
      "num_tokens": 39442780.0,
      "step": 21430
    },
    {
      "entropy": 5.210182428359985,
      "epoch": 1.790427664550618,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046790011243484387,
      "loss": 4.9488,
      "mean_token_accuracy": 0.1991382747888565,
      "num_tokens": 39453318.0,
      "step": 21435
    },
    {
      "entropy": 5.155518198013306,
      "epoch": 1.7908453057133311,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046788478632844616,
      "loss": 4.8666,
      "mean_token_accuracy": 0.21747777312994004,
      "num_tokens": 39462688.0,
      "step": 21440
    },
    {
      "entropy": 5.02219705581665,
      "epoch": 1.7912629468760441,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046786945684536806,
      "loss": 4.7338,
      "mean_token_accuracy": 0.2213583141565323,
      "num_tokens": 39472473.0,
      "step": 21445
    },
    {
      "entropy": 5.0466289043426515,
      "epoch": 1.7916805880387572,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000467854123985878,
      "loss": 4.7938,
      "mean_token_accuracy": 0.2220352843403816,
      "num_tokens": 39481576.0,
      "step": 21450
    },
    {
      "entropy": 5.167979145050049,
      "epoch": 1.79209822920147,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004678387877502443,
      "loss": 4.8616,
      "mean_token_accuracy": 0.21819595545530318,
      "num_tokens": 39490799.0,
      "step": 21455
    },
    {
      "entropy": 5.1429696559906,
      "epoch": 1.7925158703641833,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046782344813873556,
      "loss": 4.8996,
      "mean_token_accuracy": 0.20404738038778306,
      "num_tokens": 39499552.0,
      "step": 21460
    },
    {
      "entropy": 4.977811288833618,
      "epoch": 1.792933511526896,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046780810515162027,
      "loss": 4.7745,
      "mean_token_accuracy": 0.215883307158947,
      "num_tokens": 39508461.0,
      "step": 21465
    },
    {
      "entropy": 5.051908683776856,
      "epoch": 1.793351152689609,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046779275878916694,
      "loss": 4.8303,
      "mean_token_accuracy": 0.21421982049942018,
      "num_tokens": 39516766.0,
      "step": 21470
    },
    {
      "entropy": 5.088547420501709,
      "epoch": 1.7937687938523221,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046777740905164434,
      "loss": 4.7642,
      "mean_token_accuracy": 0.21686530709266663,
      "num_tokens": 39526175.0,
      "step": 21475
    },
    {
      "entropy": 5.035620307922363,
      "epoch": 1.794186435015035,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046776205593932114,
      "loss": 4.7882,
      "mean_token_accuracy": 0.22324127107858657,
      "num_tokens": 39534690.0,
      "step": 21480
    },
    {
      "entropy": 5.158098983764648,
      "epoch": 1.7946040761777482,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046774669945246616,
      "loss": 4.9007,
      "mean_token_accuracy": 0.20324090868234634,
      "num_tokens": 39544204.0,
      "step": 21485
    },
    {
      "entropy": 5.177254915237427,
      "epoch": 1.795021717340461,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004677313395913482,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20738850384950638,
      "num_tokens": 39554620.0,
      "step": 21490
    },
    {
      "entropy": 5.055862760543823,
      "epoch": 1.795439358503174,
      "grad_norm": 1.25,
      "learning_rate": 0.0004677159763562361,
      "loss": 4.718,
      "mean_token_accuracy": 0.2200713649392128,
      "num_tokens": 39563129.0,
      "step": 21495
    },
    {
      "entropy": 5.0902180671691895,
      "epoch": 1.795856999665887,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046770060974739893,
      "loss": 4.8907,
      "mean_token_accuracy": 0.20434708446264266,
      "num_tokens": 39573645.0,
      "step": 21500
    },
    {
      "entropy": 5.123977565765381,
      "epoch": 1.7962746408286,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004676852397651055,
      "loss": 4.8337,
      "mean_token_accuracy": 0.21507720947265624,
      "num_tokens": 39582887.0,
      "step": 21505
    },
    {
      "entropy": 5.190601396560669,
      "epoch": 1.7966922819913131,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004676698664096251,
      "loss": 4.9156,
      "mean_token_accuracy": 0.2061470091342926,
      "num_tokens": 39592462.0,
      "step": 21510
    },
    {
      "entropy": 5.174539852142334,
      "epoch": 1.797109923154026,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004676544896812268,
      "loss": 4.8669,
      "mean_token_accuracy": 0.2083702027797699,
      "num_tokens": 39601576.0,
      "step": 21515
    },
    {
      "entropy": 5.1355994701385494,
      "epoch": 1.7975275643167392,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046763910958017973,
      "loss": 4.8519,
      "mean_token_accuracy": 0.21412574350833893,
      "num_tokens": 39609211.0,
      "step": 21520
    },
    {
      "entropy": 5.1162254333496096,
      "epoch": 1.797945205479452,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046762372610675314,
      "loss": 4.8432,
      "mean_token_accuracy": 0.21604853719472886,
      "num_tokens": 39617884.0,
      "step": 21525
    },
    {
      "entropy": 5.074136638641358,
      "epoch": 1.798362846642165,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046760833926121645,
      "loss": 4.82,
      "mean_token_accuracy": 0.21683358252048493,
      "num_tokens": 39627102.0,
      "step": 21530
    },
    {
      "entropy": 5.138836097717285,
      "epoch": 1.798780487804878,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046759294904383887,
      "loss": 4.8479,
      "mean_token_accuracy": 0.21278984397649764,
      "num_tokens": 39635404.0,
      "step": 21535
    },
    {
      "entropy": 5.0940995693206785,
      "epoch": 1.799198128967591,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004675775554548899,
      "loss": 4.8558,
      "mean_token_accuracy": 0.2139880582690239,
      "num_tokens": 39644016.0,
      "step": 21540
    },
    {
      "entropy": 5.036130475997925,
      "epoch": 1.7996157701303042,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046756215849463905,
      "loss": 4.8028,
      "mean_token_accuracy": 0.21865651458501817,
      "num_tokens": 39653639.0,
      "step": 21545
    },
    {
      "entropy": 5.051755046844482,
      "epoch": 1.800033411293017,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046754675816335577,
      "loss": 4.7436,
      "mean_token_accuracy": 0.2184917464852333,
      "num_tokens": 39663169.0,
      "step": 21550
    },
    {
      "entropy": 5.05431637763977,
      "epoch": 1.80045105245573,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046753135446130973,
      "loss": 4.7947,
      "mean_token_accuracy": 0.22492414116859435,
      "num_tokens": 39671687.0,
      "step": 21555
    },
    {
      "entropy": 5.095961046218872,
      "epoch": 1.800868693618443,
      "grad_norm": 1.125,
      "learning_rate": 0.0004675159473887707,
      "loss": 4.7713,
      "mean_token_accuracy": 0.21979592144489288,
      "num_tokens": 39680245.0,
      "step": 21560
    },
    {
      "entropy": 5.052632570266724,
      "epoch": 1.801286334781156,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004675005369460081,
      "loss": 4.784,
      "mean_token_accuracy": 0.21950344145298004,
      "num_tokens": 39688661.0,
      "step": 21565
    },
    {
      "entropy": 5.058454895019532,
      "epoch": 1.801703975943869,
      "grad_norm": 1.140625,
      "learning_rate": 0.000467485123133292,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2113325998187065,
      "num_tokens": 39696811.0,
      "step": 21570
    },
    {
      "entropy": 5.0906209468841555,
      "epoch": 1.802121617106582,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046746970595089207,
      "loss": 4.8533,
      "mean_token_accuracy": 0.2206322208046913,
      "num_tokens": 39705866.0,
      "step": 21575
    },
    {
      "entropy": 5.105300378799439,
      "epoch": 1.8025392582692952,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004674542853990783,
      "loss": 4.8041,
      "mean_token_accuracy": 0.21692489981651306,
      "num_tokens": 39714629.0,
      "step": 21580
    },
    {
      "entropy": 5.112765121459961,
      "epoch": 1.802956899432008,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046743886147812054,
      "loss": 4.8757,
      "mean_token_accuracy": 0.21822795122861863,
      "num_tokens": 39723527.0,
      "step": 21585
    },
    {
      "entropy": 5.06984052658081,
      "epoch": 1.803374540594721,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004674234341882889,
      "loss": 4.7695,
      "mean_token_accuracy": 0.22444554418325424,
      "num_tokens": 39732122.0,
      "step": 21590
    },
    {
      "entropy": 5.03801908493042,
      "epoch": 1.803792181757434,
      "grad_norm": 1.125,
      "learning_rate": 0.0004674080035298535,
      "loss": 4.8,
      "mean_token_accuracy": 0.21933549493551255,
      "num_tokens": 39741496.0,
      "step": 21595
    },
    {
      "entropy": 5.2005010604858395,
      "epoch": 1.8042098229201469,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004673925695030843,
      "loss": 4.9247,
      "mean_token_accuracy": 0.20905920565128328,
      "num_tokens": 39749929.0,
      "step": 21600
    },
    {
      "entropy": 5.0372833728790285,
      "epoch": 1.8046274640828601,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004673771321082517,
      "loss": 4.8036,
      "mean_token_accuracy": 0.21317977756261824,
      "num_tokens": 39758878.0,
      "step": 21605
    },
    {
      "entropy": 5.118480253219604,
      "epoch": 1.805045105245573,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004673616913456258,
      "loss": 4.8255,
      "mean_token_accuracy": 0.2100500702857971,
      "num_tokens": 39768183.0,
      "step": 21610
    },
    {
      "entropy": 5.098297548294068,
      "epoch": 1.805462746408286,
      "grad_norm": 1.125,
      "learning_rate": 0.00046734624721547687,
      "loss": 4.8206,
      "mean_token_accuracy": 0.2156327798962593,
      "num_tokens": 39776494.0,
      "step": 21615
    },
    {
      "entropy": 5.082305574417115,
      "epoch": 1.805880387570999,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046733079971807536,
      "loss": 4.7798,
      "mean_token_accuracy": 0.21904877126216887,
      "num_tokens": 39784968.0,
      "step": 21620
    },
    {
      "entropy": 5.08176326751709,
      "epoch": 1.806298028733712,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046731534885369175,
      "loss": 4.8603,
      "mean_token_accuracy": 0.21849817037582397,
      "num_tokens": 39793999.0,
      "step": 21625
    },
    {
      "entropy": 5.112294483184814,
      "epoch": 1.806715669896425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004672998946225965,
      "loss": 4.8777,
      "mean_token_accuracy": 0.21223542243242263,
      "num_tokens": 39803866.0,
      "step": 21630
    },
    {
      "entropy": 5.057772397994995,
      "epoch": 1.8071333110591379,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046728443702506007,
      "loss": 4.7306,
      "mean_token_accuracy": 0.224057874083519,
      "num_tokens": 39812001.0,
      "step": 21635
    },
    {
      "entropy": 5.106938171386719,
      "epoch": 1.8075509522218511,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046726897606135307,
      "loss": 4.8848,
      "mean_token_accuracy": 0.2148329272866249,
      "num_tokens": 39821155.0,
      "step": 21640
    },
    {
      "entropy": 5.175707149505615,
      "epoch": 1.807968593384564,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046725351173174627,
      "loss": 4.9119,
      "mean_token_accuracy": 0.21178229302167892,
      "num_tokens": 39831586.0,
      "step": 21645
    },
    {
      "entropy": 5.202748727798462,
      "epoch": 1.808386234547277,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004672380440365103,
      "loss": 5.0304,
      "mean_token_accuracy": 0.20178945809602739,
      "num_tokens": 39840556.0,
      "step": 21650
    },
    {
      "entropy": 5.2672101020812985,
      "epoch": 1.80880387570999,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000467222572975916,
      "loss": 4.9251,
      "mean_token_accuracy": 0.20614464282989503,
      "num_tokens": 39849876.0,
      "step": 21655
    },
    {
      "entropy": 5.0332354545593265,
      "epoch": 1.8092215168727028,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004672070985502342,
      "loss": 4.8249,
      "mean_token_accuracy": 0.2115745574235916,
      "num_tokens": 39859623.0,
      "step": 21660
    },
    {
      "entropy": 5.065794038772583,
      "epoch": 1.809639158035416,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046719162075973576,
      "loss": 4.7327,
      "mean_token_accuracy": 0.21747012585401534,
      "num_tokens": 39868702.0,
      "step": 21665
    },
    {
      "entropy": 5.1063347339630125,
      "epoch": 1.810056799198129,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004671761396046916,
      "loss": 4.8426,
      "mean_token_accuracy": 0.2143711805343628,
      "num_tokens": 39877772.0,
      "step": 21670
    },
    {
      "entropy": 5.035979270935059,
      "epoch": 1.810474440360842,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004671606550853729,
      "loss": 4.7926,
      "mean_token_accuracy": 0.21954498142004014,
      "num_tokens": 39886698.0,
      "step": 21675
    },
    {
      "entropy": 5.065064907073975,
      "epoch": 1.810892081523555,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046714516720205057,
      "loss": 4.7857,
      "mean_token_accuracy": 0.21802543103694916,
      "num_tokens": 39895809.0,
      "step": 21680
    },
    {
      "entropy": 5.168442010879517,
      "epoch": 1.811309722686268,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046712967595499585,
      "loss": 4.8585,
      "mean_token_accuracy": 0.22208362370729445,
      "num_tokens": 39904560.0,
      "step": 21685
    },
    {
      "entropy": 5.090098190307617,
      "epoch": 1.811727363848981,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004671141813444798,
      "loss": 4.8718,
      "mean_token_accuracy": 0.20921379029750825,
      "num_tokens": 39913942.0,
      "step": 21690
    },
    {
      "entropy": 5.156583070755005,
      "epoch": 1.8121450050116938,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046709868337077386,
      "loss": 4.9121,
      "mean_token_accuracy": 0.2094907209277153,
      "num_tokens": 39923087.0,
      "step": 21695
    },
    {
      "entropy": 5.200180768966675,
      "epoch": 1.812562646174407,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004670831820341491,
      "loss": 4.8861,
      "mean_token_accuracy": 0.21534719467163085,
      "num_tokens": 39932033.0,
      "step": 21700
    },
    {
      "entropy": 5.0957419872283936,
      "epoch": 1.81298028733712,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046706767733487714,
      "loss": 4.8653,
      "mean_token_accuracy": 0.21528517156839372,
      "num_tokens": 39941289.0,
      "step": 21705
    },
    {
      "entropy": 5.073445463180542,
      "epoch": 1.813397928499833,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046705216927322925,
      "loss": 4.781,
      "mean_token_accuracy": 0.21173635125160217,
      "num_tokens": 39949769.0,
      "step": 21710
    },
    {
      "entropy": 5.146317052841186,
      "epoch": 1.813815569662546,
      "grad_norm": 1.0625,
      "learning_rate": 0.000467036657849477,
      "loss": 4.8346,
      "mean_token_accuracy": 0.21599016785621644,
      "num_tokens": 39958774.0,
      "step": 21715
    },
    {
      "entropy": 5.069385528564453,
      "epoch": 1.8142332108252588,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004670211430638919,
      "loss": 4.8425,
      "mean_token_accuracy": 0.21828534156084062,
      "num_tokens": 39969135.0,
      "step": 21720
    },
    {
      "entropy": 5.100531959533692,
      "epoch": 1.814650851987972,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004670056249167455,
      "loss": 4.8773,
      "mean_token_accuracy": 0.20688125640153884,
      "num_tokens": 39977989.0,
      "step": 21725
    },
    {
      "entropy": 5.098042297363281,
      "epoch": 1.8150684931506849,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004669901034083096,
      "loss": 4.7598,
      "mean_token_accuracy": 0.21445304155349731,
      "num_tokens": 39986773.0,
      "step": 21730
    },
    {
      "entropy": 5.113954257965088,
      "epoch": 1.815486134313398,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046697457853885576,
      "loss": 4.8212,
      "mean_token_accuracy": 0.22151059955358504,
      "num_tokens": 39995621.0,
      "step": 21735
    },
    {
      "entropy": 5.044676113128662,
      "epoch": 1.815903775476111,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004669590503086559,
      "loss": 4.769,
      "mean_token_accuracy": 0.22250396013259888,
      "num_tokens": 40004775.0,
      "step": 21740
    },
    {
      "entropy": 5.097189950942993,
      "epoch": 1.816321416638824,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046694351871798175,
      "loss": 4.7635,
      "mean_token_accuracy": 0.2260387122631073,
      "num_tokens": 40013281.0,
      "step": 21745
    },
    {
      "entropy": 5.22846040725708,
      "epoch": 1.816739057801537,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046692798376710533,
      "loss": 5.024,
      "mean_token_accuracy": 0.20973725914955138,
      "num_tokens": 40022998.0,
      "step": 21750
    },
    {
      "entropy": 5.208833122253418,
      "epoch": 1.8171566989642498,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046691244545629847,
      "loss": 4.9582,
      "mean_token_accuracy": 0.20786502212285995,
      "num_tokens": 40032485.0,
      "step": 21755
    },
    {
      "entropy": 5.182286500930786,
      "epoch": 1.817574340126963,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004668969037858333,
      "loss": 4.8375,
      "mean_token_accuracy": 0.21602776348590852,
      "num_tokens": 40041759.0,
      "step": 21760
    },
    {
      "entropy": 5.088160276412964,
      "epoch": 1.8179919812896759,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046688135875598184,
      "loss": 4.7145,
      "mean_token_accuracy": 0.2225364163517952,
      "num_tokens": 40050151.0,
      "step": 21765
    },
    {
      "entropy": 5.087700462341308,
      "epoch": 1.818409622452389,
      "grad_norm": 1.125,
      "learning_rate": 0.0004668658103670161,
      "loss": 4.7815,
      "mean_token_accuracy": 0.20981224477291108,
      "num_tokens": 40059173.0,
      "step": 21770
    },
    {
      "entropy": 5.051149320602417,
      "epoch": 1.818827263615102,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046685025861920856,
      "loss": 4.7497,
      "mean_token_accuracy": 0.2215653657913208,
      "num_tokens": 40067868.0,
      "step": 21775
    },
    {
      "entropy": 5.1422717571258545,
      "epoch": 1.8192449047778148,
      "grad_norm": 1.25,
      "learning_rate": 0.0004668347035128312,
      "loss": 4.9038,
      "mean_token_accuracy": 0.21038708090782166,
      "num_tokens": 40078022.0,
      "step": 21780
    },
    {
      "entropy": 4.992796182632446,
      "epoch": 1.819662545940528,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046681914504815653,
      "loss": 4.6696,
      "mean_token_accuracy": 0.22181298732757568,
      "num_tokens": 40087051.0,
      "step": 21785
    },
    {
      "entropy": 5.065821027755737,
      "epoch": 1.8200801871032408,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046680358322545687,
      "loss": 4.8013,
      "mean_token_accuracy": 0.22039390057325364,
      "num_tokens": 40096368.0,
      "step": 21790
    },
    {
      "entropy": 5.103058195114135,
      "epoch": 1.8204978282659539,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004667880180450045,
      "loss": 4.8396,
      "mean_token_accuracy": 0.2156355068087578,
      "num_tokens": 40104843.0,
      "step": 21795
    },
    {
      "entropy": 5.061640119552612,
      "epoch": 1.8209154694286669,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046677244950707204,
      "loss": 4.7984,
      "mean_token_accuracy": 0.21102621853351594,
      "num_tokens": 40114653.0,
      "step": 21800
    },
    {
      "entropy": 5.049965763092041,
      "epoch": 1.82133311059138,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004667568776119321,
      "loss": 4.7901,
      "mean_token_accuracy": 0.22384050786495208,
      "num_tokens": 40124043.0,
      "step": 21805
    },
    {
      "entropy": 5.084676599502563,
      "epoch": 1.821750751754093,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004667413023598571,
      "loss": 4.827,
      "mean_token_accuracy": 0.21337041556835173,
      "num_tokens": 40133853.0,
      "step": 21810
    },
    {
      "entropy": 5.049754667282104,
      "epoch": 1.8221683929168058,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004667257237511199,
      "loss": 4.6598,
      "mean_token_accuracy": 0.22967637777328492,
      "num_tokens": 40141850.0,
      "step": 21815
    },
    {
      "entropy": 5.074651098251342,
      "epoch": 1.822586034079519,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046671014178599295,
      "loss": 4.8045,
      "mean_token_accuracy": 0.21049740612506868,
      "num_tokens": 40151375.0,
      "step": 21820
    },
    {
      "entropy": 5.080843305587768,
      "epoch": 1.8230036752422318,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046669455646474936,
      "loss": 4.824,
      "mean_token_accuracy": 0.21799491494894027,
      "num_tokens": 40161006.0,
      "step": 21825
    },
    {
      "entropy": 5.123122453689575,
      "epoch": 1.8234213164049449,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046667896778766185,
      "loss": 4.824,
      "mean_token_accuracy": 0.2170561134815216,
      "num_tokens": 40169137.0,
      "step": 21830
    },
    {
      "entropy": 5.10516324043274,
      "epoch": 1.823838957567658,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046666337575500315,
      "loss": 4.9491,
      "mean_token_accuracy": 0.20729831457138062,
      "num_tokens": 40179138.0,
      "step": 21835
    },
    {
      "entropy": 5.111067104339599,
      "epoch": 1.8242565987303707,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004666477803670464,
      "loss": 4.8442,
      "mean_token_accuracy": 0.21974292695522307,
      "num_tokens": 40187870.0,
      "step": 21840
    },
    {
      "entropy": 5.094037294387817,
      "epoch": 1.824674239893084,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004666321816240646,
      "loss": 4.7834,
      "mean_token_accuracy": 0.22254382371902465,
      "num_tokens": 40197826.0,
      "step": 21845
    },
    {
      "entropy": 5.000092458724976,
      "epoch": 1.8250918810557968,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004666165795263308,
      "loss": 4.7632,
      "mean_token_accuracy": 0.21444657146930696,
      "num_tokens": 40206503.0,
      "step": 21850
    },
    {
      "entropy": 5.05329999923706,
      "epoch": 1.8255095222185098,
      "grad_norm": 1.125,
      "learning_rate": 0.0004666009740741181,
      "loss": 4.7889,
      "mean_token_accuracy": 0.229941426217556,
      "num_tokens": 40214937.0,
      "step": 21855
    },
    {
      "entropy": 5.150661993026733,
      "epoch": 1.8259271633812229,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046658536526769976,
      "loss": 4.9137,
      "mean_token_accuracy": 0.20661329329013825,
      "num_tokens": 40224602.0,
      "step": 21860
    },
    {
      "entropy": 5.106811237335205,
      "epoch": 1.8263448045439359,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004665697531073489,
      "loss": 4.8279,
      "mean_token_accuracy": 0.20967702120542525,
      "num_tokens": 40234388.0,
      "step": 21865
    },
    {
      "entropy": 5.184930515289307,
      "epoch": 1.826762445706649,
      "grad_norm": 1.234375,
      "learning_rate": 0.000466554137593339,
      "loss": 4.8266,
      "mean_token_accuracy": 0.21283209323883057,
      "num_tokens": 40242340.0,
      "step": 21870
    },
    {
      "entropy": 5.122069311141968,
      "epoch": 1.8271800868693617,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046653851872594334,
      "loss": 4.8923,
      "mean_token_accuracy": 0.21096494942903518,
      "num_tokens": 40250618.0,
      "step": 21875
    },
    {
      "entropy": 5.136884689331055,
      "epoch": 1.827597728032075,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004665228965054353,
      "loss": 4.8387,
      "mean_token_accuracy": 0.21356207728385926,
      "num_tokens": 40259535.0,
      "step": 21880
    },
    {
      "entropy": 5.234014797210693,
      "epoch": 1.8280153691947878,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004665072709320885,
      "loss": 4.9646,
      "mean_token_accuracy": 0.20299085825681687,
      "num_tokens": 40269927.0,
      "step": 21885
    },
    {
      "entropy": 5.140410614013672,
      "epoch": 1.8284330103575008,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046649164200617633,
      "loss": 4.9519,
      "mean_token_accuracy": 0.20694604218006135,
      "num_tokens": 40280297.0,
      "step": 21890
    },
    {
      "entropy": 5.074210834503174,
      "epoch": 1.8288506515202139,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004664760097279725,
      "loss": 4.7852,
      "mean_token_accuracy": 0.21920785307884216,
      "num_tokens": 40289224.0,
      "step": 21895
    },
    {
      "entropy": 5.116854572296143,
      "epoch": 1.8292682926829267,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004664603740977506,
      "loss": 4.7743,
      "mean_token_accuracy": 0.2234537273645401,
      "num_tokens": 40298063.0,
      "step": 21900
    },
    {
      "entropy": 5.003336715698242,
      "epoch": 1.82968593384564,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046644473511578443,
      "loss": 4.7934,
      "mean_token_accuracy": 0.2216170147061348,
      "num_tokens": 40306401.0,
      "step": 21905
    },
    {
      "entropy": 5.059374141693115,
      "epoch": 1.8301035750083527,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046642909278234775,
      "loss": 4.7474,
      "mean_token_accuracy": 0.22065191119909286,
      "num_tokens": 40316127.0,
      "step": 21910
    },
    {
      "entropy": 5.005469894409179,
      "epoch": 1.8305212161710658,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004664134470977143,
      "loss": 4.7059,
      "mean_token_accuracy": 0.22223503589630128,
      "num_tokens": 40324990.0,
      "step": 21915
    },
    {
      "entropy": 5.161125612258911,
      "epoch": 1.8309388573337788,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046639779806215815,
      "loss": 4.9094,
      "mean_token_accuracy": 0.20784000903367997,
      "num_tokens": 40334209.0,
      "step": 21920
    },
    {
      "entropy": 5.1733640193939205,
      "epoch": 1.8313564984964918,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004663821456759532,
      "loss": 4.8652,
      "mean_token_accuracy": 0.21297597289085388,
      "num_tokens": 40342836.0,
      "step": 21925
    },
    {
      "entropy": 5.095439052581787,
      "epoch": 1.8317741396592049,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046636648993937326,
      "loss": 4.8361,
      "mean_token_accuracy": 0.21597236692905425,
      "num_tokens": 40351184.0,
      "step": 21930
    },
    {
      "entropy": 5.148210239410401,
      "epoch": 1.8321917808219177,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004663508308526927,
      "loss": 4.8945,
      "mean_token_accuracy": 0.20933680087327958,
      "num_tokens": 40360003.0,
      "step": 21935
    },
    {
      "entropy": 5.077456998825073,
      "epoch": 1.832609421984631,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046633516841618547,
      "loss": 4.8258,
      "mean_token_accuracy": 0.21127749532461165,
      "num_tokens": 40368344.0,
      "step": 21940
    },
    {
      "entropy": 5.118193674087524,
      "epoch": 1.8330270631473438,
      "grad_norm": 1.125,
      "learning_rate": 0.0004663195026301258,
      "loss": 4.9127,
      "mean_token_accuracy": 0.2106036961078644,
      "num_tokens": 40377075.0,
      "step": 21945
    },
    {
      "entropy": 5.225760889053345,
      "epoch": 1.8334447043100568,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004663038334947879,
      "loss": 4.9403,
      "mean_token_accuracy": 0.20702860206365586,
      "num_tokens": 40387187.0,
      "step": 21950
    },
    {
      "entropy": 5.200837326049805,
      "epoch": 1.8338623454727698,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004662881610104461,
      "loss": 4.9228,
      "mean_token_accuracy": 0.2188393160700798,
      "num_tokens": 40396588.0,
      "step": 21955
    },
    {
      "entropy": 5.010299491882324,
      "epoch": 1.8342799866354826,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000466272485177375,
      "loss": 4.7568,
      "mean_token_accuracy": 0.22810060083866118,
      "num_tokens": 40406035.0,
      "step": 21960
    },
    {
      "entropy": 5.165268421173096,
      "epoch": 1.834697627798196,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046625680599584864,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2134721279144287,
      "num_tokens": 40415840.0,
      "step": 21965
    },
    {
      "entropy": 5.12898850440979,
      "epoch": 1.8351152689609087,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046624112346614166,
      "loss": 4.7847,
      "mean_token_accuracy": 0.21562200635671616,
      "num_tokens": 40424551.0,
      "step": 21970
    },
    {
      "entropy": 5.1319934844970705,
      "epoch": 1.8355329101236217,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004662254375885287,
      "loss": 4.8073,
      "mean_token_accuracy": 0.21155655533075332,
      "num_tokens": 40434237.0,
      "step": 21975
    },
    {
      "entropy": 5.030478525161743,
      "epoch": 1.8359505512863348,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046620974836328425,
      "loss": 4.8221,
      "mean_token_accuracy": 0.21185256242752076,
      "num_tokens": 40443781.0,
      "step": 21980
    },
    {
      "entropy": 5.154204225540161,
      "epoch": 1.8363681924490478,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004661940557906829,
      "loss": 4.7696,
      "mean_token_accuracy": 0.21853094547986984,
      "num_tokens": 40452459.0,
      "step": 21985
    },
    {
      "entropy": 5.095801258087159,
      "epoch": 1.8367858336117608,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046617835987099957,
      "loss": 4.8248,
      "mean_token_accuracy": 0.2131367653608322,
      "num_tokens": 40461428.0,
      "step": 21990
    },
    {
      "entropy": 5.116941452026367,
      "epoch": 1.8372034747744737,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000466162660604509,
      "loss": 4.835,
      "mean_token_accuracy": 0.2150573804974556,
      "num_tokens": 40471265.0,
      "step": 21995
    },
    {
      "entropy": 5.062653398513794,
      "epoch": 1.837621115937187,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046614695799148575,
      "loss": 4.801,
      "mean_token_accuracy": 0.21937137097120285,
      "num_tokens": 40479624.0,
      "step": 22000
    },
    {
      "entropy": 5.055170631408691,
      "epoch": 1.8380387570998997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046613125203220506,
      "loss": 4.9446,
      "mean_token_accuracy": 0.20614700615406037,
      "num_tokens": 40489537.0,
      "step": 22005
    },
    {
      "entropy": 5.156039714813232,
      "epoch": 1.8384563982626128,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046611554272694175,
      "loss": 4.8637,
      "mean_token_accuracy": 0.21128061264753342,
      "num_tokens": 40499470.0,
      "step": 22010
    },
    {
      "entropy": 5.020924758911133,
      "epoch": 1.8388740394253258,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004660998300759707,
      "loss": 4.7187,
      "mean_token_accuracy": 0.22113583832979203,
      "num_tokens": 40507651.0,
      "step": 22015
    },
    {
      "entropy": 5.087190341949463,
      "epoch": 1.8392916805880386,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046608411407956726,
      "loss": 4.8364,
      "mean_token_accuracy": 0.22154588997364044,
      "num_tokens": 40516205.0,
      "step": 22020
    },
    {
      "entropy": 5.115843772888184,
      "epoch": 1.8397093217507519,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046606839473800626,
      "loss": 4.8282,
      "mean_token_accuracy": 0.22046646922826768,
      "num_tokens": 40524431.0,
      "step": 22025
    },
    {
      "entropy": 5.105663824081421,
      "epoch": 1.8401269629134647,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004660526720515631,
      "loss": 4.8,
      "mean_token_accuracy": 0.22577325105667115,
      "num_tokens": 40532898.0,
      "step": 22030
    },
    {
      "entropy": 5.192043685913086,
      "epoch": 1.8405446040761777,
      "grad_norm": 1.125,
      "learning_rate": 0.0004660369460205128,
      "loss": 4.9901,
      "mean_token_accuracy": 0.20147363245487213,
      "num_tokens": 40541843.0,
      "step": 22035
    },
    {
      "entropy": 5.078958463668823,
      "epoch": 1.8409622452388907,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046602121664513104,
      "loss": 4.7872,
      "mean_token_accuracy": 0.2130481481552124,
      "num_tokens": 40550687.0,
      "step": 22040
    },
    {
      "entropy": 5.152485752105713,
      "epoch": 1.8413798864016038,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046600548392569283,
      "loss": 4.841,
      "mean_token_accuracy": 0.21355855464935303,
      "num_tokens": 40560422.0,
      "step": 22045
    },
    {
      "entropy": 5.070024251937866,
      "epoch": 1.8417975275643168,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046598974786247365,
      "loss": 4.76,
      "mean_token_accuracy": 0.22027797102928162,
      "num_tokens": 40569227.0,
      "step": 22050
    },
    {
      "entropy": 5.100557422637939,
      "epoch": 1.8422151687270296,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004659740084557491,
      "loss": 4.8951,
      "mean_token_accuracy": 0.2151883661746979,
      "num_tokens": 40577805.0,
      "step": 22055
    },
    {
      "entropy": 5.028522109985351,
      "epoch": 1.8426328098897429,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004659582657057947,
      "loss": 4.7802,
      "mean_token_accuracy": 0.22183261066675186,
      "num_tokens": 40586887.0,
      "step": 22060
    },
    {
      "entropy": 5.025799655914307,
      "epoch": 1.8430504510524557,
      "grad_norm": 1.046875,
      "learning_rate": 0.000465942519612886,
      "loss": 4.7887,
      "mean_token_accuracy": 0.21889824718236922,
      "num_tokens": 40595627.0,
      "step": 22065
    },
    {
      "entropy": 5.12622299194336,
      "epoch": 1.8434680922151687,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004659267701772986,
      "loss": 4.8093,
      "mean_token_accuracy": 0.2147844687104225,
      "num_tokens": 40604568.0,
      "step": 22070
    },
    {
      "entropy": 5.0702508926391605,
      "epoch": 1.8438857333778818,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004659110173993083,
      "loss": 4.8312,
      "mean_token_accuracy": 0.21565713882446289,
      "num_tokens": 40613415.0,
      "step": 22075
    },
    {
      "entropy": 5.123496246337891,
      "epoch": 1.8443033745405946,
      "grad_norm": 1.125,
      "learning_rate": 0.0004658952612791908,
      "loss": 4.8692,
      "mean_token_accuracy": 0.21623747497797013,
      "num_tokens": 40622210.0,
      "step": 22080
    },
    {
      "entropy": 5.186977481842041,
      "epoch": 1.8447210157033078,
      "grad_norm": 1.21875,
      "learning_rate": 0.000465879501817222,
      "loss": 4.9654,
      "mean_token_accuracy": 0.21096947193145751,
      "num_tokens": 40632069.0,
      "step": 22085
    },
    {
      "entropy": 5.133398723602295,
      "epoch": 1.8451386568660206,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046586373901367774,
      "loss": 4.7464,
      "mean_token_accuracy": 0.21870831102132798,
      "num_tokens": 40640190.0,
      "step": 22090
    },
    {
      "entropy": 5.06477165222168,
      "epoch": 1.8455562980287337,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000465847972868834,
      "loss": 4.8036,
      "mean_token_accuracy": 0.21078497022390366,
      "num_tokens": 40648992.0,
      "step": 22095
    },
    {
      "entropy": 5.140925121307373,
      "epoch": 1.8459739391914467,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004658322033829668,
      "loss": 4.9426,
      "mean_token_accuracy": 0.20985675752162933,
      "num_tokens": 40658428.0,
      "step": 22100
    },
    {
      "entropy": 5.151074028015136,
      "epoch": 1.8463915803541597,
      "grad_norm": 1.25,
      "learning_rate": 0.00046581643055635215,
      "loss": 4.8953,
      "mean_token_accuracy": 0.2204982817173004,
      "num_tokens": 40667692.0,
      "step": 22105
    },
    {
      "entropy": 5.15876874923706,
      "epoch": 1.8468092215168728,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046580065438926624,
      "loss": 4.8717,
      "mean_token_accuracy": 0.2060331016778946,
      "num_tokens": 40676054.0,
      "step": 22110
    },
    {
      "entropy": 5.0782664775848385,
      "epoch": 1.8472268626795856,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004657848748819851,
      "loss": 4.8467,
      "mean_token_accuracy": 0.21502454280853273,
      "num_tokens": 40684873.0,
      "step": 22115
    },
    {
      "entropy": 5.143300294876099,
      "epoch": 1.8476445038422988,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046576909203478517,
      "loss": 4.8098,
      "mean_token_accuracy": 0.21764669865369796,
      "num_tokens": 40693571.0,
      "step": 22120
    },
    {
      "entropy": 5.081498908996582,
      "epoch": 1.8480621450050116,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046575330584794263,
      "loss": 4.8445,
      "mean_token_accuracy": 0.21530949622392653,
      "num_tokens": 40701999.0,
      "step": 22125
    },
    {
      "entropy": 5.070813226699829,
      "epoch": 1.8484797861677247,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004657375163217338,
      "loss": 4.8752,
      "mean_token_accuracy": 0.21343853026628495,
      "num_tokens": 40711240.0,
      "step": 22130
    },
    {
      "entropy": 5.137397956848145,
      "epoch": 1.8488974273304377,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046572172345643527,
      "loss": 4.8241,
      "mean_token_accuracy": 0.21306135654449462,
      "num_tokens": 40720730.0,
      "step": 22135
    },
    {
      "entropy": 5.124275064468383,
      "epoch": 1.8493150684931505,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046570592725232333,
      "loss": 4.8596,
      "mean_token_accuracy": 0.2142780065536499,
      "num_tokens": 40730754.0,
      "step": 22140
    },
    {
      "entropy": 5.190464878082276,
      "epoch": 1.8497327096558638,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004656901277096746,
      "loss": 4.945,
      "mean_token_accuracy": 0.20726808309555053,
      "num_tokens": 40741824.0,
      "step": 22145
    },
    {
      "entropy": 5.074951410293579,
      "epoch": 1.8501503508185766,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004656743248287656,
      "loss": 4.6904,
      "mean_token_accuracy": 0.22409522533416748,
      "num_tokens": 40751322.0,
      "step": 22150
    },
    {
      "entropy": 5.1472546577453615,
      "epoch": 1.8505679919812896,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004656585186098731,
      "loss": 4.9606,
      "mean_token_accuracy": 0.20183337479829788,
      "num_tokens": 40760285.0,
      "step": 22155
    },
    {
      "entropy": 5.0661396980285645,
      "epoch": 1.8509856331440027,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004656427090532737,
      "loss": 4.7884,
      "mean_token_accuracy": 0.2169583559036255,
      "num_tokens": 40770235.0,
      "step": 22160
    },
    {
      "entropy": 5.140839672088623,
      "epoch": 1.8514032743067157,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046562689615924427,
      "loss": 4.9016,
      "mean_token_accuracy": 0.21454603374004363,
      "num_tokens": 40779816.0,
      "step": 22165
    },
    {
      "entropy": 5.173488521575928,
      "epoch": 1.8518209154694287,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004656110799280615,
      "loss": 4.8289,
      "mean_token_accuracy": 0.21216772347688675,
      "num_tokens": 40788527.0,
      "step": 22170
    },
    {
      "entropy": 5.062484073638916,
      "epoch": 1.8522385566321415,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046559526036000235,
      "loss": 4.7729,
      "mean_token_accuracy": 0.2211953282356262,
      "num_tokens": 40796748.0,
      "step": 22175
    },
    {
      "entropy": 4.9889490604400635,
      "epoch": 1.8526561977948548,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046557943745534374,
      "loss": 4.7428,
      "mean_token_accuracy": 0.2161347821354866,
      "num_tokens": 40805221.0,
      "step": 22180
    },
    {
      "entropy": 5.133953666687011,
      "epoch": 1.8530738389575676,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046556361121436265,
      "loss": 4.8276,
      "mean_token_accuracy": 0.2147882416844368,
      "num_tokens": 40814576.0,
      "step": 22185
    },
    {
      "entropy": 5.146309757232666,
      "epoch": 1.8534914801202806,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046554778163733626,
      "loss": 4.801,
      "mean_token_accuracy": 0.21664714962244033,
      "num_tokens": 40822925.0,
      "step": 22190
    },
    {
      "entropy": 5.036245250701905,
      "epoch": 1.8539091212829937,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046553194872454146,
      "loss": 4.8537,
      "mean_token_accuracy": 0.21503063291311264,
      "num_tokens": 40831936.0,
      "step": 22195
    },
    {
      "entropy": 5.144589138031006,
      "epoch": 1.8543267624457065,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046551611247625565,
      "loss": 4.8849,
      "mean_token_accuracy": 0.21994175016880035,
      "num_tokens": 40841393.0,
      "step": 22200
    },
    {
      "entropy": 5.202072238922119,
      "epoch": 1.8547444036084197,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004655002728927559,
      "loss": 4.9192,
      "mean_token_accuracy": 0.2138162761926651,
      "num_tokens": 40849969.0,
      "step": 22205
    },
    {
      "entropy": 4.970247268676758,
      "epoch": 1.8551620447711326,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046548442997431966,
      "loss": 4.6949,
      "mean_token_accuracy": 0.22024947106838227,
      "num_tokens": 40859054.0,
      "step": 22210
    },
    {
      "entropy": 5.047815418243408,
      "epoch": 1.8555796859338456,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004654685837212241,
      "loss": 4.8565,
      "mean_token_accuracy": 0.21049519926309584,
      "num_tokens": 40868450.0,
      "step": 22215
    },
    {
      "entropy": 5.060461568832397,
      "epoch": 1.8559973270965586,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004654527341337468,
      "loss": 4.7531,
      "mean_token_accuracy": 0.21501994878053665,
      "num_tokens": 40877470.0,
      "step": 22220
    },
    {
      "entropy": 5.248870611190796,
      "epoch": 1.8564149682592717,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046543688121216504,
      "loss": 5.0396,
      "mean_token_accuracy": 0.201912260055542,
      "num_tokens": 40887040.0,
      "step": 22225
    },
    {
      "entropy": 5.191174411773682,
      "epoch": 1.8568326094219847,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004654210249567565,
      "loss": 4.9379,
      "mean_token_accuracy": 0.21120188981294633,
      "num_tokens": 40897090.0,
      "step": 22230
    },
    {
      "entropy": 5.261825180053711,
      "epoch": 1.8572502505846975,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004654051653677988,
      "loss": 4.9978,
      "mean_token_accuracy": 0.20286723226308823,
      "num_tokens": 40906000.0,
      "step": 22235
    },
    {
      "entropy": 5.06960973739624,
      "epoch": 1.8576678917474108,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046538930244556927,
      "loss": 4.7701,
      "mean_token_accuracy": 0.22319318652153014,
      "num_tokens": 40916356.0,
      "step": 22240
    },
    {
      "entropy": 5.110470247268677,
      "epoch": 1.8580855329101236,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046537343619034593,
      "loss": 4.8142,
      "mean_token_accuracy": 0.21880934238433838,
      "num_tokens": 40925084.0,
      "step": 22245
    },
    {
      "entropy": 5.125528573989868,
      "epoch": 1.8585031740728366,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046535756660240644,
      "loss": 4.8588,
      "mean_token_accuracy": 0.20395312309265137,
      "num_tokens": 40934353.0,
      "step": 22250
    },
    {
      "entropy": 5.036670541763305,
      "epoch": 1.8589208152355496,
      "grad_norm": 1.125,
      "learning_rate": 0.0004653416936820287,
      "loss": 4.6881,
      "mean_token_accuracy": 0.22730548977851867,
      "num_tokens": 40943617.0,
      "step": 22255
    },
    {
      "entropy": 5.124774312973022,
      "epoch": 1.8593384563982625,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004653258174294904,
      "loss": 4.9103,
      "mean_token_accuracy": 0.20644330531358718,
      "num_tokens": 40953690.0,
      "step": 22260
    },
    {
      "entropy": 5.138674354553222,
      "epoch": 1.8597560975609757,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004653099378450696,
      "loss": 4.7759,
      "mean_token_accuracy": 0.22170137315988542,
      "num_tokens": 40962475.0,
      "step": 22265
    },
    {
      "entropy": 5.072968244552612,
      "epoch": 1.8601737387236885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046529405492904427,
      "loss": 4.8211,
      "mean_token_accuracy": 0.21375028491020204,
      "num_tokens": 40971544.0,
      "step": 22270
    },
    {
      "entropy": 5.139245986938477,
      "epoch": 1.8605913798864016,
      "grad_norm": 1.125,
      "learning_rate": 0.0004652781686816925,
      "loss": 4.8985,
      "mean_token_accuracy": 0.20886510610580444,
      "num_tokens": 40980478.0,
      "step": 22275
    },
    {
      "entropy": 5.146305704116822,
      "epoch": 1.8610090210491146,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004652622791032922,
      "loss": 4.8279,
      "mean_token_accuracy": 0.2157420724630356,
      "num_tokens": 40988940.0,
      "step": 22280
    },
    {
      "entropy": 5.107742738723755,
      "epoch": 1.8614266622118276,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046524638619412187,
      "loss": 4.8231,
      "mean_token_accuracy": 0.21072943061590194,
      "num_tokens": 40997387.0,
      "step": 22285
    },
    {
      "entropy": 5.111868762969971,
      "epoch": 1.8618443033745407,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046523048995445945,
      "loss": 4.827,
      "mean_token_accuracy": 0.2111380696296692,
      "num_tokens": 41006408.0,
      "step": 22290
    },
    {
      "entropy": 5.0296050071716305,
      "epoch": 1.8622619445372535,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046521459038458327,
      "loss": 4.7821,
      "mean_token_accuracy": 0.2221483752131462,
      "num_tokens": 41015034.0,
      "step": 22295
    },
    {
      "entropy": 5.129676771163941,
      "epoch": 1.8626795856999667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046519868748477186,
      "loss": 4.8455,
      "mean_token_accuracy": 0.2144380271434784,
      "num_tokens": 41024522.0,
      "step": 22300
    },
    {
      "entropy": 5.110772562026978,
      "epoch": 1.8630972268626795,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046518278125530336,
      "loss": 4.8894,
      "mean_token_accuracy": 0.21704219728708268,
      "num_tokens": 41034497.0,
      "step": 22305
    },
    {
      "entropy": 5.231056070327758,
      "epoch": 1.8635148680253926,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004651668716964564,
      "loss": 5.0336,
      "mean_token_accuracy": 0.20305024087429047,
      "num_tokens": 41044681.0,
      "step": 22310
    },
    {
      "entropy": 5.174465703964233,
      "epoch": 1.8639325091881056,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046515095880850937,
      "loss": 4.853,
      "mean_token_accuracy": 0.20907117575407028,
      "num_tokens": 41053820.0,
      "step": 22315
    },
    {
      "entropy": 5.095247030258179,
      "epoch": 1.8643501503508184,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046513504259174104,
      "loss": 4.7629,
      "mean_token_accuracy": 0.22106999903917313,
      "num_tokens": 41062412.0,
      "step": 22320
    },
    {
      "entropy": 5.074376106262207,
      "epoch": 1.8647677915135317,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004651191230464299,
      "loss": 4.8831,
      "mean_token_accuracy": 0.22122192680835723,
      "num_tokens": 41070717.0,
      "step": 22325
    },
    {
      "entropy": 5.129698610305786,
      "epoch": 1.8651854326762445,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004651032001728546,
      "loss": 4.8756,
      "mean_token_accuracy": 0.21393635720014573,
      "num_tokens": 41080165.0,
      "step": 22330
    },
    {
      "entropy": 5.157280540466308,
      "epoch": 1.8656030738389575,
      "grad_norm": 1.125,
      "learning_rate": 0.0004650872739712941,
      "loss": 4.8455,
      "mean_token_accuracy": 0.21366207599639891,
      "num_tokens": 41089639.0,
      "step": 22335
    },
    {
      "entropy": 5.11905722618103,
      "epoch": 1.8660207150016705,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046507134444202694,
      "loss": 4.8874,
      "mean_token_accuracy": 0.20630863606929778,
      "num_tokens": 41099424.0,
      "step": 22340
    },
    {
      "entropy": 5.060626220703125,
      "epoch": 1.8664383561643836,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004650554115853321,
      "loss": 4.7174,
      "mean_token_accuracy": 0.21813855022192002,
      "num_tokens": 41108180.0,
      "step": 22345
    },
    {
      "entropy": 5.040714406967163,
      "epoch": 1.8668559973270966,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046503947540148857,
      "loss": 4.757,
      "mean_token_accuracy": 0.2261373892426491,
      "num_tokens": 41117180.0,
      "step": 22350
    },
    {
      "entropy": 4.9876524925231935,
      "epoch": 1.8672736384898094,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046502353589077523,
      "loss": 4.7081,
      "mean_token_accuracy": 0.21818677335977554,
      "num_tokens": 41125479.0,
      "step": 22355
    },
    {
      "entropy": 5.038670492172241,
      "epoch": 1.8676912796525227,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046500759305347116,
      "loss": 4.8519,
      "mean_token_accuracy": 0.2167334347963333,
      "num_tokens": 41134124.0,
      "step": 22360
    },
    {
      "entropy": 5.134174776077271,
      "epoch": 1.8681089208152355,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004649916468898555,
      "loss": 4.7713,
      "mean_token_accuracy": 0.2101711869239807,
      "num_tokens": 41143473.0,
      "step": 22365
    },
    {
      "entropy": 5.099495458602905,
      "epoch": 1.8685265619779485,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004649756974002073,
      "loss": 4.8523,
      "mean_token_accuracy": 0.20964787304401397,
      "num_tokens": 41153669.0,
      "step": 22370
    },
    {
      "entropy": 5.087727355957031,
      "epoch": 1.8689442031406616,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004649597445848059,
      "loss": 4.8241,
      "mean_token_accuracy": 0.2173657312989235,
      "num_tokens": 41163705.0,
      "step": 22375
    },
    {
      "entropy": 5.130826139450074,
      "epoch": 1.8693618443033744,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004649437884439305,
      "loss": 4.8675,
      "mean_token_accuracy": 0.21221335232257843,
      "num_tokens": 41172596.0,
      "step": 22380
    },
    {
      "entropy": 5.095294857025147,
      "epoch": 1.8697794854660876,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004649278289778604,
      "loss": 4.8452,
      "mean_token_accuracy": 0.212396739423275,
      "num_tokens": 41181869.0,
      "step": 22385
    },
    {
      "entropy": 5.011569023132324,
      "epoch": 1.8701971266288004,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046491186618687514,
      "loss": 4.7149,
      "mean_token_accuracy": 0.22152774184942245,
      "num_tokens": 41190450.0,
      "step": 22390
    },
    {
      "entropy": 5.062025499343872,
      "epoch": 1.8706147677915135,
      "grad_norm": 1.125,
      "learning_rate": 0.00046489590007125397,
      "loss": 4.8183,
      "mean_token_accuracy": 0.21918369680643082,
      "num_tokens": 41200605.0,
      "step": 22395
    },
    {
      "entropy": 5.201643371582032,
      "epoch": 1.8710324089542265,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004648799306312766,
      "loss": 4.9344,
      "mean_token_accuracy": 0.20455965548753738,
      "num_tokens": 41210764.0,
      "step": 22400
    },
    {
      "entropy": 5.226814842224121,
      "epoch": 1.8714500501169395,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046486395786722237,
      "loss": 4.9166,
      "mean_token_accuracy": 0.2078649252653122,
      "num_tokens": 41220865.0,
      "step": 22405
    },
    {
      "entropy": 5.204370832443237,
      "epoch": 1.8718676912796526,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004648479817793711,
      "loss": 4.9426,
      "mean_token_accuracy": 0.2064755916595459,
      "num_tokens": 41230622.0,
      "step": 22410
    },
    {
      "entropy": 5.080234146118164,
      "epoch": 1.8722853324423654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004648320023680023,
      "loss": 4.8778,
      "mean_token_accuracy": 0.21320890933275222,
      "num_tokens": 41239987.0,
      "step": 22415
    },
    {
      "entropy": 5.143846893310547,
      "epoch": 1.8727029736050786,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046481601963339593,
      "loss": 4.9091,
      "mean_token_accuracy": 0.20308684110641478,
      "num_tokens": 41249237.0,
      "step": 22420
    },
    {
      "entropy": 5.127580833435059,
      "epoch": 1.8731206147677915,
      "grad_norm": 1.125,
      "learning_rate": 0.00046480003357583154,
      "loss": 4.8038,
      "mean_token_accuracy": 0.21347064822912215,
      "num_tokens": 41258011.0,
      "step": 22425
    },
    {
      "entropy": 5.15661940574646,
      "epoch": 1.8735382559305045,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004647840441955892,
      "loss": 4.8363,
      "mean_token_accuracy": 0.21491085290908812,
      "num_tokens": 41267317.0,
      "step": 22430
    },
    {
      "entropy": 5.081780433654785,
      "epoch": 1.8739558970932175,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004647680514929486,
      "loss": 4.84,
      "mean_token_accuracy": 0.20791762620210646,
      "num_tokens": 41275929.0,
      "step": 22435
    },
    {
      "entropy": 5.035600852966309,
      "epoch": 1.8743735382559303,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046475205546818996,
      "loss": 4.7776,
      "mean_token_accuracy": 0.2174837589263916,
      "num_tokens": 41285076.0,
      "step": 22440
    },
    {
      "entropy": 5.094008493423462,
      "epoch": 1.8747911794186436,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004647360561215931,
      "loss": 4.8448,
      "mean_token_accuracy": 0.21133733689785003,
      "num_tokens": 41294159.0,
      "step": 22445
    },
    {
      "entropy": 4.991512680053711,
      "epoch": 1.8752088205813564,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046472005345343814,
      "loss": 4.7699,
      "mean_token_accuracy": 0.22591231912374496,
      "num_tokens": 41303363.0,
      "step": 22450
    },
    {
      "entropy": 5.123915719985962,
      "epoch": 1.8756264617440697,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004647040474640054,
      "loss": 4.7849,
      "mean_token_accuracy": 0.21887823939323425,
      "num_tokens": 41312240.0,
      "step": 22455
    },
    {
      "entropy": 5.099921655654907,
      "epoch": 1.8760441029067825,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004646880381535749,
      "loss": 4.8117,
      "mean_token_accuracy": 0.21084340959787368,
      "num_tokens": 41321377.0,
      "step": 22460
    },
    {
      "entropy": 5.099807024002075,
      "epoch": 1.8764617440694955,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000464672025522427,
      "loss": 4.8801,
      "mean_token_accuracy": 0.21756961196660995,
      "num_tokens": 41329942.0,
      "step": 22465
    },
    {
      "entropy": 5.136385917663574,
      "epoch": 1.8768793852322085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000464656009570842,
      "loss": 4.8395,
      "mean_token_accuracy": 0.21014131605625153,
      "num_tokens": 41339737.0,
      "step": 22470
    },
    {
      "entropy": 5.059055805206299,
      "epoch": 1.8772970263949214,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004646399902991002,
      "loss": 4.804,
      "mean_token_accuracy": 0.21960167586803436,
      "num_tokens": 41349432.0,
      "step": 22475
    },
    {
      "entropy": 5.079119253158569,
      "epoch": 1.8777146675576346,
      "grad_norm": 1.125,
      "learning_rate": 0.0004646239677074821,
      "loss": 4.7954,
      "mean_token_accuracy": 0.20606041103601455,
      "num_tokens": 41358628.0,
      "step": 22480
    },
    {
      "entropy": 5.228392219543457,
      "epoch": 1.8781323087203474,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046460794179626827,
      "loss": 4.9181,
      "mean_token_accuracy": 0.20637483447790145,
      "num_tokens": 41368380.0,
      "step": 22485
    },
    {
      "entropy": 5.046833038330078,
      "epoch": 1.8785499498830605,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004645919125657391,
      "loss": 4.7422,
      "mean_token_accuracy": 0.21548664420843125,
      "num_tokens": 41376923.0,
      "step": 22490
    },
    {
      "entropy": 5.079728078842163,
      "epoch": 1.8789675910457735,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004645758800161754,
      "loss": 4.8069,
      "mean_token_accuracy": 0.21571187525987626,
      "num_tokens": 41386318.0,
      "step": 22495
    },
    {
      "entropy": 5.102402830123902,
      "epoch": 1.8793852322084865,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004645598441478577,
      "loss": 4.729,
      "mean_token_accuracy": 0.22379518002271653,
      "num_tokens": 41395849.0,
      "step": 22500
    },
    {
      "entropy": 5.078215265274048,
      "epoch": 1.8798028733711996,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046454380496106665,
      "loss": 4.9162,
      "mean_token_accuracy": 0.2144307866692543,
      "num_tokens": 41404115.0,
      "step": 22505
    },
    {
      "entropy": 5.131972312927246,
      "epoch": 1.8802205145339124,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004645277624560832,
      "loss": 4.9132,
      "mean_token_accuracy": 0.20927826464176177,
      "num_tokens": 41412776.0,
      "step": 22510
    },
    {
      "entropy": 5.13053879737854,
      "epoch": 1.8806381556966256,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046451171663318815,
      "loss": 4.786,
      "mean_token_accuracy": 0.2160789519548416,
      "num_tokens": 41422180.0,
      "step": 22515
    },
    {
      "entropy": 5.16524920463562,
      "epoch": 1.8810557968593384,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046449566749266237,
      "loss": 4.9916,
      "mean_token_accuracy": 0.20623315423727034,
      "num_tokens": 41431597.0,
      "step": 22520
    },
    {
      "entropy": 5.177836799621582,
      "epoch": 1.8814734380220515,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004644796150347868,
      "loss": 4.927,
      "mean_token_accuracy": 0.20903895646333695,
      "num_tokens": 41441395.0,
      "step": 22525
    },
    {
      "entropy": 5.168490886688232,
      "epoch": 1.8818910791847645,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046446355925984257,
      "loss": 4.9015,
      "mean_token_accuracy": 0.21641702950000763,
      "num_tokens": 41450245.0,
      "step": 22530
    },
    {
      "entropy": 5.139573478698731,
      "epoch": 1.8823087203474773,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046444750016811064,
      "loss": 4.8249,
      "mean_token_accuracy": 0.20833204835653304,
      "num_tokens": 41459758.0,
      "step": 22535
    },
    {
      "entropy": 5.1063884735107425,
      "epoch": 1.8827263615101906,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004644314377598722,
      "loss": 4.8105,
      "mean_token_accuracy": 0.216145658493042,
      "num_tokens": 41468505.0,
      "step": 22540
    },
    {
      "entropy": 5.1112123966217045,
      "epoch": 1.8831440026729034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004644153720354084,
      "loss": 4.8239,
      "mean_token_accuracy": 0.21609459072351456,
      "num_tokens": 41476625.0,
      "step": 22545
    },
    {
      "entropy": 5.11238694190979,
      "epoch": 1.8835616438356164,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004643993029950005,
      "loss": 4.8897,
      "mean_token_accuracy": 0.2071613624691963,
      "num_tokens": 41485567.0,
      "step": 22550
    },
    {
      "entropy": 5.134337568283081,
      "epoch": 1.8839792849983295,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046438323063892985,
      "loss": 4.8443,
      "mean_token_accuracy": 0.2156350389122963,
      "num_tokens": 41493568.0,
      "step": 22555
    },
    {
      "entropy": 5.171909523010254,
      "epoch": 1.8843969261610425,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046436715496747774,
      "loss": 4.8137,
      "mean_token_accuracy": 0.21646583080291748,
      "num_tokens": 41502318.0,
      "step": 22560
    },
    {
      "entropy": 5.005716800689697,
      "epoch": 1.8848145673237555,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004643510759809257,
      "loss": 4.7238,
      "mean_token_accuracy": 0.23194727599620818,
      "num_tokens": 41510819.0,
      "step": 22565
    },
    {
      "entropy": 5.034723424911499,
      "epoch": 1.8852322084864683,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004643349936795551,
      "loss": 4.7793,
      "mean_token_accuracy": 0.2180106073617935,
      "num_tokens": 41519470.0,
      "step": 22570
    },
    {
      "entropy": 5.156602144241333,
      "epoch": 1.8856498496491816,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004643189080636475,
      "loss": 4.983,
      "mean_token_accuracy": 0.20536357462406157,
      "num_tokens": 41529191.0,
      "step": 22575
    },
    {
      "entropy": 5.062708759307862,
      "epoch": 1.8860674908118944,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046430281913348466,
      "loss": 4.7937,
      "mean_token_accuracy": 0.2235099509358406,
      "num_tokens": 41538764.0,
      "step": 22580
    },
    {
      "entropy": 5.082252883911133,
      "epoch": 1.8864851319746074,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004642867268893481,
      "loss": 4.7992,
      "mean_token_accuracy": 0.21916080564260482,
      "num_tokens": 41547552.0,
      "step": 22585
    },
    {
      "entropy": 4.985545063018799,
      "epoch": 1.8869027731373205,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004642706313315194,
      "loss": 4.7062,
      "mean_token_accuracy": 0.22725413888692855,
      "num_tokens": 41557341.0,
      "step": 22590
    },
    {
      "entropy": 5.132285642623901,
      "epoch": 1.8873204143000333,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046425453246028056,
      "loss": 4.8626,
      "mean_token_accuracy": 0.20760590881109237,
      "num_tokens": 41567644.0,
      "step": 22595
    },
    {
      "entropy": 5.067331695556641,
      "epoch": 1.8877380554627465,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004642384302759133,
      "loss": 4.7406,
      "mean_token_accuracy": 0.21966723054647447,
      "num_tokens": 41577301.0,
      "step": 22600
    },
    {
      "entropy": 5.1120007038116455,
      "epoch": 1.8881556966254593,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046422232477869957,
      "loss": 4.8031,
      "mean_token_accuracy": 0.2138280987739563,
      "num_tokens": 41586768.0,
      "step": 22605
    },
    {
      "entropy": 4.985317754745483,
      "epoch": 1.8885733377881724,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004642062159689213,
      "loss": 4.7016,
      "mean_token_accuracy": 0.2256334736943245,
      "num_tokens": 41596259.0,
      "step": 22610
    },
    {
      "entropy": 5.068572950363159,
      "epoch": 1.8889909789508854,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046419010384686043,
      "loss": 4.7677,
      "mean_token_accuracy": 0.2180028960108757,
      "num_tokens": 41605946.0,
      "step": 22615
    },
    {
      "entropy": 5.040532445907592,
      "epoch": 1.8894086201135984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000464173988412799,
      "loss": 4.7357,
      "mean_token_accuracy": 0.2216590717434883,
      "num_tokens": 41615139.0,
      "step": 22620
    },
    {
      "entropy": 5.191213226318359,
      "epoch": 1.8898262612763115,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046415786966701926,
      "loss": 4.9742,
      "mean_token_accuracy": 0.20127045214176179,
      "num_tokens": 41625011.0,
      "step": 22625
    },
    {
      "entropy": 5.153307342529297,
      "epoch": 1.8902439024390243,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004641417476098034,
      "loss": 4.8363,
      "mean_token_accuracy": 0.21872023940086366,
      "num_tokens": 41634072.0,
      "step": 22630
    },
    {
      "entropy": 5.105899000167847,
      "epoch": 1.8906615436017375,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004641256222414335,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2109067350625992,
      "num_tokens": 41643868.0,
      "step": 22635
    },
    {
      "entropy": 5.126273918151855,
      "epoch": 1.8910791847644504,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004641094935621919,
      "loss": 4.7658,
      "mean_token_accuracy": 0.2184901714324951,
      "num_tokens": 41652412.0,
      "step": 22640
    },
    {
      "entropy": 5.099703025817871,
      "epoch": 1.8914968259271634,
      "grad_norm": 1.203125,
      "learning_rate": 0.000464093361572361,
      "loss": 4.8431,
      "mean_token_accuracy": 0.21561671197414398,
      "num_tokens": 41662224.0,
      "step": 22645
    },
    {
      "entropy": 5.053807163238526,
      "epoch": 1.8919144670898764,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046407722627222326,
      "loss": 4.8188,
      "mean_token_accuracy": 0.22447024285793304,
      "num_tokens": 41671943.0,
      "step": 22650
    },
    {
      "entropy": 5.143688106536866,
      "epoch": 1.8923321082525892,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046406108766206103,
      "loss": 4.8671,
      "mean_token_accuracy": 0.20514324605464934,
      "num_tokens": 41680865.0,
      "step": 22655
    },
    {
      "entropy": 5.067242288589478,
      "epoch": 1.8927497494153025,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046404494574215683,
      "loss": 4.6485,
      "mean_token_accuracy": 0.23057181835174562,
      "num_tokens": 41690123.0,
      "step": 22660
    },
    {
      "entropy": 5.09148268699646,
      "epoch": 1.8931673905780153,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004640288005127933,
      "loss": 4.8457,
      "mean_token_accuracy": 0.21267474293708802,
      "num_tokens": 41698725.0,
      "step": 22665
    },
    {
      "entropy": 4.96150770187378,
      "epoch": 1.8935850317407283,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004640126519742532,
      "loss": 4.7079,
      "mean_token_accuracy": 0.2235168471932411,
      "num_tokens": 41707428.0,
      "step": 22670
    },
    {
      "entropy": 5.085896062850952,
      "epoch": 1.8940026729034414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000463996500126819,
      "loss": 4.8721,
      "mean_token_accuracy": 0.20762041062116623,
      "num_tokens": 41716560.0,
      "step": 22675
    },
    {
      "entropy": 5.139844703674316,
      "epoch": 1.8944203140661544,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004639803449707737,
      "loss": 4.9111,
      "mean_token_accuracy": 0.21244049370288848,
      "num_tokens": 41726150.0,
      "step": 22680
    },
    {
      "entropy": 5.112206697463989,
      "epoch": 1.8948379552288674,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046396418650639993,
      "loss": 4.8453,
      "mean_token_accuracy": 0.2138979136943817,
      "num_tokens": 41736493.0,
      "step": 22685
    },
    {
      "entropy": 5.1351056575775145,
      "epoch": 1.8952555963915803,
      "grad_norm": 1.125,
      "learning_rate": 0.00046394802473398065,
      "loss": 4.9509,
      "mean_token_accuracy": 0.210500368475914,
      "num_tokens": 41747565.0,
      "step": 22690
    },
    {
      "entropy": 5.101741981506348,
      "epoch": 1.8956732375542935,
      "grad_norm": 1.125,
      "learning_rate": 0.0004639318596537986,
      "loss": 4.8316,
      "mean_token_accuracy": 0.21154515892267228,
      "num_tokens": 41755641.0,
      "step": 22695
    },
    {
      "entropy": 5.211078691482544,
      "epoch": 1.8960908787170063,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000463915691266137,
      "loss": 5.0134,
      "mean_token_accuracy": 0.20288529992103577,
      "num_tokens": 41765083.0,
      "step": 22700
    },
    {
      "entropy": 5.158375072479248,
      "epoch": 1.8965085198797194,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004638995195712789,
      "loss": 4.8745,
      "mean_token_accuracy": 0.21156893521547318,
      "num_tokens": 41775160.0,
      "step": 22705
    },
    {
      "entropy": 5.024574422836304,
      "epoch": 1.8969261610424324,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004638833445695073,
      "loss": 4.6861,
      "mean_token_accuracy": 0.2295734241604805,
      "num_tokens": 41783663.0,
      "step": 22710
    },
    {
      "entropy": 5.134573841094971,
      "epoch": 1.8973438022051452,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046386716626110544,
      "loss": 4.8397,
      "mean_token_accuracy": 0.21001927256584169,
      "num_tokens": 41791974.0,
      "step": 22715
    },
    {
      "entropy": 5.072363233566284,
      "epoch": 1.8977614433678585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046385098464635645,
      "loss": 4.7844,
      "mean_token_accuracy": 0.22227238565683366,
      "num_tokens": 41800928.0,
      "step": 22720
    },
    {
      "entropy": 5.1701271533966064,
      "epoch": 1.8981790845305713,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004638347997255436,
      "loss": 4.8471,
      "mean_token_accuracy": 0.21245009303092957,
      "num_tokens": 41809512.0,
      "step": 22725
    },
    {
      "entropy": 5.095557594299317,
      "epoch": 1.8985967256932843,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004638186114989504,
      "loss": 4.7691,
      "mean_token_accuracy": 0.2288046032190323,
      "num_tokens": 41817796.0,
      "step": 22730
    },
    {
      "entropy": 5.130031442642212,
      "epoch": 1.8990143668559973,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004638024199668601,
      "loss": 4.8406,
      "mean_token_accuracy": 0.2113763391971588,
      "num_tokens": 41827250.0,
      "step": 22735
    },
    {
      "entropy": 5.135564947128296,
      "epoch": 1.8994320080187104,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004637862251295561,
      "loss": 4.8248,
      "mean_token_accuracy": 0.22563838809728623,
      "num_tokens": 41836540.0,
      "step": 22740
    },
    {
      "entropy": 5.039070892333984,
      "epoch": 1.8998496491814234,
      "grad_norm": 1.171875,
      "learning_rate": 0.000463770026987322,
      "loss": 4.7492,
      "mean_token_accuracy": 0.2189675062894821,
      "num_tokens": 41845182.0,
      "step": 22745
    },
    {
      "entropy": 5.1694965839385985,
      "epoch": 1.9002672903441362,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046375382554044136,
      "loss": 4.9314,
      "mean_token_accuracy": 0.19836183488368989,
      "num_tokens": 41853664.0,
      "step": 22750
    },
    {
      "entropy": 5.167787504196167,
      "epoch": 1.9006849315068495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004637376207891978,
      "loss": 4.768,
      "mean_token_accuracy": 0.21598386615514756,
      "num_tokens": 41862930.0,
      "step": 22755
    },
    {
      "entropy": 5.131521701812744,
      "epoch": 1.9011025726695623,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000463721412733875,
      "loss": 4.9125,
      "mean_token_accuracy": 0.21454621106386185,
      "num_tokens": 41873301.0,
      "step": 22760
    },
    {
      "entropy": 5.157780981063842,
      "epoch": 1.9015202138322753,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046370520137475673,
      "loss": 4.8362,
      "mean_token_accuracy": 0.2160434231162071,
      "num_tokens": 41882368.0,
      "step": 22765
    },
    {
      "entropy": 5.083187103271484,
      "epoch": 1.9019378549949884,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004636889867121268,
      "loss": 4.8004,
      "mean_token_accuracy": 0.2229881539940834,
      "num_tokens": 41891693.0,
      "step": 22770
    },
    {
      "entropy": 5.086151456832885,
      "epoch": 1.9023554961577012,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046367276874626894,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2193836435675621,
      "num_tokens": 41900394.0,
      "step": 22775
    },
    {
      "entropy": 5.208340883255005,
      "epoch": 1.9027731373204144,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004636565474774673,
      "loss": 4.8709,
      "mean_token_accuracy": 0.21138095259666442,
      "num_tokens": 41909836.0,
      "step": 22780
    },
    {
      "entropy": 5.123951435089111,
      "epoch": 1.9031907784831272,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046364032290600556,
      "loss": 4.8698,
      "mean_token_accuracy": 0.21581250578165054,
      "num_tokens": 41918178.0,
      "step": 22785
    },
    {
      "entropy": 5.233667182922363,
      "epoch": 1.9036084196458403,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000463624095032168,
      "loss": 5.0174,
      "mean_token_accuracy": 0.20357561409473418,
      "num_tokens": 41928827.0,
      "step": 22790
    },
    {
      "entropy": 5.177658891677856,
      "epoch": 1.9040260608085533,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046360786385623854,
      "loss": 4.8935,
      "mean_token_accuracy": 0.20707383006811142,
      "num_tokens": 41938236.0,
      "step": 22795
    },
    {
      "entropy": 5.092713499069214,
      "epoch": 1.9044437019712663,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004635916293785015,
      "loss": 4.7933,
      "mean_token_accuracy": 0.21985691487789155,
      "num_tokens": 41947086.0,
      "step": 22800
    },
    {
      "entropy": 5.085988187789917,
      "epoch": 1.9048613431339794,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004635753915992409,
      "loss": 4.8528,
      "mean_token_accuracy": 0.21169158518314363,
      "num_tokens": 41956768.0,
      "step": 22805
    },
    {
      "entropy": 5.160266876220703,
      "epoch": 1.9052789842966922,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000463559150518741,
      "loss": 4.9142,
      "mean_token_accuracy": 0.21468410044908523,
      "num_tokens": 41966956.0,
      "step": 22810
    },
    {
      "entropy": 5.1000221252441404,
      "epoch": 1.9056966254594054,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004635429061372864,
      "loss": 4.8242,
      "mean_token_accuracy": 0.22098021656274797,
      "num_tokens": 41975754.0,
      "step": 22815
    },
    {
      "entropy": 5.084816837310791,
      "epoch": 1.9061142666221182,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046352665845516113,
      "loss": 4.7476,
      "mean_token_accuracy": 0.22500994503498079,
      "num_tokens": 41983947.0,
      "step": 22820
    },
    {
      "entropy": 5.169951248168945,
      "epoch": 1.9065319077848313,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004635104074726498,
      "loss": 4.9516,
      "mean_token_accuracy": 0.2107323557138443,
      "num_tokens": 41994354.0,
      "step": 22825
    },
    {
      "entropy": 5.068253135681152,
      "epoch": 1.9069495489475443,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004634941531900369,
      "loss": 4.73,
      "mean_token_accuracy": 0.22464282363653182,
      "num_tokens": 42003023.0,
      "step": 22830
    },
    {
      "entropy": 5.059603548049926,
      "epoch": 1.9073671901102571,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046347789560760696,
      "loss": 4.8371,
      "mean_token_accuracy": 0.22200269848108292,
      "num_tokens": 42012558.0,
      "step": 22835
    },
    {
      "entropy": 5.067018508911133,
      "epoch": 1.9077848312729704,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004634616347256446,
      "loss": 4.7918,
      "mean_token_accuracy": 0.21644943654537202,
      "num_tokens": 42020298.0,
      "step": 22840
    },
    {
      "entropy": 5.118303108215332,
      "epoch": 1.9082024724356832,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046344537054443443,
      "loss": 4.8665,
      "mean_token_accuracy": 0.20785260200500488,
      "num_tokens": 42029420.0,
      "step": 22845
    },
    {
      "entropy": 5.206516170501709,
      "epoch": 1.9086201135983962,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004634291030642613,
      "loss": 4.8751,
      "mean_token_accuracy": 0.2147187441587448,
      "num_tokens": 42038311.0,
      "step": 22850
    },
    {
      "entropy": 5.134695339202881,
      "epoch": 1.9090377547611093,
      "grad_norm": 1.125,
      "learning_rate": 0.0004634128322854098,
      "loss": 4.8571,
      "mean_token_accuracy": 0.2156744748353958,
      "num_tokens": 42046806.0,
      "step": 22855
    },
    {
      "entropy": 4.922200679779053,
      "epoch": 1.9094553959238223,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004633965582081649,
      "loss": 4.6473,
      "mean_token_accuracy": 0.23595648258924484,
      "num_tokens": 42055522.0,
      "step": 22860
    },
    {
      "entropy": 5.094328117370606,
      "epoch": 1.9098730370865353,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046338028083281154,
      "loss": 4.8295,
      "mean_token_accuracy": 0.2213165983557701,
      "num_tokens": 42065363.0,
      "step": 22865
    },
    {
      "entropy": 5.149624872207641,
      "epoch": 1.9102906782492481,
      "grad_norm": 1.125,
      "learning_rate": 0.00046336400015963453,
      "loss": 4.8326,
      "mean_token_accuracy": 0.22181724905967712,
      "num_tokens": 42074355.0,
      "step": 22870
    },
    {
      "entropy": 5.072008323669434,
      "epoch": 1.9107083194119614,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000463347716188919,
      "loss": 4.7449,
      "mean_token_accuracy": 0.2181555524468422,
      "num_tokens": 42083814.0,
      "step": 22875
    },
    {
      "entropy": 5.097264051437378,
      "epoch": 1.9111259605746742,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046333142892094996,
      "loss": 4.835,
      "mean_token_accuracy": 0.21335448771715165,
      "num_tokens": 42093532.0,
      "step": 22880
    },
    {
      "entropy": 5.171812248229981,
      "epoch": 1.9115436017373872,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046331513835601263,
      "loss": 4.8549,
      "mean_token_accuracy": 0.21694796532392502,
      "num_tokens": 42102832.0,
      "step": 22885
    },
    {
      "entropy": 5.118483924865723,
      "epoch": 1.9119612429001003,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046329884449439207,
      "loss": 4.8246,
      "mean_token_accuracy": 0.21543822139501573,
      "num_tokens": 42111654.0,
      "step": 22890
    },
    {
      "entropy": 5.055588436126709,
      "epoch": 1.912378884062813,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046328254733637357,
      "loss": 4.8614,
      "mean_token_accuracy": 0.21525460630655288,
      "num_tokens": 42120916.0,
      "step": 22895
    },
    {
      "entropy": 5.242621278762817,
      "epoch": 1.9127965252255263,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004632662468822425,
      "loss": 4.9083,
      "mean_token_accuracy": 0.20686921030282973,
      "num_tokens": 42130078.0,
      "step": 22900
    },
    {
      "entropy": 5.152799463272094,
      "epoch": 1.9132141663882392,
      "grad_norm": 1.125,
      "learning_rate": 0.0004632499431322842,
      "loss": 4.7874,
      "mean_token_accuracy": 0.21787153482437133,
      "num_tokens": 42139137.0,
      "step": 22905
    },
    {
      "entropy": 5.087819957733155,
      "epoch": 1.9136318075509522,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000463233636086784,
      "loss": 4.9304,
      "mean_token_accuracy": 0.21028387397527695,
      "num_tokens": 42148474.0,
      "step": 22910
    },
    {
      "entropy": 5.018934440612793,
      "epoch": 1.9140494487136652,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046321732574602735,
      "loss": 4.8243,
      "mean_token_accuracy": 0.21241675317287445,
      "num_tokens": 42157848.0,
      "step": 22915
    },
    {
      "entropy": 5.060224151611328,
      "epoch": 1.9144670898763783,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046320101211029996,
      "loss": 4.7325,
      "mean_token_accuracy": 0.22270556837320327,
      "num_tokens": 42166117.0,
      "step": 22920
    },
    {
      "entropy": 5.052152633666992,
      "epoch": 1.9148847310390913,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046318469517988724,
      "loss": 4.7574,
      "mean_token_accuracy": 0.2177221104502678,
      "num_tokens": 42174951.0,
      "step": 22925
    },
    {
      "entropy": 5.100954961776734,
      "epoch": 1.915302372201804,
      "grad_norm": 1.203125,
      "learning_rate": 0.000463168374955075,
      "loss": 4.7666,
      "mean_token_accuracy": 0.22048801630735398,
      "num_tokens": 42183542.0,
      "step": 22930
    },
    {
      "entropy": 5.130425310134887,
      "epoch": 1.9157200133645174,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046315205143614873,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21227136850357056,
      "num_tokens": 42193361.0,
      "step": 22935
    },
    {
      "entropy": 5.148499870300293,
      "epoch": 1.9161376545272302,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046313572462339437,
      "loss": 4.8793,
      "mean_token_accuracy": 0.21634650230407715,
      "num_tokens": 42201790.0,
      "step": 22940
    },
    {
      "entropy": 5.179864454269409,
      "epoch": 1.9165552956899432,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046311939451709776,
      "loss": 4.8564,
      "mean_token_accuracy": 0.21805322617292405,
      "num_tokens": 42209743.0,
      "step": 22945
    },
    {
      "entropy": 5.139715337753296,
      "epoch": 1.9169729368526562,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004631030611175447,
      "loss": 4.7905,
      "mean_token_accuracy": 0.20970847308635712,
      "num_tokens": 42219056.0,
      "step": 22950
    },
    {
      "entropy": 5.038080549240112,
      "epoch": 1.917390578015369,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046308672442502103,
      "loss": 4.7213,
      "mean_token_accuracy": 0.2199682131409645,
      "num_tokens": 42227611.0,
      "step": 22955
    },
    {
      "entropy": 4.987605142593384,
      "epoch": 1.9178082191780823,
      "grad_norm": 1.234375,
      "learning_rate": 0.000463070384439813,
      "loss": 4.746,
      "mean_token_accuracy": 0.2312915503978729,
      "num_tokens": 42235979.0,
      "step": 22960
    },
    {
      "entropy": 5.006364011764527,
      "epoch": 1.9182258603407951,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004630540411622064,
      "loss": 4.7506,
      "mean_token_accuracy": 0.22576317191123962,
      "num_tokens": 42244889.0,
      "step": 22965
    },
    {
      "entropy": 5.038729429244995,
      "epoch": 1.9186435015035082,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004630376945924875,
      "loss": 4.8033,
      "mean_token_accuracy": 0.220181143283844,
      "num_tokens": 42253751.0,
      "step": 22970
    },
    {
      "entropy": 5.159863901138306,
      "epoch": 1.9190611426662212,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004630213447309424,
      "loss": 4.9097,
      "mean_token_accuracy": 0.21114323139190674,
      "num_tokens": 42263379.0,
      "step": 22975
    },
    {
      "entropy": 5.169176483154297,
      "epoch": 1.9194787838289342,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046300499157785725,
      "loss": 4.8463,
      "mean_token_accuracy": 0.21897483617067337,
      "num_tokens": 42272733.0,
      "step": 22980
    },
    {
      "entropy": 5.076908588409424,
      "epoch": 1.9198964249916473,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046298863513351854,
      "loss": 4.8288,
      "mean_token_accuracy": 0.2090869963169098,
      "num_tokens": 42282001.0,
      "step": 22985
    },
    {
      "entropy": 5.071258926391602,
      "epoch": 1.92031406615436,
      "grad_norm": 1.125,
      "learning_rate": 0.00046297227539821245,
      "loss": 4.7663,
      "mean_token_accuracy": 0.21555761694908143,
      "num_tokens": 42290645.0,
      "step": 22990
    },
    {
      "entropy": 5.1414306640625,
      "epoch": 1.9207317073170733,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004629559123722254,
      "loss": 4.8913,
      "mean_token_accuracy": 0.2141265705227852,
      "num_tokens": 42299606.0,
      "step": 22995
    },
    {
      "entropy": 5.101401710510254,
      "epoch": 1.9211493484797861,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046293954605584393,
      "loss": 4.7278,
      "mean_token_accuracy": 0.22137150913476944,
      "num_tokens": 42308007.0,
      "step": 23000
    },
    {
      "entropy": 5.109527969360352,
      "epoch": 1.9215669896424992,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004629231764493544,
      "loss": 4.7844,
      "mean_token_accuracy": 0.22113825380802155,
      "num_tokens": 42316495.0,
      "step": 23005
    },
    {
      "entropy": 5.177698945999145,
      "epoch": 1.9219846308052122,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004629068035530435,
      "loss": 4.895,
      "mean_token_accuracy": 0.20493163913488388,
      "num_tokens": 42325729.0,
      "step": 23010
    },
    {
      "entropy": 5.076220083236694,
      "epoch": 1.922402271967925,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004628904273671978,
      "loss": 4.7866,
      "mean_token_accuracy": 0.2151107296347618,
      "num_tokens": 42334128.0,
      "step": 23015
    },
    {
      "entropy": 5.145128917694092,
      "epoch": 1.9228199131306383,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004628740478921041,
      "loss": 4.8176,
      "mean_token_accuracy": 0.21257152110338212,
      "num_tokens": 42343177.0,
      "step": 23020
    },
    {
      "entropy": 5.160645627975464,
      "epoch": 1.923237554293351,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046285766512804894,
      "loss": 4.8457,
      "mean_token_accuracy": 0.2104414477944374,
      "num_tokens": 42352245.0,
      "step": 23025
    },
    {
      "entropy": 5.0565454959869385,
      "epoch": 1.9236551954560641,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046284127907531925,
      "loss": 4.7217,
      "mean_token_accuracy": 0.23035981506109238,
      "num_tokens": 42361175.0,
      "step": 23030
    },
    {
      "entropy": 5.120312690734863,
      "epoch": 1.9240728366187771,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046282488973420187,
      "loss": 4.972,
      "mean_token_accuracy": 0.20566562563180923,
      "num_tokens": 42370610.0,
      "step": 23035
    },
    {
      "entropy": 5.148265027999878,
      "epoch": 1.9244904777814902,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004628084971049837,
      "loss": 4.8805,
      "mean_token_accuracy": 0.21179884523153306,
      "num_tokens": 42379049.0,
      "step": 23040
    },
    {
      "entropy": 5.124570846557617,
      "epoch": 1.9249081189442032,
      "grad_norm": 1.125,
      "learning_rate": 0.00046279210118795175,
      "loss": 4.8205,
      "mean_token_accuracy": 0.2140533894300461,
      "num_tokens": 42388329.0,
      "step": 23045
    },
    {
      "entropy": 5.052695083618164,
      "epoch": 1.925325760106916,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000462775701983393,
      "loss": 4.77,
      "mean_token_accuracy": 0.22846606224775315,
      "num_tokens": 42398041.0,
      "step": 23050
    },
    {
      "entropy": 5.112186145782471,
      "epoch": 1.9257434012696293,
      "grad_norm": 1.125,
      "learning_rate": 0.0004627592994915946,
      "loss": 4.8694,
      "mean_token_accuracy": 0.21410893946886062,
      "num_tokens": 42407479.0,
      "step": 23055
    },
    {
      "entropy": 5.069595432281494,
      "epoch": 1.926161042432342,
      "grad_norm": 1.125,
      "learning_rate": 0.00046274289371284357,
      "loss": 4.7406,
      "mean_token_accuracy": 0.22168111205101013,
      "num_tokens": 42416435.0,
      "step": 23060
    },
    {
      "entropy": 5.0169322967529295,
      "epoch": 1.9265786835950551,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046272648464742733,
      "loss": 4.7876,
      "mean_token_accuracy": 0.21984623223543168,
      "num_tokens": 42425416.0,
      "step": 23065
    },
    {
      "entropy": 5.091073322296142,
      "epoch": 1.9269963247577682,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004627100722956329,
      "loss": 4.7768,
      "mean_token_accuracy": 0.2191094309091568,
      "num_tokens": 42434143.0,
      "step": 23070
    },
    {
      "entropy": 5.115967178344727,
      "epoch": 1.927413965920481,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046269365665774773,
      "loss": 4.847,
      "mean_token_accuracy": 0.21547989547252655,
      "num_tokens": 42443162.0,
      "step": 23075
    },
    {
      "entropy": 5.02263445854187,
      "epoch": 1.9278316070831942,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046267723773405915,
      "loss": 4.6843,
      "mean_token_accuracy": 0.22496939897537233,
      "num_tokens": 42451891.0,
      "step": 23080
    },
    {
      "entropy": 5.151050233840943,
      "epoch": 1.928249248245907,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046266081552485467,
      "loss": 4.866,
      "mean_token_accuracy": 0.2165645867586136,
      "num_tokens": 42461014.0,
      "step": 23085
    },
    {
      "entropy": 5.144297981262207,
      "epoch": 1.92866688940862,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046264439003042176,
      "loss": 4.8249,
      "mean_token_accuracy": 0.21189694553613664,
      "num_tokens": 42471235.0,
      "step": 23090
    },
    {
      "entropy": 5.1321056365966795,
      "epoch": 1.9290845305713331,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046262796125104785,
      "loss": 4.8205,
      "mean_token_accuracy": 0.21407441794872284,
      "num_tokens": 42480381.0,
      "step": 23095
    },
    {
      "entropy": 5.041715908050537,
      "epoch": 1.9295021717340461,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004626115291870206,
      "loss": 4.8181,
      "mean_token_accuracy": 0.21682097166776657,
      "num_tokens": 42489411.0,
      "step": 23100
    },
    {
      "entropy": 5.056657028198242,
      "epoch": 1.9299198128967592,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046259509383862784,
      "loss": 4.7043,
      "mean_token_accuracy": 0.2239028573036194,
      "num_tokens": 42498106.0,
      "step": 23105
    },
    {
      "entropy": 5.1469556331634525,
      "epoch": 1.930337454059472,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004625786552061571,
      "loss": 4.9752,
      "mean_token_accuracy": 0.20204979032278061,
      "num_tokens": 42508773.0,
      "step": 23110
    },
    {
      "entropy": 5.2137988090515135,
      "epoch": 1.9307550952221852,
      "grad_norm": 1.25,
      "learning_rate": 0.0004625622132898962,
      "loss": 4.9669,
      "mean_token_accuracy": 0.20461070835590361,
      "num_tokens": 42519368.0,
      "step": 23115
    },
    {
      "entropy": 5.260952472686768,
      "epoch": 1.931172736384898,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046254576809013297,
      "loss": 4.8852,
      "mean_token_accuracy": 0.20588243156671523,
      "num_tokens": 42529111.0,
      "step": 23120
    },
    {
      "entropy": 5.041258335113525,
      "epoch": 1.931590377547611,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004625293196071554,
      "loss": 4.7224,
      "mean_token_accuracy": 0.22711544930934907,
      "num_tokens": 42538680.0,
      "step": 23125
    },
    {
      "entropy": 5.215601444244385,
      "epoch": 1.9320080187103241,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004625128678412513,
      "loss": 5.0571,
      "mean_token_accuracy": 0.19603193998336793,
      "num_tokens": 42548239.0,
      "step": 23130
    },
    {
      "entropy": 5.158619546890259,
      "epoch": 1.932425659873037,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046249641279270866,
      "loss": 4.8801,
      "mean_token_accuracy": 0.21344099640846254,
      "num_tokens": 42557590.0,
      "step": 23135
    },
    {
      "entropy": 5.219175148010254,
      "epoch": 1.9328433010357502,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046247995446181575,
      "loss": 4.8991,
      "mean_token_accuracy": 0.20537127405405045,
      "num_tokens": 42567454.0,
      "step": 23140
    },
    {
      "entropy": 5.046689510345459,
      "epoch": 1.933260942198463,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004624634928488605,
      "loss": 4.792,
      "mean_token_accuracy": 0.21768397837877274,
      "num_tokens": 42576304.0,
      "step": 23145
    },
    {
      "entropy": 5.0691314220428465,
      "epoch": 1.933678583361176,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004624470279541312,
      "loss": 4.699,
      "mean_token_accuracy": 0.22020441591739653,
      "num_tokens": 42585861.0,
      "step": 23150
    },
    {
      "entropy": 4.988879013061523,
      "epoch": 1.934096224523889,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046243055977791597,
      "loss": 4.8069,
      "mean_token_accuracy": 0.22018093168735503,
      "num_tokens": 42595094.0,
      "step": 23155
    },
    {
      "entropy": 5.154203987121582,
      "epoch": 1.934513865686602,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004624140883205033,
      "loss": 4.8911,
      "mean_token_accuracy": 0.20997425317764282,
      "num_tokens": 42604718.0,
      "step": 23160
    },
    {
      "entropy": 5.204624223709106,
      "epoch": 1.9349315068493151,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046239761358218135,
      "loss": 4.9263,
      "mean_token_accuracy": 0.2097051963210106,
      "num_tokens": 42614324.0,
      "step": 23165
    },
    {
      "entropy": 5.0888190269470215,
      "epoch": 1.935349148012028,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004623811355632386,
      "loss": 4.7648,
      "mean_token_accuracy": 0.22128525525331497,
      "num_tokens": 42623605.0,
      "step": 23170
    },
    {
      "entropy": 5.005456495285034,
      "epoch": 1.9357667891747412,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046236465426396353,
      "loss": 4.7257,
      "mean_token_accuracy": 0.22216408848762512,
      "num_tokens": 42632381.0,
      "step": 23175
    },
    {
      "entropy": 5.120740270614624,
      "epoch": 1.936184430337454,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004623481696846446,
      "loss": 4.9259,
      "mean_token_accuracy": 0.2087822750210762,
      "num_tokens": 42642082.0,
      "step": 23180
    },
    {
      "entropy": 5.064899444580078,
      "epoch": 1.936602071500167,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004623316818255706,
      "loss": 4.7477,
      "mean_token_accuracy": 0.22436324506998062,
      "num_tokens": 42650472.0,
      "step": 23185
    },
    {
      "entropy": 5.0485231399536135,
      "epoch": 1.93701971266288,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004623151906870299,
      "loss": 4.7611,
      "mean_token_accuracy": 0.2178101882338524,
      "num_tokens": 42660412.0,
      "step": 23190
    },
    {
      "entropy": 4.996782207489014,
      "epoch": 1.937437353825593,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004622986962693113,
      "loss": 4.7457,
      "mean_token_accuracy": 0.21906332224607467,
      "num_tokens": 42669339.0,
      "step": 23195
    },
    {
      "entropy": 5.051999235153199,
      "epoch": 1.9378549949883062,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004622821985727036,
      "loss": 4.7782,
      "mean_token_accuracy": 0.21365466415882112,
      "num_tokens": 42677669.0,
      "step": 23200
    },
    {
      "entropy": 5.095537900924683,
      "epoch": 1.938272636151019,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004622656975974956,
      "loss": 4.8172,
      "mean_token_accuracy": 0.21632085889577865,
      "num_tokens": 42686609.0,
      "step": 23205
    },
    {
      "entropy": 5.0840003967285154,
      "epoch": 1.938690277313732,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046224919334397616,
      "loss": 4.8273,
      "mean_token_accuracy": 0.2174352750182152,
      "num_tokens": 42696081.0,
      "step": 23210
    },
    {
      "entropy": 5.017590856552124,
      "epoch": 1.939107918476445,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046223268581243416,
      "loss": 4.7995,
      "mean_token_accuracy": 0.22510260194540024,
      "num_tokens": 42704883.0,
      "step": 23215
    },
    {
      "entropy": 4.996682024002075,
      "epoch": 1.939525559639158,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046221617500315866,
      "loss": 4.7118,
      "mean_token_accuracy": 0.22741928845643997,
      "num_tokens": 42714057.0,
      "step": 23220
    },
    {
      "entropy": 5.0624480724334715,
      "epoch": 1.939943200801871,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046219966091643865,
      "loss": 4.8163,
      "mean_token_accuracy": 0.21959919929504396,
      "num_tokens": 42724135.0,
      "step": 23225
    },
    {
      "entropy": 5.065033864974976,
      "epoch": 1.940360841964584,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046218314355256324,
      "loss": 4.7783,
      "mean_token_accuracy": 0.2151641607284546,
      "num_tokens": 42734321.0,
      "step": 23230
    },
    {
      "entropy": 5.076207447052002,
      "epoch": 1.9407784831272972,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004621666229118216,
      "loss": 4.7106,
      "mean_token_accuracy": 0.23353231996297835,
      "num_tokens": 42742191.0,
      "step": 23235
    },
    {
      "entropy": 5.085351085662841,
      "epoch": 1.94119612429001,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046215009899450296,
      "loss": 4.8124,
      "mean_token_accuracy": 0.21575014740228654,
      "num_tokens": 42751182.0,
      "step": 23240
    },
    {
      "entropy": 5.1473472118377686,
      "epoch": 1.941613765452723,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004621335718008965,
      "loss": 4.9086,
      "mean_token_accuracy": 0.21346849501132964,
      "num_tokens": 42760867.0,
      "step": 23245
    },
    {
      "entropy": 5.062694501876831,
      "epoch": 1.942031406615436,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046211704133129164,
      "loss": 4.7833,
      "mean_token_accuracy": 0.21630997806787491,
      "num_tokens": 42769758.0,
      "step": 23250
    },
    {
      "entropy": 5.056417608261109,
      "epoch": 1.9424490477781489,
      "grad_norm": 1.125,
      "learning_rate": 0.00046210050758597776,
      "loss": 4.8742,
      "mean_token_accuracy": 0.22492533773183823,
      "num_tokens": 42778441.0,
      "step": 23255
    },
    {
      "entropy": 5.043932104110718,
      "epoch": 1.9428666889408621,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046208397056524424,
      "loss": 4.7461,
      "mean_token_accuracy": 0.2202758178114891,
      "num_tokens": 42787709.0,
      "step": 23260
    },
    {
      "entropy": 5.212898206710816,
      "epoch": 1.943284330103575,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004620674302693807,
      "loss": 4.9165,
      "mean_token_accuracy": 0.20309990346431733,
      "num_tokens": 42797783.0,
      "step": 23265
    },
    {
      "entropy": 5.173751497268677,
      "epoch": 1.943701971266288,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004620508866986765,
      "loss": 4.9104,
      "mean_token_accuracy": 0.20725826770067216,
      "num_tokens": 42807253.0,
      "step": 23270
    },
    {
      "entropy": 5.072569179534912,
      "epoch": 1.944119612429001,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046203433985342145,
      "loss": 4.8296,
      "mean_token_accuracy": 0.21375942528247832,
      "num_tokens": 42815673.0,
      "step": 23275
    },
    {
      "entropy": 5.064637804031372,
      "epoch": 1.944537253591714,
      "grad_norm": 1.125,
      "learning_rate": 0.0004620177897339051,
      "loss": 4.8362,
      "mean_token_accuracy": 0.21205582618713378,
      "num_tokens": 42825226.0,
      "step": 23280
    },
    {
      "entropy": 5.108976793289185,
      "epoch": 1.944954894754427,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004620012363404173,
      "loss": 4.8695,
      "mean_token_accuracy": 0.21222736239433287,
      "num_tokens": 42835440.0,
      "step": 23285
    },
    {
      "entropy": 5.168034887313842,
      "epoch": 1.9453725359171399,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046198467967324783,
      "loss": 4.8051,
      "mean_token_accuracy": 0.2179797351360321,
      "num_tokens": 42844479.0,
      "step": 23290
    },
    {
      "entropy": 5.070064687728882,
      "epoch": 1.9457901770798531,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004619681197326863,
      "loss": 4.7901,
      "mean_token_accuracy": 0.21847739517688752,
      "num_tokens": 42853589.0,
      "step": 23295
    },
    {
      "entropy": 5.0464332580566404,
      "epoch": 1.946207818242566,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004619515565190229,
      "loss": 4.7448,
      "mean_token_accuracy": 0.22460867017507552,
      "num_tokens": 42863061.0,
      "step": 23300
    },
    {
      "entropy": 5.030014085769653,
      "epoch": 1.946625459405279,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046193499003254737,
      "loss": 4.7787,
      "mean_token_accuracy": 0.21905454695224763,
      "num_tokens": 42872593.0,
      "step": 23305
    },
    {
      "entropy": 5.089883136749267,
      "epoch": 1.947043100567992,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004619184202735499,
      "loss": 4.8155,
      "mean_token_accuracy": 0.2151014193892479,
      "num_tokens": 42882105.0,
      "step": 23310
    },
    {
      "entropy": 5.083291864395141,
      "epoch": 1.9474607417307048,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004619018472423205,
      "loss": 4.7633,
      "mean_token_accuracy": 0.23033597022295,
      "num_tokens": 42890421.0,
      "step": 23315
    },
    {
      "entropy": 5.087955045700073,
      "epoch": 1.947878382893418,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046188527093914925,
      "loss": 4.8349,
      "mean_token_accuracy": 0.21434160619974135,
      "num_tokens": 42899142.0,
      "step": 23320
    },
    {
      "entropy": 5.144410133361816,
      "epoch": 1.948296024056131,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004618686913643264,
      "loss": 4.8741,
      "mean_token_accuracy": 0.20559311807155609,
      "num_tokens": 42908328.0,
      "step": 23325
    },
    {
      "entropy": 5.020255470275879,
      "epoch": 1.948713665218844,
      "grad_norm": 1.25,
      "learning_rate": 0.0004618521085181421,
      "loss": 4.6652,
      "mean_token_accuracy": 0.22853380888700486,
      "num_tokens": 42916824.0,
      "step": 23330
    },
    {
      "entropy": 5.273700380325318,
      "epoch": 1.949131306381557,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046183552240088683,
      "loss": 5.1906,
      "mean_token_accuracy": 0.19008521884679794,
      "num_tokens": 42927907.0,
      "step": 23335
    },
    {
      "entropy": 5.225623655319214,
      "epoch": 1.94954894754427,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046181893301285075,
      "loss": 4.8504,
      "mean_token_accuracy": 0.22004323154687883,
      "num_tokens": 42936748.0,
      "step": 23340
    },
    {
      "entropy": 5.051156616210937,
      "epoch": 1.949966588706983,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004618023403543245,
      "loss": 4.809,
      "mean_token_accuracy": 0.21531423628330232,
      "num_tokens": 42945985.0,
      "step": 23345
    },
    {
      "entropy": 5.101368093490601,
      "epoch": 1.9503842298696958,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046178574442559835,
      "loss": 4.8607,
      "mean_token_accuracy": 0.21228118985891342,
      "num_tokens": 42955762.0,
      "step": 23350
    },
    {
      "entropy": 5.042493915557861,
      "epoch": 1.950801871032409,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004617691452269629,
      "loss": 4.7738,
      "mean_token_accuracy": 0.22558300346136093,
      "num_tokens": 42964432.0,
      "step": 23355
    },
    {
      "entropy": 5.207236957550049,
      "epoch": 1.951219512195122,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004617525427587087,
      "loss": 4.9224,
      "mean_token_accuracy": 0.20816811621189119,
      "num_tokens": 42974837.0,
      "step": 23360
    },
    {
      "entropy": 5.194874238967896,
      "epoch": 1.951637153357835,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046173593702112655,
      "loss": 4.7553,
      "mean_token_accuracy": 0.22281856834888458,
      "num_tokens": 42984055.0,
      "step": 23365
    },
    {
      "entropy": 5.050923824310303,
      "epoch": 1.952054794520548,
      "grad_norm": 1.15625,
      "learning_rate": 0.000461719328014507,
      "loss": 4.9088,
      "mean_token_accuracy": 0.20912348181009294,
      "num_tokens": 42993449.0,
      "step": 23370
    },
    {
      "entropy": 5.151818418502808,
      "epoch": 1.9524724356832608,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046170271573914084,
      "loss": 4.9045,
      "mean_token_accuracy": 0.20896595418453218,
      "num_tokens": 43003011.0,
      "step": 23375
    },
    {
      "entropy": 5.19326491355896,
      "epoch": 1.952890076845974,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046168610019531894,
      "loss": 4.846,
      "mean_token_accuracy": 0.21714895814657212,
      "num_tokens": 43011409.0,
      "step": 23380
    },
    {
      "entropy": 5.060755443572998,
      "epoch": 1.9533077180086869,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004616694813833322,
      "loss": 4.7944,
      "mean_token_accuracy": 0.21538111418485642,
      "num_tokens": 43020146.0,
      "step": 23385
    },
    {
      "entropy": 5.0993999481201175,
      "epoch": 1.9537253591714,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004616528593034713,
      "loss": 4.8006,
      "mean_token_accuracy": 0.2168735146522522,
      "num_tokens": 43028925.0,
      "step": 23390
    },
    {
      "entropy": 5.165320301055909,
      "epoch": 1.954143000334113,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004616362339560276,
      "loss": 4.9395,
      "mean_token_accuracy": 0.20728831589221955,
      "num_tokens": 43038262.0,
      "step": 23395
    },
    {
      "entropy": 5.138162755966187,
      "epoch": 1.954560641496826,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046161960534129193,
      "loss": 4.8393,
      "mean_token_accuracy": 0.21116591542959212,
      "num_tokens": 43048316.0,
      "step": 23400
    },
    {
      "entropy": 5.08573088645935,
      "epoch": 1.954978282659539,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004616029734595555,
      "loss": 4.7605,
      "mean_token_accuracy": 0.2163418024778366,
      "num_tokens": 43058113.0,
      "step": 23405
    },
    {
      "entropy": 5.0388812065124515,
      "epoch": 1.9553959238222518,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004615863383111093,
      "loss": 4.7161,
      "mean_token_accuracy": 0.2319012776017189,
      "num_tokens": 43066409.0,
      "step": 23410
    },
    {
      "entropy": 4.993842935562133,
      "epoch": 1.955813564984965,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004615696998962447,
      "loss": 4.6885,
      "mean_token_accuracy": 0.2303948998451233,
      "num_tokens": 43074618.0,
      "step": 23415
    },
    {
      "entropy": 5.114896202087403,
      "epoch": 1.9562312061476779,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004615530582152529,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2147941216826439,
      "num_tokens": 43084565.0,
      "step": 23420
    },
    {
      "entropy": 5.083013296127319,
      "epoch": 1.956648847310391,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004615364132684253,
      "loss": 4.8913,
      "mean_token_accuracy": 0.21170151680707933,
      "num_tokens": 43093943.0,
      "step": 23425
    },
    {
      "entropy": 5.068285131454468,
      "epoch": 1.957066488473104,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004615197650560532,
      "loss": 4.724,
      "mean_token_accuracy": 0.22397579103708268,
      "num_tokens": 43102470.0,
      "step": 23430
    },
    {
      "entropy": 5.129438972473144,
      "epoch": 1.9574841296358167,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004615031135784282,
      "loss": 4.8973,
      "mean_token_accuracy": 0.21205626130104066,
      "num_tokens": 43111753.0,
      "step": 23435
    },
    {
      "entropy": 5.159179973602295,
      "epoch": 1.95790177079853,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046148645883584155,
      "loss": 4.9548,
      "mean_token_accuracy": 0.2063845530152321,
      "num_tokens": 43121640.0,
      "step": 23440
    },
    {
      "entropy": 5.016538047790528,
      "epoch": 1.9583194119612428,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004614698008285851,
      "loss": 4.6678,
      "mean_token_accuracy": 0.2281038224697113,
      "num_tokens": 43130181.0,
      "step": 23445
    },
    {
      "entropy": 5.016611337661743,
      "epoch": 1.9587370531239559,
      "grad_norm": 1.125,
      "learning_rate": 0.00046145313955695027,
      "loss": 4.718,
      "mean_token_accuracy": 0.22918325215578078,
      "num_tokens": 43139211.0,
      "step": 23450
    },
    {
      "entropy": 5.070492076873779,
      "epoch": 1.9591546942866689,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004614364750212288,
      "loss": 4.8652,
      "mean_token_accuracy": 0.21776935458183289,
      "num_tokens": 43148460.0,
      "step": 23455
    },
    {
      "entropy": 5.074105358123779,
      "epoch": 1.959572335449382,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046141980722171246,
      "loss": 4.8098,
      "mean_token_accuracy": 0.221587635576725,
      "num_tokens": 43156982.0,
      "step": 23460
    },
    {
      "entropy": 5.072505140304566,
      "epoch": 1.959989976612095,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046140313615869287,
      "loss": 4.7581,
      "mean_token_accuracy": 0.21552228629589082,
      "num_tokens": 43166282.0,
      "step": 23465
    },
    {
      "entropy": 5.136740493774414,
      "epoch": 1.9604076177748078,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046138646183246213,
      "loss": 4.8139,
      "mean_token_accuracy": 0.21173383593559264,
      "num_tokens": 43175237.0,
      "step": 23470
    },
    {
      "entropy": 5.154710865020752,
      "epoch": 1.960825258937521,
      "grad_norm": 1.140625,
      "learning_rate": 0.000461369784243312,
      "loss": 4.8972,
      "mean_token_accuracy": 0.2091078966856003,
      "num_tokens": 43184144.0,
      "step": 23475
    },
    {
      "entropy": 5.0866495132446286,
      "epoch": 1.9612429001002338,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046135310339153455,
      "loss": 4.7873,
      "mean_token_accuracy": 0.21624716967344285,
      "num_tokens": 43192710.0,
      "step": 23480
    },
    {
      "entropy": 5.067236852645874,
      "epoch": 1.9616605412629469,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004613364192774216,
      "loss": 4.8254,
      "mean_token_accuracy": 0.21628091037273406,
      "num_tokens": 43202404.0,
      "step": 23485
    },
    {
      "entropy": 5.09458703994751,
      "epoch": 1.96207818242566,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046131973190126536,
      "loss": 4.8845,
      "mean_token_accuracy": 0.21153393685817717,
      "num_tokens": 43211848.0,
      "step": 23490
    },
    {
      "entropy": 5.128136062622071,
      "epoch": 1.9624958235883727,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046130304126335795,
      "loss": 4.8302,
      "mean_token_accuracy": 0.22019153833389282,
      "num_tokens": 43220747.0,
      "step": 23495
    },
    {
      "entropy": 5.127920722961425,
      "epoch": 1.962913464751086,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004612863473639915,
      "loss": 4.8284,
      "mean_token_accuracy": 0.21856954842805862,
      "num_tokens": 43230972.0,
      "step": 23500
    },
    {
      "entropy": 5.097757148742676,
      "epoch": 1.9633311059137988,
      "grad_norm": 1.125,
      "learning_rate": 0.00046126965020345834,
      "loss": 4.8231,
      "mean_token_accuracy": 0.21511345803737641,
      "num_tokens": 43240512.0,
      "step": 23505
    },
    {
      "entropy": 5.104374265670776,
      "epoch": 1.9637487470765118,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004612529497820508,
      "loss": 4.8221,
      "mean_token_accuracy": 0.2125831052660942,
      "num_tokens": 43249526.0,
      "step": 23510
    },
    {
      "entropy": 5.062287473678589,
      "epoch": 1.9641663882392248,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004612362461000611,
      "loss": 4.7963,
      "mean_token_accuracy": 0.2130104511976242,
      "num_tokens": 43257913.0,
      "step": 23515
    },
    {
      "entropy": 5.024948072433472,
      "epoch": 1.9645840294019379,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046121953915778175,
      "loss": 4.772,
      "mean_token_accuracy": 0.22096486985683442,
      "num_tokens": 43266463.0,
      "step": 23520
    },
    {
      "entropy": 5.14390377998352,
      "epoch": 1.965001670564651,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004612028289555052,
      "loss": 4.7771,
      "mean_token_accuracy": 0.22172381579875947,
      "num_tokens": 43275608.0,
      "step": 23525
    },
    {
      "entropy": 5.103160858154297,
      "epoch": 1.9654193117273637,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046118611549352407,
      "loss": 4.8346,
      "mean_token_accuracy": 0.21415783017873763,
      "num_tokens": 43285332.0,
      "step": 23530
    },
    {
      "entropy": 5.126321935653687,
      "epoch": 1.965836952890077,
      "grad_norm": 1.125,
      "learning_rate": 0.0004611693987721308,
      "loss": 4.9088,
      "mean_token_accuracy": 0.21334072649478913,
      "num_tokens": 43294667.0,
      "step": 23535
    },
    {
      "entropy": 5.191002273559571,
      "epoch": 1.9662545940527898,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046115267879161823,
      "loss": 4.8367,
      "mean_token_accuracy": 0.2161051258444786,
      "num_tokens": 43304119.0,
      "step": 23540
    },
    {
      "entropy": 5.129081726074219,
      "epoch": 1.9666722352155028,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046113595555227885,
      "loss": 4.852,
      "mean_token_accuracy": 0.2182842344045639,
      "num_tokens": 43312371.0,
      "step": 23545
    },
    {
      "entropy": 5.09284520149231,
      "epoch": 1.9670898763782159,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004611192290544055,
      "loss": 4.911,
      "mean_token_accuracy": 0.21301385313272475,
      "num_tokens": 43321767.0,
      "step": 23550
    },
    {
      "entropy": 5.166186285018921,
      "epoch": 1.9675075175409287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046110249929829104,
      "loss": 4.8294,
      "mean_token_accuracy": 0.22284647524356843,
      "num_tokens": 43330944.0,
      "step": 23555
    },
    {
      "entropy": 5.109782361984253,
      "epoch": 1.967925158703642,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004610857662842283,
      "loss": 4.7856,
      "mean_token_accuracy": 0.21181908696889878,
      "num_tokens": 43339995.0,
      "step": 23560
    },
    {
      "entropy": 5.10671648979187,
      "epoch": 1.9683427998663547,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004610690300125103,
      "loss": 4.7722,
      "mean_token_accuracy": 0.21955495923757554,
      "num_tokens": 43348890.0,
      "step": 23565
    },
    {
      "entropy": 5.011828184127808,
      "epoch": 1.9687604410290678,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046105229048342994,
      "loss": 4.7513,
      "mean_token_accuracy": 0.22358892261981964,
      "num_tokens": 43357852.0,
      "step": 23570
    },
    {
      "entropy": 5.013255882263183,
      "epoch": 1.9691780821917808,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046103554769728026,
      "loss": 4.7399,
      "mean_token_accuracy": 0.21745948642492294,
      "num_tokens": 43367786.0,
      "step": 23575
    },
    {
      "entropy": 5.171214914321899,
      "epoch": 1.9695957233544938,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004610188016543544,
      "loss": 4.9101,
      "mean_token_accuracy": 0.21016066521406174,
      "num_tokens": 43377444.0,
      "step": 23580
    },
    {
      "entropy": 5.0969428539276125,
      "epoch": 1.9700133645172069,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046100205235494554,
      "loss": 4.7744,
      "mean_token_accuracy": 0.2248961105942726,
      "num_tokens": 43385885.0,
      "step": 23585
    },
    {
      "entropy": 5.076955699920655,
      "epoch": 1.9704310056799197,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046098529979934677,
      "loss": 4.8495,
      "mean_token_accuracy": 0.21212347894906997,
      "num_tokens": 43395225.0,
      "step": 23590
    },
    {
      "entropy": 5.1246436595916744,
      "epoch": 1.970848646842633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004609685439878515,
      "loss": 4.8326,
      "mean_token_accuracy": 0.2133474603295326,
      "num_tokens": 43404407.0,
      "step": 23595
    },
    {
      "entropy": 5.153379774093628,
      "epoch": 1.9712662880053458,
      "grad_norm": 1.109375,
      "learning_rate": 0.000460951784920753,
      "loss": 4.8494,
      "mean_token_accuracy": 0.2073942169547081,
      "num_tokens": 43414046.0,
      "step": 23600
    },
    {
      "entropy": 5.094531059265137,
      "epoch": 1.9716839291680588,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004609350225983447,
      "loss": 4.7581,
      "mean_token_accuracy": 0.21485242992639542,
      "num_tokens": 43424092.0,
      "step": 23605
    },
    {
      "entropy": 5.081820392608643,
      "epoch": 1.9721015703307718,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046091825702092,
      "loss": 4.7455,
      "mean_token_accuracy": 0.2169409826397896,
      "num_tokens": 43434215.0,
      "step": 23610
    },
    {
      "entropy": 5.131243324279785,
      "epoch": 1.9725192114934846,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046090148818877244,
      "loss": 4.8665,
      "mean_token_accuracy": 0.209820020198822,
      "num_tokens": 43443676.0,
      "step": 23615
    },
    {
      "entropy": 5.059386110305786,
      "epoch": 1.972936852656198,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004608847161021955,
      "loss": 4.7756,
      "mean_token_accuracy": 0.22310948967933655,
      "num_tokens": 43453240.0,
      "step": 23620
    },
    {
      "entropy": 5.037251615524292,
      "epoch": 1.9733544938189107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000460867940761483,
      "loss": 4.7928,
      "mean_token_accuracy": 0.21104137152433394,
      "num_tokens": 43462719.0,
      "step": 23625
    },
    {
      "entropy": 5.117337417602539,
      "epoch": 1.9737721349816237,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004608511621669283,
      "loss": 4.7987,
      "mean_token_accuracy": 0.21520818918943405,
      "num_tokens": 43472097.0,
      "step": 23630
    },
    {
      "entropy": 5.030080890655517,
      "epoch": 1.9741897761443368,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046083438031882534,
      "loss": 4.744,
      "mean_token_accuracy": 0.22228530645370484,
      "num_tokens": 43481014.0,
      "step": 23635
    },
    {
      "entropy": 4.9992071151733395,
      "epoch": 1.9746074173070498,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004608175952174679,
      "loss": 4.7395,
      "mean_token_accuracy": 0.22581353783607483,
      "num_tokens": 43489415.0,
      "step": 23640
    },
    {
      "entropy": 5.122688055038452,
      "epoch": 1.9750250584697628,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004608008068631498,
      "loss": 4.8887,
      "mean_token_accuracy": 0.2175940066576004,
      "num_tokens": 43499252.0,
      "step": 23645
    },
    {
      "entropy": 5.158153533935547,
      "epoch": 1.9754426996324757,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046078401525616493,
      "loss": 4.8655,
      "mean_token_accuracy": 0.21221470534801484,
      "num_tokens": 43508391.0,
      "step": 23650
    },
    {
      "entropy": 5.288312005996704,
      "epoch": 1.975860340795189,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004607672203968072,
      "loss": 5.0561,
      "mean_token_accuracy": 0.19970345497131348,
      "num_tokens": 43518536.0,
      "step": 23655
    },
    {
      "entropy": 5.140969133377075,
      "epoch": 1.9762779819579017,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046075042228537074,
      "loss": 4.919,
      "mean_token_accuracy": 0.21012751758098602,
      "num_tokens": 43528079.0,
      "step": 23660
    },
    {
      "entropy": 5.121465539932251,
      "epoch": 1.9766956231206148,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004607336209221494,
      "loss": 4.8934,
      "mean_token_accuracy": 0.21396371275186538,
      "num_tokens": 43537150.0,
      "step": 23665
    },
    {
      "entropy": 5.179365587234497,
      "epoch": 1.9771132642833278,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004607168163074376,
      "loss": 4.8784,
      "mean_token_accuracy": 0.21173638254404067,
      "num_tokens": 43546858.0,
      "step": 23670
    },
    {
      "entropy": 5.0407532215118405,
      "epoch": 1.9775309054460406,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004607000084415294,
      "loss": 4.7503,
      "mean_token_accuracy": 0.22333619892597198,
      "num_tokens": 43556091.0,
      "step": 23675
    },
    {
      "entropy": 5.042842435836792,
      "epoch": 1.9779485466087539,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046068319732471894,
      "loss": 4.8187,
      "mean_token_accuracy": 0.22339455485343934,
      "num_tokens": 43565467.0,
      "step": 23680
    },
    {
      "entropy": 5.0732628345489506,
      "epoch": 1.9783661877714667,
      "grad_norm": 1.125,
      "learning_rate": 0.0004606663829573006,
      "loss": 4.7326,
      "mean_token_accuracy": 0.22098861932754515,
      "num_tokens": 43574860.0,
      "step": 23685
    },
    {
      "entropy": 5.071172285079956,
      "epoch": 1.9787838289341797,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004606495653395688,
      "loss": 4.8519,
      "mean_token_accuracy": 0.2206648603081703,
      "num_tokens": 43584514.0,
      "step": 23690
    },
    {
      "entropy": 5.054350423812866,
      "epoch": 1.9792014700968927,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004606327444718179,
      "loss": 4.7168,
      "mean_token_accuracy": 0.2161804184317589,
      "num_tokens": 43593805.0,
      "step": 23695
    },
    {
      "entropy": 5.065666151046753,
      "epoch": 1.9796191112596058,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004606159203543423,
      "loss": 4.7634,
      "mean_token_accuracy": 0.2226391151547432,
      "num_tokens": 43602757.0,
      "step": 23700
    },
    {
      "entropy": 5.104462623596191,
      "epoch": 1.9800367524223188,
      "grad_norm": 1.125,
      "learning_rate": 0.0004605990929874366,
      "loss": 4.8336,
      "mean_token_accuracy": 0.21691469103097916,
      "num_tokens": 43612008.0,
      "step": 23705
    },
    {
      "entropy": 5.141250514984131,
      "epoch": 1.9804543935850316,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004605822623713954,
      "loss": 4.885,
      "mean_token_accuracy": 0.21985891759395598,
      "num_tokens": 43621596.0,
      "step": 23710
    },
    {
      "entropy": 5.167520952224732,
      "epoch": 1.9808720347477449,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046056542850651333,
      "loss": 4.9032,
      "mean_token_accuracy": 0.20809148997068405,
      "num_tokens": 43630725.0,
      "step": 23715
    },
    {
      "entropy": 5.090035486221313,
      "epoch": 1.9812896759104577,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000460548591393085,
      "loss": 4.8909,
      "mean_token_accuracy": 0.21180587410926818,
      "num_tokens": 43639560.0,
      "step": 23720
    },
    {
      "entropy": 5.10419282913208,
      "epoch": 1.9817073170731707,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046053175103140533,
      "loss": 4.8109,
      "mean_token_accuracy": 0.21690985709428787,
      "num_tokens": 43648840.0,
      "step": 23725
    },
    {
      "entropy": 5.059973192214966,
      "epoch": 1.9821249582358837,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046051490742176894,
      "loss": 4.7544,
      "mean_token_accuracy": 0.22549170553684234,
      "num_tokens": 43657714.0,
      "step": 23730
    },
    {
      "entropy": 5.114449882507325,
      "epoch": 1.9825425993985966,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004604980605644708,
      "loss": 4.857,
      "mean_token_accuracy": 0.21850906312465668,
      "num_tokens": 43667230.0,
      "step": 23735
    },
    {
      "entropy": 5.095452070236206,
      "epoch": 1.9829602405613098,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004604812104598059,
      "loss": 4.8096,
      "mean_token_accuracy": 0.2222469449043274,
      "num_tokens": 43676347.0,
      "step": 23740
    },
    {
      "entropy": 5.0969836711883545,
      "epoch": 1.9833778817240226,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004604643571080691,
      "loss": 4.7952,
      "mean_token_accuracy": 0.22204215228557586,
      "num_tokens": 43684824.0,
      "step": 23745
    },
    {
      "entropy": 5.082304191589356,
      "epoch": 1.9837955228867357,
      "grad_norm": 1.125,
      "learning_rate": 0.00046044750050955546,
      "loss": 4.8737,
      "mean_token_accuracy": 0.21575431227684022,
      "num_tokens": 43693995.0,
      "step": 23750
    },
    {
      "entropy": 5.198403072357178,
      "epoch": 1.9842131640494487,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046043064066456015,
      "loss": 4.8133,
      "mean_token_accuracy": 0.21286311596632004,
      "num_tokens": 43702781.0,
      "step": 23755
    },
    {
      "entropy": 5.076531076431275,
      "epoch": 1.9846308052121617,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046041377757337826,
      "loss": 4.7523,
      "mean_token_accuracy": 0.2218612626194954,
      "num_tokens": 43711387.0,
      "step": 23760
    },
    {
      "entropy": 5.0204404354095455,
      "epoch": 1.9850484463748748,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046039691123630496,
      "loss": 4.8091,
      "mean_token_accuracy": 0.2184247612953186,
      "num_tokens": 43721016.0,
      "step": 23765
    },
    {
      "entropy": 5.006798934936524,
      "epoch": 1.9854660875375876,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046038004165363554,
      "loss": 4.7665,
      "mean_token_accuracy": 0.22335927188396454,
      "num_tokens": 43729874.0,
      "step": 23770
    },
    {
      "entropy": 5.091734266281128,
      "epoch": 1.9858837287003008,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046036316882566545,
      "loss": 4.7953,
      "mean_token_accuracy": 0.21907998323440553,
      "num_tokens": 43738245.0,
      "step": 23775
    },
    {
      "entropy": 5.1277776718139645,
      "epoch": 1.9863013698630136,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046034629275268997,
      "loss": 4.7627,
      "mean_token_accuracy": 0.21880977004766464,
      "num_tokens": 43746891.0,
      "step": 23780
    },
    {
      "entropy": 5.09840579032898,
      "epoch": 1.9867190110257267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046032941343500443,
      "loss": 4.8998,
      "mean_token_accuracy": 0.21013183742761612,
      "num_tokens": 43755509.0,
      "step": 23785
    },
    {
      "entropy": 5.104866218566895,
      "epoch": 1.9871366521884397,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046031253087290455,
      "loss": 4.8301,
      "mean_token_accuracy": 0.21243332028388978,
      "num_tokens": 43764329.0,
      "step": 23790
    },
    {
      "entropy": 5.061763668060303,
      "epoch": 1.9875542933511525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046029564506668567,
      "loss": 4.7605,
      "mean_token_accuracy": 0.22213039696216583,
      "num_tokens": 43773588.0,
      "step": 23795
    },
    {
      "entropy": 5.046582555770874,
      "epoch": 1.9879719345138658,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046027875601664355,
      "loss": 4.7706,
      "mean_token_accuracy": 0.22307678461074829,
      "num_tokens": 43781980.0,
      "step": 23800
    },
    {
      "entropy": 5.135012245178222,
      "epoch": 1.9883895756765786,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004602618637230737,
      "loss": 4.7923,
      "mean_token_accuracy": 0.2158823937177658,
      "num_tokens": 43791810.0,
      "step": 23805
    },
    {
      "entropy": 5.067705583572388,
      "epoch": 1.9888072168392918,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000460244968186272,
      "loss": 4.7634,
      "mean_token_accuracy": 0.2253384470939636,
      "num_tokens": 43800866.0,
      "step": 23810
    },
    {
      "entropy": 5.109630823135376,
      "epoch": 1.9892248580020047,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046022806940653415,
      "loss": 4.8623,
      "mean_token_accuracy": 0.2128104791045189,
      "num_tokens": 43811247.0,
      "step": 23815
    },
    {
      "entropy": 5.09610686302185,
      "epoch": 1.9896424991647177,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000460211167384156,
      "loss": 4.7814,
      "mean_token_accuracy": 0.21330292373895646,
      "num_tokens": 43819740.0,
      "step": 23820
    },
    {
      "entropy": 5.017685270309448,
      "epoch": 1.9900601403274307,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046019426211943334,
      "loss": 4.7439,
      "mean_token_accuracy": 0.21566085517406464,
      "num_tokens": 43828662.0,
      "step": 23825
    },
    {
      "entropy": 5.1192974090576175,
      "epoch": 1.9904777814901435,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004601773536126623,
      "loss": 4.836,
      "mean_token_accuracy": 0.20676441192626954,
      "num_tokens": 43838143.0,
      "step": 23830
    },
    {
      "entropy": 5.136837673187256,
      "epoch": 1.9908954226528568,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004601604418641387,
      "loss": 4.7783,
      "mean_token_accuracy": 0.21766939759254456,
      "num_tokens": 43847582.0,
      "step": 23835
    },
    {
      "entropy": 5.081834697723389,
      "epoch": 1.9913130638155696,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046014352687415866,
      "loss": 4.8156,
      "mean_token_accuracy": 0.2094303011894226,
      "num_tokens": 43856992.0,
      "step": 23840
    },
    {
      "entropy": 5.0467997074127195,
      "epoch": 1.9917307049782826,
      "grad_norm": 1.125,
      "learning_rate": 0.00046012660864301846,
      "loss": 4.7554,
      "mean_token_accuracy": 0.22319515496492387,
      "num_tokens": 43866217.0,
      "step": 23845
    },
    {
      "entropy": 5.064989233016968,
      "epoch": 1.9921483461409957,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046010968717101404,
      "loss": 4.8778,
      "mean_token_accuracy": 0.20834244787693024,
      "num_tokens": 43875107.0,
      "step": 23850
    },
    {
      "entropy": 5.023962688446045,
      "epoch": 1.9925659873037087,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046009276245844176,
      "loss": 4.7737,
      "mean_token_accuracy": 0.22396346032619477,
      "num_tokens": 43884189.0,
      "step": 23855
    },
    {
      "entropy": 5.090383863449096,
      "epoch": 1.9929836284664217,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004600758345055979,
      "loss": 4.7389,
      "mean_token_accuracy": 0.2218128353357315,
      "num_tokens": 43893356.0,
      "step": 23860
    },
    {
      "entropy": 5.111717796325683,
      "epoch": 1.9934012696291346,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046005890331277875,
      "loss": 4.8115,
      "mean_token_accuracy": 0.21627305895090104,
      "num_tokens": 43902288.0,
      "step": 23865
    },
    {
      "entropy": 4.963195848464966,
      "epoch": 1.9938189107918478,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004600419688802807,
      "loss": 4.6792,
      "mean_token_accuracy": 0.2285003185272217,
      "num_tokens": 43910940.0,
      "step": 23870
    },
    {
      "entropy": 5.147058296203613,
      "epoch": 1.9942365519545606,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046002503120840037,
      "loss": 4.909,
      "mean_token_accuracy": 0.2083713412284851,
      "num_tokens": 43919899.0,
      "step": 23875
    },
    {
      "entropy": 5.1353593349456785,
      "epoch": 1.9946541931172737,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000460008090297434,
      "loss": 4.8535,
      "mean_token_accuracy": 0.21396878361701965,
      "num_tokens": 43930361.0,
      "step": 23880
    },
    {
      "entropy": 4.960782384872436,
      "epoch": 1.9950718342799867,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045999114614767847,
      "loss": 4.6004,
      "mean_token_accuracy": 0.23124884068965912,
      "num_tokens": 43939171.0,
      "step": 23885
    },
    {
      "entropy": 5.0088263034820555,
      "epoch": 1.9954894754426995,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004599741987594302,
      "loss": 4.8399,
      "mean_token_accuracy": 0.2086770012974739,
      "num_tokens": 43949224.0,
      "step": 23890
    },
    {
      "entropy": 5.099758338928223,
      "epoch": 1.9959071166054128,
      "grad_norm": 1.25,
      "learning_rate": 0.0004599572481329859,
      "loss": 4.7928,
      "mean_token_accuracy": 0.2251793012022972,
      "num_tokens": 43958379.0,
      "step": 23895
    },
    {
      "entropy": 5.158888292312622,
      "epoch": 1.9963247577681256,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004599402942686423,
      "loss": 4.8741,
      "mean_token_accuracy": 0.2107936128973961,
      "num_tokens": 43967161.0,
      "step": 23900
    },
    {
      "entropy": 5.125488948822022,
      "epoch": 1.9967423989308386,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004599233371666963,
      "loss": 4.8388,
      "mean_token_accuracy": 0.2242676466703415,
      "num_tokens": 43975943.0,
      "step": 23905
    },
    {
      "entropy": 5.034035539627075,
      "epoch": 1.9971600400935516,
      "grad_norm": 1.125,
      "learning_rate": 0.0004599063768274447,
      "loss": 4.7499,
      "mean_token_accuracy": 0.2178575038909912,
      "num_tokens": 43985510.0,
      "step": 23910
    },
    {
      "entropy": 5.0844379425048825,
      "epoch": 1.9975776812562647,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004598894132511843,
      "loss": 4.8348,
      "mean_token_accuracy": 0.21325312405824662,
      "num_tokens": 43994936.0,
      "step": 23915
    },
    {
      "entropy": 5.059509038925171,
      "epoch": 1.9979953224189777,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004598724464382123,
      "loss": 4.7899,
      "mean_token_accuracy": 0.2162877544760704,
      "num_tokens": 44003989.0,
      "step": 23920
    },
    {
      "entropy": 5.036414957046508,
      "epoch": 1.9984129635816905,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045985547638882553,
      "loss": 4.7489,
      "mean_token_accuracy": 0.22508286982774733,
      "num_tokens": 44012651.0,
      "step": 23925
    },
    {
      "entropy": 5.1153096675872805,
      "epoch": 1.9988306047444038,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045983850310332117,
      "loss": 4.8286,
      "mean_token_accuracy": 0.22592005431652068,
      "num_tokens": 44021287.0,
      "step": 23930
    },
    {
      "entropy": 5.086086654663086,
      "epoch": 1.9992482459071166,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004598215265819963,
      "loss": 4.8469,
      "mean_token_accuracy": 0.2135340690612793,
      "num_tokens": 44030917.0,
      "step": 23935
    },
    {
      "entropy": 5.072421503067017,
      "epoch": 1.9996658870698296,
      "grad_norm": 1.125,
      "learning_rate": 0.0004598045468251481,
      "loss": 4.825,
      "mean_token_accuracy": 0.21529478579759598,
      "num_tokens": 44040465.0,
      "step": 23940
    },
    {
      "entropy": 5.1055474281311035,
      "epoch": 2.0000835282325427,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004597875638330739,
      "loss": 4.7729,
      "mean_token_accuracy": 0.21463800519704818,
      "num_tokens": 44050358.0,
      "step": 23945
    },
    {
      "entropy": 5.05221471786499,
      "epoch": 2.0005011693952555,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045977057760607096,
      "loss": 4.6612,
      "mean_token_accuracy": 0.22703286111354828,
      "num_tokens": 44059459.0,
      "step": 23950
    },
    {
      "entropy": 5.139415979385376,
      "epoch": 2.0009188105579687,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004597535881444366,
      "loss": 4.893,
      "mean_token_accuracy": 0.21267905384302138,
      "num_tokens": 44069417.0,
      "step": 23955
    },
    {
      "entropy": 5.042020511627197,
      "epoch": 2.0013364517206815,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045973659544846834,
      "loss": 4.6803,
      "mean_token_accuracy": 0.22336072027683257,
      "num_tokens": 44079405.0,
      "step": 23960
    },
    {
      "entropy": 5.0707355499267575,
      "epoch": 2.001754092883395,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004597195995184637,
      "loss": 4.7936,
      "mean_token_accuracy": 0.21804890185594558,
      "num_tokens": 44088274.0,
      "step": 23965
    },
    {
      "entropy": 4.950699138641357,
      "epoch": 2.0021717340461076,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004597026003547201,
      "loss": 4.6613,
      "mean_token_accuracy": 0.2300811767578125,
      "num_tokens": 44097340.0,
      "step": 23970
    },
    {
      "entropy": 5.15016622543335,
      "epoch": 2.0025893752088204,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045968559795753493,
      "loss": 4.8149,
      "mean_token_accuracy": 0.21799637675285338,
      "num_tokens": 44106700.0,
      "step": 23975
    },
    {
      "entropy": 5.1454099178314205,
      "epoch": 2.0030070163715337,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045966859232720626,
      "loss": 4.8329,
      "mean_token_accuracy": 0.21430485546588898,
      "num_tokens": 44117030.0,
      "step": 23980
    },
    {
      "entropy": 5.1125571727752686,
      "epoch": 2.0034246575342465,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004596515834640316,
      "loss": 4.7694,
      "mean_token_accuracy": 0.21809237450361252,
      "num_tokens": 44125859.0,
      "step": 23985
    },
    {
      "entropy": 5.060871362686157,
      "epoch": 2.0038422986969597,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004596345713683086,
      "loss": 4.7165,
      "mean_token_accuracy": 0.22405832707881929,
      "num_tokens": 44134745.0,
      "step": 23990
    },
    {
      "entropy": 5.173876237869263,
      "epoch": 2.0042599398596725,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045961755604033536,
      "loss": 4.8508,
      "mean_token_accuracy": 0.21076624244451522,
      "num_tokens": 44144260.0,
      "step": 23995
    },
    {
      "entropy": 5.126508092880249,
      "epoch": 2.0046775810223854,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004596005374804094,
      "loss": 4.7696,
      "mean_token_accuracy": 0.21011869013309478,
      "num_tokens": 44153831.0,
      "step": 24000
    },
    {
      "epoch": 2.0046775810223854,
      "eval_entropy": 4.846252209805402,
      "eval_loss": 4.818875312805176,
      "eval_mean_token_accuracy": 0.22776300444634026,
      "eval_num_tokens": 44153831.0,
      "eval_runtime": 27.4755,
      "eval_samples_per_second": 1374.425,
      "eval_steps_per_second": 171.826,
      "step": 24000
    },
    {
      "entropy": 5.159799671173095,
      "epoch": 2.0050952221850986,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045958351568882883,
      "loss": 4.8289,
      "mean_token_accuracy": 0.2141694501042366,
      "num_tokens": 44163180.0,
      "step": 24005
    },
    {
      "entropy": 5.077951145172119,
      "epoch": 2.0055128633478114,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004595664906658917,
      "loss": 4.7492,
      "mean_token_accuracy": 0.22520564496517181,
      "num_tokens": 44171848.0,
      "step": 24010
    },
    {
      "entropy": 5.14252552986145,
      "epoch": 2.0059305045105247,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045954946241189606,
      "loss": 4.774,
      "mean_token_accuracy": 0.21326659619808197,
      "num_tokens": 44180924.0,
      "step": 24015
    },
    {
      "entropy": 5.149975204467774,
      "epoch": 2.0063481456732375,
      "grad_norm": 1.125,
      "learning_rate": 0.00045953243092713986,
      "loss": 4.7682,
      "mean_token_accuracy": 0.22159954458475112,
      "num_tokens": 44190060.0,
      "step": 24020
    },
    {
      "entropy": 5.154547548294067,
      "epoch": 2.0067657868359507,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045951539621192137,
      "loss": 4.8628,
      "mean_token_accuracy": 0.2184820607304573,
      "num_tokens": 44199545.0,
      "step": 24025
    },
    {
      "entropy": 4.976793909072876,
      "epoch": 2.0071834279986636,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004594983582665387,
      "loss": 4.6315,
      "mean_token_accuracy": 0.22316248267889022,
      "num_tokens": 44208276.0,
      "step": 24030
    },
    {
      "entropy": 5.058961677551269,
      "epoch": 2.0076010691613764,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004594813170912902,
      "loss": 4.6726,
      "mean_token_accuracy": 0.2235078662633896,
      "num_tokens": 44217716.0,
      "step": 24035
    },
    {
      "entropy": 5.10311918258667,
      "epoch": 2.0080187103240896,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045946427268647424,
      "loss": 4.8216,
      "mean_token_accuracy": 0.21809617877006532,
      "num_tokens": 44226652.0,
      "step": 24040
    },
    {
      "entropy": 5.108706665039063,
      "epoch": 2.0084363514868024,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045944722505238914,
      "loss": 4.8611,
      "mean_token_accuracy": 0.2148088589310646,
      "num_tokens": 44236431.0,
      "step": 24045
    },
    {
      "entropy": 4.952599954605103,
      "epoch": 2.0088539926495157,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045943017418933324,
      "loss": 4.5748,
      "mean_token_accuracy": 0.23136397898197175,
      "num_tokens": 44245270.0,
      "step": 24050
    },
    {
      "entropy": 4.947251272201538,
      "epoch": 2.0092716338122285,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004594131200976053,
      "loss": 4.6279,
      "mean_token_accuracy": 0.22437117248773575,
      "num_tokens": 44254352.0,
      "step": 24055
    },
    {
      "entropy": 5.131717729568481,
      "epoch": 2.0096892749749413,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045939606277750356,
      "loss": 4.852,
      "mean_token_accuracy": 0.21826615184545517,
      "num_tokens": 44264011.0,
      "step": 24060
    },
    {
      "entropy": 5.045847988128662,
      "epoch": 2.0101069161376546,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004593790022293269,
      "loss": 4.6556,
      "mean_token_accuracy": 0.22457920014858246,
      "num_tokens": 44272661.0,
      "step": 24065
    },
    {
      "entropy": 5.108726358413696,
      "epoch": 2.0105245573003674,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004593619384533738,
      "loss": 4.8189,
      "mean_token_accuracy": 0.22058157324790956,
      "num_tokens": 44281343.0,
      "step": 24070
    },
    {
      "entropy": 5.076183795928955,
      "epoch": 2.0109421984630806,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045934487144994306,
      "loss": 4.7349,
      "mean_token_accuracy": 0.22182834297418594,
      "num_tokens": 44290174.0,
      "step": 24075
    },
    {
      "entropy": 5.0596600532531735,
      "epoch": 2.0113598396257935,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045932780121933354,
      "loss": 4.7269,
      "mean_token_accuracy": 0.21953545808792113,
      "num_tokens": 44299701.0,
      "step": 24080
    },
    {
      "entropy": 5.088707780838012,
      "epoch": 2.0117774807885067,
      "grad_norm": 1.125,
      "learning_rate": 0.0004593107277618439,
      "loss": 4.7761,
      "mean_token_accuracy": 0.22243632078170777,
      "num_tokens": 44308280.0,
      "step": 24085
    },
    {
      "entropy": 5.160494613647461,
      "epoch": 2.0121951219512195,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004592936510777731,
      "loss": 4.8395,
      "mean_token_accuracy": 0.20523992627859117,
      "num_tokens": 44317857.0,
      "step": 24090
    },
    {
      "entropy": 5.135222244262695,
      "epoch": 2.0126127631139323,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045927657116742016,
      "loss": 4.7826,
      "mean_token_accuracy": 0.21525317728519439,
      "num_tokens": 44327024.0,
      "step": 24095
    },
    {
      "entropy": 5.067237520217896,
      "epoch": 2.0130304042766456,
      "grad_norm": 1.40625,
      "learning_rate": 0.000459259488031084,
      "loss": 4.7817,
      "mean_token_accuracy": 0.2183425173163414,
      "num_tokens": 44336853.0,
      "step": 24100
    },
    {
      "entropy": 5.125655603408814,
      "epoch": 2.0134480454393584,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004592424016690637,
      "loss": 4.7956,
      "mean_token_accuracy": 0.21796344518661498,
      "num_tokens": 44345784.0,
      "step": 24105
    },
    {
      "entropy": 5.12822413444519,
      "epoch": 2.0138656866020717,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045922531208165843,
      "loss": 4.6879,
      "mean_token_accuracy": 0.22880934029817582,
      "num_tokens": 44353886.0,
      "step": 24110
    },
    {
      "entropy": 5.146082735061645,
      "epoch": 2.0142833277647845,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004592082192691673,
      "loss": 4.8413,
      "mean_token_accuracy": 0.21512123048305512,
      "num_tokens": 44365119.0,
      "step": 24115
    },
    {
      "entropy": 5.062642002105713,
      "epoch": 2.0147009689274973,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045919112323188955,
      "loss": 4.7698,
      "mean_token_accuracy": 0.21784849762916564,
      "num_tokens": 44375312.0,
      "step": 24120
    },
    {
      "entropy": 5.075678539276123,
      "epoch": 2.0151186100902105,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004591740239701244,
      "loss": 4.7387,
      "mean_token_accuracy": 0.2085263893008232,
      "num_tokens": 44383986.0,
      "step": 24125
    },
    {
      "entropy": 5.067912483215332,
      "epoch": 2.0155362512529233,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045915692148417136,
      "loss": 4.6481,
      "mean_token_accuracy": 0.23062726855278015,
      "num_tokens": 44393043.0,
      "step": 24130
    },
    {
      "entropy": 5.045528221130371,
      "epoch": 2.0159538924156366,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004591398157743297,
      "loss": 4.7319,
      "mean_token_accuracy": 0.2121296465396881,
      "num_tokens": 44401888.0,
      "step": 24135
    },
    {
      "entropy": 5.085045385360718,
      "epoch": 2.0163715335783494,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045912270684089903,
      "loss": 4.7598,
      "mean_token_accuracy": 0.21979102194309236,
      "num_tokens": 44410380.0,
      "step": 24140
    },
    {
      "entropy": 5.086677503585816,
      "epoch": 2.0167891747410627,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004591055946841786,
      "loss": 4.8591,
      "mean_token_accuracy": 0.21456977277994155,
      "num_tokens": 44419843.0,
      "step": 24145
    },
    {
      "entropy": 5.068501043319702,
      "epoch": 2.0172068159037755,
      "grad_norm": 1.125,
      "learning_rate": 0.00045908847930446815,
      "loss": 4.7047,
      "mean_token_accuracy": 0.22170326709747315,
      "num_tokens": 44428282.0,
      "step": 24150
    },
    {
      "entropy": 5.023894262313843,
      "epoch": 2.0176244570664883,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004590713607020674,
      "loss": 4.6117,
      "mean_token_accuracy": 0.2326388403773308,
      "num_tokens": 44436107.0,
      "step": 24155
    },
    {
      "entropy": 5.046874475479126,
      "epoch": 2.0180420982292016,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045905423887727584,
      "loss": 4.7443,
      "mean_token_accuracy": 0.22386959344148635,
      "num_tokens": 44445681.0,
      "step": 24160
    },
    {
      "entropy": 5.076388025283814,
      "epoch": 2.0184597393919144,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004590371138303933,
      "loss": 4.7218,
      "mean_token_accuracy": 0.22517000883817673,
      "num_tokens": 44454796.0,
      "step": 24165
    },
    {
      "entropy": 5.055276727676391,
      "epoch": 2.0188773805546276,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004590199855617195,
      "loss": 4.6755,
      "mean_token_accuracy": 0.2275378555059433,
      "num_tokens": 44463244.0,
      "step": 24170
    },
    {
      "entropy": 5.019612264633179,
      "epoch": 2.0192950217173404,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004590028540715544,
      "loss": 4.7587,
      "mean_token_accuracy": 0.21936322003602982,
      "num_tokens": 44472093.0,
      "step": 24175
    },
    {
      "entropy": 5.038987493515014,
      "epoch": 2.0197126628800532,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004589857193601978,
      "loss": 4.6126,
      "mean_token_accuracy": 0.23693909347057343,
      "num_tokens": 44481013.0,
      "step": 24180
    },
    {
      "entropy": 5.002915477752685,
      "epoch": 2.0201303040427665,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045896858142794976,
      "loss": 4.6629,
      "mean_token_accuracy": 0.2228978157043457,
      "num_tokens": 44489488.0,
      "step": 24185
    },
    {
      "entropy": 5.062248802185058,
      "epoch": 2.0205479452054793,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004589514402751103,
      "loss": 4.7898,
      "mean_token_accuracy": 0.22093962579965593,
      "num_tokens": 44499215.0,
      "step": 24190
    },
    {
      "entropy": 5.095105791091919,
      "epoch": 2.0209655863681926,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004589342959019794,
      "loss": 4.7517,
      "mean_token_accuracy": 0.21623753756284714,
      "num_tokens": 44509480.0,
      "step": 24195
    },
    {
      "entropy": 5.145942115783692,
      "epoch": 2.0213832275309054,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004589171483088572,
      "loss": 4.7921,
      "mean_token_accuracy": 0.2150246813893318,
      "num_tokens": 44518060.0,
      "step": 24200
    },
    {
      "entropy": 5.1203423023223875,
      "epoch": 2.0218008686936186,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000458899997496044,
      "loss": 4.7679,
      "mean_token_accuracy": 0.2144925445318222,
      "num_tokens": 44527792.0,
      "step": 24205
    },
    {
      "entropy": 5.020914554595947,
      "epoch": 2.0222185098563314,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045888284346384003,
      "loss": 4.7076,
      "mean_token_accuracy": 0.23036398738622665,
      "num_tokens": 44537916.0,
      "step": 24210
    },
    {
      "entropy": 5.014121961593628,
      "epoch": 2.0226361510190443,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004588656862125454,
      "loss": 4.6837,
      "mean_token_accuracy": 0.2278803601861,
      "num_tokens": 44546314.0,
      "step": 24215
    },
    {
      "entropy": 5.033180379867554,
      "epoch": 2.0230537921817575,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004588485257424608,
      "loss": 4.6878,
      "mean_token_accuracy": 0.2243029549717903,
      "num_tokens": 44555184.0,
      "step": 24220
    },
    {
      "entropy": 5.099652910232544,
      "epoch": 2.0234714333444703,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004588313620538863,
      "loss": 4.763,
      "mean_token_accuracy": 0.21658070534467697,
      "num_tokens": 44563626.0,
      "step": 24225
    },
    {
      "entropy": 5.11422176361084,
      "epoch": 2.0238890745071836,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004588141951471226,
      "loss": 4.8174,
      "mean_token_accuracy": 0.218257375061512,
      "num_tokens": 44573332.0,
      "step": 24230
    },
    {
      "entropy": 5.168767976760864,
      "epoch": 2.0243067156698964,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045879702502247014,
      "loss": 4.8152,
      "mean_token_accuracy": 0.20930910408496856,
      "num_tokens": 44582615.0,
      "step": 24235
    },
    {
      "entropy": 5.129198741912842,
      "epoch": 2.024724356832609,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004587798516802296,
      "loss": 4.8617,
      "mean_token_accuracy": 0.20644531399011612,
      "num_tokens": 44592505.0,
      "step": 24240
    },
    {
      "entropy": 5.0671467781066895,
      "epoch": 2.0251419979953225,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045876267512070146,
      "loss": 4.7264,
      "mean_token_accuracy": 0.22946203798055648,
      "num_tokens": 44601920.0,
      "step": 24245
    },
    {
      "entropy": 5.057325124740601,
      "epoch": 2.0255596391580353,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045874549534418644,
      "loss": 4.7432,
      "mean_token_accuracy": 0.23068273812532425,
      "num_tokens": 44610778.0,
      "step": 24250
    },
    {
      "entropy": 5.081636905670166,
      "epoch": 2.0259772803207485,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004587283123509855,
      "loss": 4.6934,
      "mean_token_accuracy": 0.22408584505319595,
      "num_tokens": 44620042.0,
      "step": 24255
    },
    {
      "entropy": 5.0407813549041744,
      "epoch": 2.0263949214834613,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004587111261413992,
      "loss": 4.6641,
      "mean_token_accuracy": 0.2267714262008667,
      "num_tokens": 44628756.0,
      "step": 24260
    },
    {
      "entropy": 4.944365739822388,
      "epoch": 2.0268125626461746,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004586939367157285,
      "loss": 4.6487,
      "mean_token_accuracy": 0.22555789947509766,
      "num_tokens": 44637512.0,
      "step": 24265
    },
    {
      "entropy": 5.041509342193604,
      "epoch": 2.0272302038088874,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045867674407427435,
      "loss": 4.7432,
      "mean_token_accuracy": 0.22468582838773726,
      "num_tokens": 44646071.0,
      "step": 24270
    },
    {
      "entropy": 5.044828844070435,
      "epoch": 2.0276478449716,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045865954821733767,
      "loss": 4.6732,
      "mean_token_accuracy": 0.2238421469926834,
      "num_tokens": 44655302.0,
      "step": 24275
    },
    {
      "entropy": 5.011698818206787,
      "epoch": 2.0280654861343135,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045864234914521956,
      "loss": 4.6767,
      "mean_token_accuracy": 0.22831906974315644,
      "num_tokens": 44663566.0,
      "step": 24280
    },
    {
      "entropy": 5.024549436569214,
      "epoch": 2.0284831272970263,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004586251468582211,
      "loss": 4.6908,
      "mean_token_accuracy": 0.2212030276656151,
      "num_tokens": 44673489.0,
      "step": 24285
    },
    {
      "entropy": 5.032116651535034,
      "epoch": 2.0289007684597395,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004586079413566434,
      "loss": 4.6947,
      "mean_token_accuracy": 0.23159761875867843,
      "num_tokens": 44682213.0,
      "step": 24290
    },
    {
      "entropy": 5.223638725280762,
      "epoch": 2.0293184096224524,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045859073264078765,
      "loss": 4.9097,
      "mean_token_accuracy": 0.21537072360515594,
      "num_tokens": 44690953.0,
      "step": 24295
    },
    {
      "entropy": 5.123017263412476,
      "epoch": 2.029736050785165,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045857352071095514,
      "loss": 4.7207,
      "mean_token_accuracy": 0.22544974982738494,
      "num_tokens": 44700142.0,
      "step": 24300
    },
    {
      "entropy": 5.185457324981689,
      "epoch": 2.0301536919478784,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004585563055674472,
      "loss": 4.8975,
      "mean_token_accuracy": 0.20797480046749114,
      "num_tokens": 44710077.0,
      "step": 24305
    },
    {
      "entropy": 5.108078050613403,
      "epoch": 2.0305713331105912,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045853908721056526,
      "loss": 4.7815,
      "mean_token_accuracy": 0.21796504259109498,
      "num_tokens": 44719264.0,
      "step": 24310
    },
    {
      "entropy": 5.1001969337463375,
      "epoch": 2.0309889742733045,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045852186564061055,
      "loss": 4.7541,
      "mean_token_accuracy": 0.21696643382310868,
      "num_tokens": 44729134.0,
      "step": 24315
    },
    {
      "entropy": 5.049740839004516,
      "epoch": 2.0314066154360173,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004585046408578847,
      "loss": 4.658,
      "mean_token_accuracy": 0.23334278166294098,
      "num_tokens": 44738285.0,
      "step": 24320
    },
    {
      "entropy": 5.045674419403076,
      "epoch": 2.0318242565987306,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045848741286268923,
      "loss": 4.718,
      "mean_token_accuracy": 0.2259058326482773,
      "num_tokens": 44746780.0,
      "step": 24325
    },
    {
      "entropy": 5.043691825866699,
      "epoch": 2.0322418977614434,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045847018165532576,
      "loss": 4.8197,
      "mean_token_accuracy": 0.2109440013766289,
      "num_tokens": 44755806.0,
      "step": 24330
    },
    {
      "entropy": 5.128110408782959,
      "epoch": 2.032659538924156,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045845294723609596,
      "loss": 4.697,
      "mean_token_accuracy": 0.22675138860940933,
      "num_tokens": 44765400.0,
      "step": 24335
    },
    {
      "entropy": 5.123219585418701,
      "epoch": 2.0330771800868694,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004584357096053014,
      "loss": 4.6841,
      "mean_token_accuracy": 0.22410551011562346,
      "num_tokens": 44773847.0,
      "step": 24340
    },
    {
      "entropy": 5.163650560379028,
      "epoch": 2.0334948212495823,
      "grad_norm": 1.109375,
      "learning_rate": 0.000458418468763244,
      "loss": 4.8979,
      "mean_token_accuracy": 0.21430244892835618,
      "num_tokens": 44783075.0,
      "step": 24345
    },
    {
      "entropy": 5.070710754394531,
      "epoch": 2.0339124624122955,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004584012247102255,
      "loss": 4.786,
      "mean_token_accuracy": 0.2145507037639618,
      "num_tokens": 44792708.0,
      "step": 24350
    },
    {
      "entropy": 5.048905372619629,
      "epoch": 2.0343301035750083,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004583839774465479,
      "loss": 4.6956,
      "mean_token_accuracy": 0.22304155826568603,
      "num_tokens": 44801825.0,
      "step": 24355
    },
    {
      "entropy": 5.052341365814209,
      "epoch": 2.034747744737721,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045836672697251293,
      "loss": 4.7053,
      "mean_token_accuracy": 0.2203857645392418,
      "num_tokens": 44810137.0,
      "step": 24360
    },
    {
      "entropy": 5.05956621170044,
      "epoch": 2.0351653859004344,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004583494732884227,
      "loss": 4.7894,
      "mean_token_accuracy": 0.22953833788633346,
      "num_tokens": 44819397.0,
      "step": 24365
    },
    {
      "entropy": 5.099778079986573,
      "epoch": 2.035583027063147,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004583322163945792,
      "loss": 4.7707,
      "mean_token_accuracy": 0.21689544916152953,
      "num_tokens": 44828246.0,
      "step": 24370
    },
    {
      "entropy": 5.062007713317871,
      "epoch": 2.0360006682258605,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004583149562912847,
      "loss": 4.6286,
      "mean_token_accuracy": 0.2276034340262413,
      "num_tokens": 44836916.0,
      "step": 24375
    },
    {
      "entropy": 5.033902978897094,
      "epoch": 2.0364183093885733,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004582976929788411,
      "loss": 4.7772,
      "mean_token_accuracy": 0.22177194356918334,
      "num_tokens": 44846067.0,
      "step": 24380
    },
    {
      "entropy": 5.093341493606568,
      "epoch": 2.0368359505512865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004582804264575508,
      "loss": 4.7116,
      "mean_token_accuracy": 0.2244448184967041,
      "num_tokens": 44855426.0,
      "step": 24385
    },
    {
      "entropy": 5.0170389175415036,
      "epoch": 2.0372535917139993,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000458263156727716,
      "loss": 4.6638,
      "mean_token_accuracy": 0.22493226379156112,
      "num_tokens": 44864455.0,
      "step": 24390
    },
    {
      "entropy": 4.929323530197143,
      "epoch": 2.037671232876712,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004582458837896391,
      "loss": 4.6087,
      "mean_token_accuracy": 0.2306794136762619,
      "num_tokens": 44873568.0,
      "step": 24395
    },
    {
      "entropy": 5.04471173286438,
      "epoch": 2.0380888740394254,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045822860764362235,
      "loss": 4.7487,
      "mean_token_accuracy": 0.21844628006219863,
      "num_tokens": 44884136.0,
      "step": 24400
    },
    {
      "entropy": 5.108980131149292,
      "epoch": 2.038506515202138,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004582113282899683,
      "loss": 4.7445,
      "mean_token_accuracy": 0.21526617258787156,
      "num_tokens": 44893464.0,
      "step": 24405
    },
    {
      "entropy": 5.123139953613281,
      "epoch": 2.0389241563648515,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004581940457289795,
      "loss": 4.7583,
      "mean_token_accuracy": 0.21729944944381713,
      "num_tokens": 44902424.0,
      "step": 24410
    },
    {
      "entropy": 5.064098501205445,
      "epoch": 2.0393417975275643,
      "grad_norm": 1.125,
      "learning_rate": 0.0004581767599609584,
      "loss": 4.7361,
      "mean_token_accuracy": 0.22502943128347397,
      "num_tokens": 44912290.0,
      "step": 24415
    },
    {
      "entropy": 4.969353818893433,
      "epoch": 2.039759438690277,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004581594709862076,
      "loss": 4.6007,
      "mean_token_accuracy": 0.23360099345445634,
      "num_tokens": 44921371.0,
      "step": 24420
    },
    {
      "entropy": 4.998706865310669,
      "epoch": 2.0401770798529903,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045814217880502973,
      "loss": 4.724,
      "mean_token_accuracy": 0.23039331436157226,
      "num_tokens": 44930776.0,
      "step": 24425
    },
    {
      "entropy": 5.102911424636841,
      "epoch": 2.040594721015703,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004581248834177276,
      "loss": 4.7692,
      "mean_token_accuracy": 0.2213438019156456,
      "num_tokens": 44939936.0,
      "step": 24430
    },
    {
      "entropy": 5.0911266803741455,
      "epoch": 2.0410123621784164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045810758482460397,
      "loss": 4.7097,
      "mean_token_accuracy": 0.22081074565649034,
      "num_tokens": 44948329.0,
      "step": 24435
    },
    {
      "entropy": 4.990226411819458,
      "epoch": 2.0414300033411292,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004580902830259617,
      "loss": 4.7176,
      "mean_token_accuracy": 0.22129664719104766,
      "num_tokens": 44956958.0,
      "step": 24440
    },
    {
      "entropy": 5.087716817855835,
      "epoch": 2.0418476445038425,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045807297802210364,
      "loss": 4.7284,
      "mean_token_accuracy": 0.22019705325365066,
      "num_tokens": 44966159.0,
      "step": 24445
    },
    {
      "entropy": 5.114686679840088,
      "epoch": 2.0422652856665553,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004580556698133327,
      "loss": 4.7847,
      "mean_token_accuracy": 0.2120949074625969,
      "num_tokens": 44975155.0,
      "step": 24450
    },
    {
      "entropy": 5.07527379989624,
      "epoch": 2.042682926829268,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004580383583999519,
      "loss": 4.752,
      "mean_token_accuracy": 0.21210570335388185,
      "num_tokens": 44984100.0,
      "step": 24455
    },
    {
      "entropy": 5.0727557182312015,
      "epoch": 2.0431005679919814,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004580210437822643,
      "loss": 4.7395,
      "mean_token_accuracy": 0.21775793731212617,
      "num_tokens": 44992957.0,
      "step": 24460
    },
    {
      "entropy": 5.063355541229248,
      "epoch": 2.043518209154694,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004580037259605731,
      "loss": 4.7345,
      "mean_token_accuracy": 0.21403424739837645,
      "num_tokens": 45002791.0,
      "step": 24465
    },
    {
      "entropy": 5.011760616302491,
      "epoch": 2.0439358503174074,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004579864049351813,
      "loss": 4.6996,
      "mean_token_accuracy": 0.23012530356645583,
      "num_tokens": 45012319.0,
      "step": 24470
    },
    {
      "entropy": 5.111622524261475,
      "epoch": 2.0443534914801202,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004579690807063923,
      "loss": 4.6938,
      "mean_token_accuracy": 0.21774869561195373,
      "num_tokens": 45021799.0,
      "step": 24475
    },
    {
      "entropy": 5.053289270401001,
      "epoch": 2.0447711326428335,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004579517532745093,
      "loss": 4.6881,
      "mean_token_accuracy": 0.22600627839565277,
      "num_tokens": 45031734.0,
      "step": 24480
    },
    {
      "entropy": 4.9967950820922855,
      "epoch": 2.0451887738055463,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045793442263983563,
      "loss": 4.6921,
      "mean_token_accuracy": 0.228014974296093,
      "num_tokens": 45040549.0,
      "step": 24485
    },
    {
      "entropy": 5.118269348144532,
      "epoch": 2.045606414968259,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004579170888026747,
      "loss": 4.8345,
      "mean_token_accuracy": 0.21848441213369368,
      "num_tokens": 45050372.0,
      "step": 24490
    },
    {
      "entropy": 5.216223526000976,
      "epoch": 2.0460240561309724,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004578997517633299,
      "loss": 4.8655,
      "mean_token_accuracy": 0.2111721768975258,
      "num_tokens": 45059655.0,
      "step": 24495
    },
    {
      "entropy": 4.974164581298828,
      "epoch": 2.046441697293685,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045788241152210486,
      "loss": 4.6499,
      "mean_token_accuracy": 0.22819115221500397,
      "num_tokens": 45068043.0,
      "step": 24500
    },
    {
      "entropy": 5.10622239112854,
      "epoch": 2.0468593384563984,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000457865068079303,
      "loss": 4.7603,
      "mean_token_accuracy": 0.22028163075447083,
      "num_tokens": 45076570.0,
      "step": 24505
    },
    {
      "entropy": 5.149793386459351,
      "epoch": 2.0472769796191113,
      "grad_norm": 1.125,
      "learning_rate": 0.000457847721435228,
      "loss": 4.8051,
      "mean_token_accuracy": 0.21760532706975938,
      "num_tokens": 45086170.0,
      "step": 24510
    },
    {
      "entropy": 5.059961414337158,
      "epoch": 2.047694620781824,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045783037159018364,
      "loss": 4.7948,
      "mean_token_accuracy": 0.21720093190670015,
      "num_tokens": 45095580.0,
      "step": 24515
    },
    {
      "entropy": 5.055165147781372,
      "epoch": 2.0481122619445373,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045781301854447347,
      "loss": 4.6946,
      "mean_token_accuracy": 0.22534335106611253,
      "num_tokens": 45104624.0,
      "step": 24520
    },
    {
      "entropy": 5.074354124069214,
      "epoch": 2.04852990310725,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045779566229840144,
      "loss": 4.7683,
      "mean_token_accuracy": 0.21669333130121232,
      "num_tokens": 45114383.0,
      "step": 24525
    },
    {
      "entropy": 5.029681825637818,
      "epoch": 2.0489475442699634,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045777830285227125,
      "loss": 4.6792,
      "mean_token_accuracy": 0.22379084974527358,
      "num_tokens": 45124420.0,
      "step": 24530
    },
    {
      "entropy": 5.056218194961548,
      "epoch": 2.049365185432676,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004577609402063868,
      "loss": 4.7203,
      "mean_token_accuracy": 0.21294601261615753,
      "num_tokens": 45133171.0,
      "step": 24535
    },
    {
      "entropy": 5.097293472290039,
      "epoch": 2.049782826595389,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004577435743610522,
      "loss": 4.7154,
      "mean_token_accuracy": 0.22587222903966903,
      "num_tokens": 45142060.0,
      "step": 24540
    },
    {
      "entropy": 5.069958353042603,
      "epoch": 2.0502004677581023,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004577262053165713,
      "loss": 4.7252,
      "mean_token_accuracy": 0.2236767068505287,
      "num_tokens": 45151231.0,
      "step": 24545
    },
    {
      "entropy": 5.14122462272644,
      "epoch": 2.050618108920815,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004577088330732483,
      "loss": 4.8185,
      "mean_token_accuracy": 0.21697262823581695,
      "num_tokens": 45161175.0,
      "step": 24550
    },
    {
      "entropy": 5.085650444030762,
      "epoch": 2.0510357500835283,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004576914576313872,
      "loss": 4.8181,
      "mean_token_accuracy": 0.20956612676382064,
      "num_tokens": 45171018.0,
      "step": 24555
    },
    {
      "entropy": 5.089688873291015,
      "epoch": 2.051453391246241,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004576740789912922,
      "loss": 4.7395,
      "mean_token_accuracy": 0.22412060499191283,
      "num_tokens": 45180386.0,
      "step": 24560
    },
    {
      "entropy": 5.022232532501221,
      "epoch": 2.0518710324089544,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004576566971532676,
      "loss": 4.6151,
      "mean_token_accuracy": 0.22990407794713974,
      "num_tokens": 45188879.0,
      "step": 24565
    },
    {
      "entropy": 5.042891883850098,
      "epoch": 2.052288673571667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004576393121176177,
      "loss": 4.777,
      "mean_token_accuracy": 0.22039705663919448,
      "num_tokens": 45198685.0,
      "step": 24570
    },
    {
      "entropy": 5.114610099792481,
      "epoch": 2.05270631473438,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045762192388464675,
      "loss": 4.8249,
      "mean_token_accuracy": 0.20987554490566254,
      "num_tokens": 45207712.0,
      "step": 24575
    },
    {
      "entropy": 5.107165718078614,
      "epoch": 2.0531239558970933,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004576045324546592,
      "loss": 4.6864,
      "mean_token_accuracy": 0.2203238993883133,
      "num_tokens": 45216791.0,
      "step": 24580
    },
    {
      "entropy": 5.1812914371490475,
      "epoch": 2.053541597059806,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045758713782795946,
      "loss": 4.8549,
      "mean_token_accuracy": 0.21007587760686874,
      "num_tokens": 45227125.0,
      "step": 24585
    },
    {
      "entropy": 5.0550346851348875,
      "epoch": 2.0539592382225194,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045756974000485216,
      "loss": 4.6973,
      "mean_token_accuracy": 0.23060209155082703,
      "num_tokens": 45236021.0,
      "step": 24590
    },
    {
      "entropy": 5.00059814453125,
      "epoch": 2.054376879385232,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045755233898564176,
      "loss": 4.6898,
      "mean_token_accuracy": 0.2278942734003067,
      "num_tokens": 45244858.0,
      "step": 24595
    },
    {
      "entropy": 5.11146650314331,
      "epoch": 2.0547945205479454,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045753493477063297,
      "loss": 4.8923,
      "mean_token_accuracy": 0.21278541535139084,
      "num_tokens": 45254474.0,
      "step": 24600
    },
    {
      "entropy": 5.138153696060181,
      "epoch": 2.0552121617106582,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004575175273601304,
      "loss": 4.6973,
      "mean_token_accuracy": 0.22173168063163756,
      "num_tokens": 45262238.0,
      "step": 24605
    },
    {
      "entropy": 5.012244367599488,
      "epoch": 2.055629802873371,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045750011675443886,
      "loss": 4.6891,
      "mean_token_accuracy": 0.23184198886156082,
      "num_tokens": 45270520.0,
      "step": 24610
    },
    {
      "entropy": 5.027739000320435,
      "epoch": 2.0560474440360843,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004574827029538631,
      "loss": 4.7643,
      "mean_token_accuracy": 0.21977008879184723,
      "num_tokens": 45279208.0,
      "step": 24615
    },
    {
      "entropy": 5.012616682052612,
      "epoch": 2.056465085198797,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004574652859587079,
      "loss": 4.6766,
      "mean_token_accuracy": 0.22627577185630798,
      "num_tokens": 45288175.0,
      "step": 24620
    },
    {
      "entropy": 5.091184282302857,
      "epoch": 2.0568827263615104,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045744786576927825,
      "loss": 4.7666,
      "mean_token_accuracy": 0.22483773827552794,
      "num_tokens": 45298484.0,
      "step": 24625
    },
    {
      "entropy": 5.124270963668823,
      "epoch": 2.057300367524223,
      "grad_norm": 1.125,
      "learning_rate": 0.0004574304423858791,
      "loss": 4.7962,
      "mean_token_accuracy": 0.21811570525169371,
      "num_tokens": 45308275.0,
      "step": 24630
    },
    {
      "entropy": 4.952074003219605,
      "epoch": 2.057718008686936,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045741301580881557,
      "loss": 4.6606,
      "mean_token_accuracy": 0.22750235944986344,
      "num_tokens": 45316428.0,
      "step": 24635
    },
    {
      "entropy": 5.0064483165740965,
      "epoch": 2.0581356498496493,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004573955860383925,
      "loss": 4.683,
      "mean_token_accuracy": 0.22110970914363862,
      "num_tokens": 45326038.0,
      "step": 24640
    },
    {
      "entropy": 5.188286256790161,
      "epoch": 2.058553291012362,
      "grad_norm": 1.125,
      "learning_rate": 0.0004573781530749152,
      "loss": 4.8579,
      "mean_token_accuracy": 0.20936672389507294,
      "num_tokens": 45335614.0,
      "step": 24645
    },
    {
      "entropy": 5.102451467514038,
      "epoch": 2.0589709321750753,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045736071691868883,
      "loss": 4.7595,
      "mean_token_accuracy": 0.225762015581131,
      "num_tokens": 45344379.0,
      "step": 24650
    },
    {
      "entropy": 5.081534099578858,
      "epoch": 2.059388573337788,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004573432775700185,
      "loss": 4.7776,
      "mean_token_accuracy": 0.21974516808986663,
      "num_tokens": 45353723.0,
      "step": 24655
    },
    {
      "entropy": 5.083104038238526,
      "epoch": 2.0598062145005014,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004573258350292097,
      "loss": 4.73,
      "mean_token_accuracy": 0.22443019449710847,
      "num_tokens": 45362721.0,
      "step": 24660
    },
    {
      "entropy": 5.043258762359619,
      "epoch": 2.060223855663214,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004573083892965677,
      "loss": 4.7707,
      "mean_token_accuracy": 0.22253322154283522,
      "num_tokens": 45372531.0,
      "step": 24665
    },
    {
      "entropy": 5.075724983215332,
      "epoch": 2.060641496825927,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004572909403723978,
      "loss": 4.6889,
      "mean_token_accuracy": 0.22679054141044616,
      "num_tokens": 45381330.0,
      "step": 24670
    },
    {
      "entropy": 5.070838785171508,
      "epoch": 2.0610591379886403,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004572734882570057,
      "loss": 4.7796,
      "mean_token_accuracy": 0.21869994401931764,
      "num_tokens": 45390897.0,
      "step": 24675
    },
    {
      "entropy": 5.022576570510864,
      "epoch": 2.061476779151353,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045725603295069666,
      "loss": 4.7402,
      "mean_token_accuracy": 0.21508883088827133,
      "num_tokens": 45400602.0,
      "step": 24680
    },
    {
      "entropy": 5.071195793151856,
      "epoch": 2.0618944203140663,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004572385744537765,
      "loss": 4.6846,
      "mean_token_accuracy": 0.23209428340196608,
      "num_tokens": 45409013.0,
      "step": 24685
    },
    {
      "entropy": 5.063447427749634,
      "epoch": 2.062312061476779,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045722111276655066,
      "loss": 4.8755,
      "mean_token_accuracy": 0.21157441586256026,
      "num_tokens": 45418862.0,
      "step": 24690
    },
    {
      "entropy": 5.108270502090454,
      "epoch": 2.062729702639492,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045720364788932486,
      "loss": 4.8144,
      "mean_token_accuracy": 0.2168186590075493,
      "num_tokens": 45427304.0,
      "step": 24695
    },
    {
      "entropy": 5.116755390167237,
      "epoch": 2.063147343802205,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045718617982240496,
      "loss": 4.712,
      "mean_token_accuracy": 0.22181703597307206,
      "num_tokens": 45435835.0,
      "step": 24700
    },
    {
      "entropy": 5.07468614578247,
      "epoch": 2.063564984964918,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004571687085660967,
      "loss": 4.7576,
      "mean_token_accuracy": 0.2205907478928566,
      "num_tokens": 45444565.0,
      "step": 24705
    },
    {
      "entropy": 5.002976608276367,
      "epoch": 2.0639826261276313,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045715123412070594,
      "loss": 4.6419,
      "mean_token_accuracy": 0.22457585036754607,
      "num_tokens": 45453046.0,
      "step": 24710
    },
    {
      "entropy": 5.046605205535888,
      "epoch": 2.064400267290344,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004571337564865385,
      "loss": 4.7201,
      "mean_token_accuracy": 0.2306382715702057,
      "num_tokens": 45461377.0,
      "step": 24715
    },
    {
      "entropy": 5.056305027008056,
      "epoch": 2.0648179084530573,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004571162756639004,
      "loss": 4.7849,
      "mean_token_accuracy": 0.21717772036790847,
      "num_tokens": 45470487.0,
      "step": 24720
    },
    {
      "entropy": 5.0168859481811525,
      "epoch": 2.06523554961577,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004570987916530977,
      "loss": 4.6483,
      "mean_token_accuracy": 0.22359618842601775,
      "num_tokens": 45479477.0,
      "step": 24725
    },
    {
      "entropy": 5.068761157989502,
      "epoch": 2.065653190778483,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045708130445443646,
      "loss": 4.7023,
      "mean_token_accuracy": 0.2270632341504097,
      "num_tokens": 45488354.0,
      "step": 24730
    },
    {
      "entropy": 5.151559925079345,
      "epoch": 2.0660708319411962,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045706381406822284,
      "loss": 4.7905,
      "mean_token_accuracy": 0.22037963271141053,
      "num_tokens": 45498076.0,
      "step": 24735
    },
    {
      "entropy": 5.1230710506439205,
      "epoch": 2.066488473103909,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000457046320494763,
      "loss": 4.7932,
      "mean_token_accuracy": 0.22443168014287948,
      "num_tokens": 45507259.0,
      "step": 24740
    },
    {
      "entropy": 5.022133445739746,
      "epoch": 2.0669061142666223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045702882373436317,
      "loss": 4.6706,
      "mean_token_accuracy": 0.22593455165624618,
      "num_tokens": 45517673.0,
      "step": 24745
    },
    {
      "entropy": 5.015563869476319,
      "epoch": 2.067323755429335,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004570113237873297,
      "loss": 4.7569,
      "mean_token_accuracy": 0.21150522381067277,
      "num_tokens": 45527599.0,
      "step": 24750
    },
    {
      "entropy": 5.126435232162476,
      "epoch": 2.067741396592048,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045699382065396893,
      "loss": 4.7025,
      "mean_token_accuracy": 0.224578557908535,
      "num_tokens": 45537154.0,
      "step": 24755
    },
    {
      "entropy": 5.011871194839477,
      "epoch": 2.068159037754761,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004569763143345872,
      "loss": 4.7056,
      "mean_token_accuracy": 0.23024413287639617,
      "num_tokens": 45546043.0,
      "step": 24760
    },
    {
      "entropy": 5.0302996158599855,
      "epoch": 2.068576678917474,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045695880482949107,
      "loss": 4.7071,
      "mean_token_accuracy": 0.22780821919441224,
      "num_tokens": 45554997.0,
      "step": 24765
    },
    {
      "entropy": 5.120640468597412,
      "epoch": 2.0689943200801872,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004569412921389871,
      "loss": 4.7451,
      "mean_token_accuracy": 0.21468440294265748,
      "num_tokens": 45563730.0,
      "step": 24770
    },
    {
      "entropy": 5.0246484756469725,
      "epoch": 2.0694119612429,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004569237762633818,
      "loss": 4.7233,
      "mean_token_accuracy": 0.22900073677301408,
      "num_tokens": 45572300.0,
      "step": 24775
    },
    {
      "entropy": 5.00896577835083,
      "epoch": 2.0698296024056133,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045690625720298166,
      "loss": 4.7292,
      "mean_token_accuracy": 0.22671122401952742,
      "num_tokens": 45580549.0,
      "step": 24780
    },
    {
      "entropy": 5.051060152053833,
      "epoch": 2.070247243568326,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004568887349580937,
      "loss": 4.7611,
      "mean_token_accuracy": 0.21613606363534926,
      "num_tokens": 45589414.0,
      "step": 24785
    },
    {
      "entropy": 4.974513578414917,
      "epoch": 2.070664884731039,
      "grad_norm": 1.125,
      "learning_rate": 0.00045687120952902447,
      "loss": 4.6594,
      "mean_token_accuracy": 0.2261205032467842,
      "num_tokens": 45598564.0,
      "step": 24790
    },
    {
      "entropy": 5.064595031738281,
      "epoch": 2.071082525893752,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045685368091608075,
      "loss": 4.7514,
      "mean_token_accuracy": 0.22206619381904602,
      "num_tokens": 45608441.0,
      "step": 24795
    },
    {
      "entropy": 5.072186851501465,
      "epoch": 2.071500167056465,
      "grad_norm": 1.125,
      "learning_rate": 0.00045683614911956946,
      "loss": 4.664,
      "mean_token_accuracy": 0.23070189654827117,
      "num_tokens": 45618259.0,
      "step": 24800
    },
    {
      "entropy": 5.087060451507568,
      "epoch": 2.0719178082191783,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004568186141397975,
      "loss": 4.8997,
      "mean_token_accuracy": 0.21006667613983154,
      "num_tokens": 45627347.0,
      "step": 24805
    },
    {
      "entropy": 5.136538171768189,
      "epoch": 2.072335449381891,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045680107597707186,
      "loss": 4.824,
      "mean_token_accuracy": 0.21371307224035263,
      "num_tokens": 45637074.0,
      "step": 24810
    },
    {
      "entropy": 5.025939130783081,
      "epoch": 2.072753090544604,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045678353463169953,
      "loss": 4.706,
      "mean_token_accuracy": 0.21963571459054948,
      "num_tokens": 45646653.0,
      "step": 24815
    },
    {
      "entropy": 5.10043511390686,
      "epoch": 2.073170731707317,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004567659901039876,
      "loss": 4.8607,
      "mean_token_accuracy": 0.2129446819424629,
      "num_tokens": 45657035.0,
      "step": 24820
    },
    {
      "entropy": 5.048998117446899,
      "epoch": 2.07358837287003,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004567484423942433,
      "loss": 4.7267,
      "mean_token_accuracy": 0.22189089357852937,
      "num_tokens": 45666050.0,
      "step": 24825
    },
    {
      "entropy": 5.06818208694458,
      "epoch": 2.074006014032743,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045673089150277365,
      "loss": 4.7461,
      "mean_token_accuracy": 0.22534300088882447,
      "num_tokens": 45674439.0,
      "step": 24830
    },
    {
      "entropy": 5.08548150062561,
      "epoch": 2.074423655195456,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045671333742988604,
      "loss": 4.7491,
      "mean_token_accuracy": 0.21800862103700638,
      "num_tokens": 45683298.0,
      "step": 24835
    },
    {
      "entropy": 5.137650108337402,
      "epoch": 2.0748412963581693,
      "grad_norm": 1.125,
      "learning_rate": 0.0004566957801758878,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2169272005558014,
      "num_tokens": 45692879.0,
      "step": 24840
    },
    {
      "entropy": 5.02471923828125,
      "epoch": 2.075258937520882,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004566782197410861,
      "loss": 4.7362,
      "mean_token_accuracy": 0.21793881207704544,
      "num_tokens": 45702166.0,
      "step": 24845
    },
    {
      "entropy": 5.0677014827728275,
      "epoch": 2.075676578683595,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004566606561257885,
      "loss": 4.6823,
      "mean_token_accuracy": 0.22917154133319856,
      "num_tokens": 45711729.0,
      "step": 24850
    },
    {
      "entropy": 5.047508430480957,
      "epoch": 2.076094219846308,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004566430893303025,
      "loss": 4.6633,
      "mean_token_accuracy": 0.22051584124565124,
      "num_tokens": 45721446.0,
      "step": 24855
    },
    {
      "entropy": 5.074662685394287,
      "epoch": 2.076511861009021,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004566255193549355,
      "loss": 4.7822,
      "mean_token_accuracy": 0.20849283337593078,
      "num_tokens": 45730565.0,
      "step": 24860
    },
    {
      "entropy": 5.101299238204956,
      "epoch": 2.076929502171734,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045660794619999524,
      "loss": 4.7683,
      "mean_token_accuracy": 0.2191971704363823,
      "num_tokens": 45739775.0,
      "step": 24865
    },
    {
      "entropy": 5.028861808776855,
      "epoch": 2.077347143334447,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004565903698657893,
      "loss": 4.7024,
      "mean_token_accuracy": 0.2226012572646141,
      "num_tokens": 45749032.0,
      "step": 24870
    },
    {
      "entropy": 4.999443197250367,
      "epoch": 2.07776478449716,
      "grad_norm": 1.125,
      "learning_rate": 0.00045657279035262534,
      "loss": 4.7452,
      "mean_token_accuracy": 0.2231003537774086,
      "num_tokens": 45758184.0,
      "step": 24875
    },
    {
      "entropy": 5.068873310089112,
      "epoch": 2.078182425659873,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000456555207660811,
      "loss": 4.6957,
      "mean_token_accuracy": 0.22811125963926315,
      "num_tokens": 45767017.0,
      "step": 24880
    },
    {
      "entropy": 5.046181297302246,
      "epoch": 2.078600066822586,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004565376217906543,
      "loss": 4.6943,
      "mean_token_accuracy": 0.2216106027364731,
      "num_tokens": 45776388.0,
      "step": 24885
    },
    {
      "entropy": 5.054635143280029,
      "epoch": 2.079017707985299,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045652003274246306,
      "loss": 4.81,
      "mean_token_accuracy": 0.2152481883764267,
      "num_tokens": 45787264.0,
      "step": 24890
    },
    {
      "entropy": 5.030606412887574,
      "epoch": 2.079435349148012,
      "grad_norm": 1.125,
      "learning_rate": 0.0004565024405165452,
      "loss": 4.722,
      "mean_token_accuracy": 0.22723445743322374,
      "num_tokens": 45795587.0,
      "step": 24895
    },
    {
      "entropy": 5.080720043182373,
      "epoch": 2.0798529903107252,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045648484511320853,
      "loss": 4.7173,
      "mean_token_accuracy": 0.22858177572488786,
      "num_tokens": 45804153.0,
      "step": 24900
    },
    {
      "entropy": 5.024067497253418,
      "epoch": 2.080270631473438,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045646724653276117,
      "loss": 4.7099,
      "mean_token_accuracy": 0.2205055460333824,
      "num_tokens": 45812508.0,
      "step": 24905
    },
    {
      "entropy": 5.023562908172607,
      "epoch": 2.080688272636151,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045644964477551124,
      "loss": 4.6966,
      "mean_token_accuracy": 0.2215425968170166,
      "num_tokens": 45823247.0,
      "step": 24910
    },
    {
      "entropy": 5.10070767402649,
      "epoch": 2.081105913798864,
      "grad_norm": 1.125,
      "learning_rate": 0.000456432039841767,
      "loss": 4.7525,
      "mean_token_accuracy": 0.2252424329519272,
      "num_tokens": 45832040.0,
      "step": 24915
    },
    {
      "entropy": 5.046929836273193,
      "epoch": 2.081523554961577,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004564144317318364,
      "loss": 4.7486,
      "mean_token_accuracy": 0.21867081373929978,
      "num_tokens": 45841957.0,
      "step": 24920
    },
    {
      "entropy": 5.001021146774292,
      "epoch": 2.08194119612429,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045639682044602777,
      "loss": 4.6547,
      "mean_token_accuracy": 0.22724236398935319,
      "num_tokens": 45851120.0,
      "step": 24925
    },
    {
      "entropy": 5.048725986480713,
      "epoch": 2.082358837287003,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045637920598464956,
      "loss": 4.7311,
      "mean_token_accuracy": 0.22061003893613815,
      "num_tokens": 45860199.0,
      "step": 24930
    },
    {
      "entropy": 5.051855754852295,
      "epoch": 2.082776478449716,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045636158834800987,
      "loss": 4.7986,
      "mean_token_accuracy": 0.21922173649072646,
      "num_tokens": 45869007.0,
      "step": 24935
    },
    {
      "entropy": 5.027041625976563,
      "epoch": 2.083194119612429,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045634396753641736,
      "loss": 4.6754,
      "mean_token_accuracy": 0.23204258680343628,
      "num_tokens": 45877624.0,
      "step": 24940
    },
    {
      "entropy": 5.0708075046539305,
      "epoch": 2.083611760775142,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004563263435501805,
      "loss": 4.8164,
      "mean_token_accuracy": 0.21438514292240143,
      "num_tokens": 45886797.0,
      "step": 24945
    },
    {
      "entropy": 5.092451047897339,
      "epoch": 2.084029401937855,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004563087163896076,
      "loss": 4.7279,
      "mean_token_accuracy": 0.22395619004964828,
      "num_tokens": 45895335.0,
      "step": 24950
    },
    {
      "entropy": 5.107417392730713,
      "epoch": 2.084447043100568,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045629108605500747,
      "loss": 4.7551,
      "mean_token_accuracy": 0.22062230110168457,
      "num_tokens": 45904917.0,
      "step": 24955
    },
    {
      "entropy": 5.065419387817383,
      "epoch": 2.084864684263281,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045627345254668864,
      "loss": 4.7202,
      "mean_token_accuracy": 0.22735391259193422,
      "num_tokens": 45914369.0,
      "step": 24960
    },
    {
      "entropy": 5.108623933792114,
      "epoch": 2.085282325425994,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045625581586495977,
      "loss": 4.8263,
      "mean_token_accuracy": 0.21361476331949233,
      "num_tokens": 45923765.0,
      "step": 24965
    },
    {
      "entropy": 5.079609537124634,
      "epoch": 2.085699966588707,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004562381760101297,
      "loss": 4.6884,
      "mean_token_accuracy": 0.22072772681713104,
      "num_tokens": 45932908.0,
      "step": 24970
    },
    {
      "entropy": 5.048910474777221,
      "epoch": 2.08611760775142,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004562205329825072,
      "loss": 4.7086,
      "mean_token_accuracy": 0.2304063618183136,
      "num_tokens": 45942313.0,
      "step": 24975
    },
    {
      "entropy": 4.9344817161560055,
      "epoch": 2.086535248914133,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004562028867824012,
      "loss": 4.6465,
      "mean_token_accuracy": 0.23233472853899,
      "num_tokens": 45951433.0,
      "step": 24980
    },
    {
      "entropy": 4.932235383987427,
      "epoch": 2.086952890076846,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045618523741012044,
      "loss": 4.6275,
      "mean_token_accuracy": 0.23418813943862915,
      "num_tokens": 45960376.0,
      "step": 24985
    },
    {
      "entropy": 4.941289329528809,
      "epoch": 2.087370531239559,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004561675848659741,
      "loss": 4.6625,
      "mean_token_accuracy": 0.2362625256180763,
      "num_tokens": 45969614.0,
      "step": 24990
    },
    {
      "entropy": 5.10434741973877,
      "epoch": 2.0877881724022718,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045614992915027103,
      "loss": 4.8225,
      "mean_token_accuracy": 0.21008726060390473,
      "num_tokens": 45979068.0,
      "step": 24995
    },
    {
      "entropy": 5.003912115097046,
      "epoch": 2.088205813564985,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045613227026332045,
      "loss": 4.6786,
      "mean_token_accuracy": 0.22846003323793412,
      "num_tokens": 45988101.0,
      "step": 25000
    },
    {
      "entropy": 5.078441143035889,
      "epoch": 2.088623454727698,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004561146082054314,
      "loss": 4.7179,
      "mean_token_accuracy": 0.22706440538167955,
      "num_tokens": 45998051.0,
      "step": 25005
    },
    {
      "entropy": 5.0882996082305905,
      "epoch": 2.089041095890411,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004560969429769132,
      "loss": 4.7894,
      "mean_token_accuracy": 0.21642557084560393,
      "num_tokens": 46007496.0,
      "step": 25010
    },
    {
      "entropy": 5.061066961288452,
      "epoch": 2.089458737053124,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045607927457807503,
      "loss": 4.6588,
      "mean_token_accuracy": 0.2360288083553314,
      "num_tokens": 46016307.0,
      "step": 25015
    },
    {
      "entropy": 5.083470678329467,
      "epoch": 2.089876378215837,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004560616030092261,
      "loss": 4.7784,
      "mean_token_accuracy": 0.21557476967573166,
      "num_tokens": 46025558.0,
      "step": 25020
    },
    {
      "entropy": 5.025048398971558,
      "epoch": 2.09029401937855,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004560439282706759,
      "loss": 4.7439,
      "mean_token_accuracy": 0.21875559240579606,
      "num_tokens": 46034874.0,
      "step": 25025
    },
    {
      "entropy": 5.058929824829102,
      "epoch": 2.090711660541263,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045602625036273385,
      "loss": 4.7609,
      "mean_token_accuracy": 0.222522634267807,
      "num_tokens": 46044392.0,
      "step": 25030
    },
    {
      "entropy": 5.028374147415161,
      "epoch": 2.091129301703976,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045600856928570944,
      "loss": 4.7071,
      "mean_token_accuracy": 0.22396897822618483,
      "num_tokens": 46053677.0,
      "step": 25035
    },
    {
      "entropy": 5.112160634994507,
      "epoch": 2.091546942866689,
      "grad_norm": 1.125,
      "learning_rate": 0.00045599088503991205,
      "loss": 4.7788,
      "mean_token_accuracy": 0.22538436502218245,
      "num_tokens": 46062602.0,
      "step": 25040
    },
    {
      "entropy": 5.085211181640625,
      "epoch": 2.091964584029402,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004559731976256514,
      "loss": 4.7413,
      "mean_token_accuracy": 0.22741070091724397,
      "num_tokens": 46071434.0,
      "step": 25045
    },
    {
      "entropy": 5.008741283416748,
      "epoch": 2.092382225192115,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045595550704323706,
      "loss": 4.7651,
      "mean_token_accuracy": 0.21946151107549666,
      "num_tokens": 46080898.0,
      "step": 25050
    },
    {
      "entropy": 5.081285905838013,
      "epoch": 2.0927998663548277,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045593781329297876,
      "loss": 4.7891,
      "mean_token_accuracy": 0.21948437243700028,
      "num_tokens": 46089981.0,
      "step": 25055
    },
    {
      "entropy": 5.098827743530274,
      "epoch": 2.093217507517541,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045592011637518625,
      "loss": 4.7388,
      "mean_token_accuracy": 0.21568899899721145,
      "num_tokens": 46099110.0,
      "step": 25060
    },
    {
      "entropy": 5.030089044570923,
      "epoch": 2.093635148680254,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045590241629016933,
      "loss": 4.738,
      "mean_token_accuracy": 0.22415300607681274,
      "num_tokens": 46109158.0,
      "step": 25065
    },
    {
      "entropy": 4.923915719985962,
      "epoch": 2.094052789842967,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045588471303823787,
      "loss": 4.6019,
      "mean_token_accuracy": 0.22757447510957718,
      "num_tokens": 46118433.0,
      "step": 25070
    },
    {
      "entropy": 5.0514741897583,
      "epoch": 2.09447043100568,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045586700661970174,
      "loss": 4.7209,
      "mean_token_accuracy": 0.23069310337305068,
      "num_tokens": 46127785.0,
      "step": 25075
    },
    {
      "entropy": 5.05109920501709,
      "epoch": 2.094888072168393,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045584929703487107,
      "loss": 4.6954,
      "mean_token_accuracy": 0.23049048483371734,
      "num_tokens": 46136543.0,
      "step": 25080
    },
    {
      "entropy": 4.986274099349975,
      "epoch": 2.095305713331106,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004558315842840556,
      "loss": 4.6911,
      "mean_token_accuracy": 0.22666891515254975,
      "num_tokens": 46145591.0,
      "step": 25085
    },
    {
      "entropy": 5.1065877914428714,
      "epoch": 2.0957233544938187,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004558138683675657,
      "loss": 4.8155,
      "mean_token_accuracy": 0.2169689029455185,
      "num_tokens": 46156343.0,
      "step": 25090
    },
    {
      "entropy": 5.143506956100464,
      "epoch": 2.096140995656532,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045579614928571135,
      "loss": 4.9011,
      "mean_token_accuracy": 0.2053508535027504,
      "num_tokens": 46167153.0,
      "step": 25095
    },
    {
      "entropy": 5.082786655426025,
      "epoch": 2.096558636819245,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004557784270388028,
      "loss": 4.7379,
      "mean_token_accuracy": 0.22076621204614638,
      "num_tokens": 46176459.0,
      "step": 25100
    },
    {
      "entropy": 5.043760633468628,
      "epoch": 2.096976277981958,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045576070162715025,
      "loss": 4.7093,
      "mean_token_accuracy": 0.2249722361564636,
      "num_tokens": 46185563.0,
      "step": 25105
    },
    {
      "entropy": 5.075391864776611,
      "epoch": 2.097393919144671,
      "grad_norm": 1.15625,
      "learning_rate": 0.000455742973051064,
      "loss": 4.7694,
      "mean_token_accuracy": 0.22494118809700012,
      "num_tokens": 46193945.0,
      "step": 25110
    },
    {
      "entropy": 5.098305082321167,
      "epoch": 2.0978115603073837,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004557252413108545,
      "loss": 4.8122,
      "mean_token_accuracy": 0.22280343025922775,
      "num_tokens": 46203186.0,
      "step": 25115
    },
    {
      "entropy": 4.9947656154632565,
      "epoch": 2.098229201470097,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045570750640683207,
      "loss": 4.6706,
      "mean_token_accuracy": 0.2285703256726265,
      "num_tokens": 46213219.0,
      "step": 25120
    },
    {
      "entropy": 4.903379011154175,
      "epoch": 2.0986468426328098,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004556897683393073,
      "loss": 4.5374,
      "mean_token_accuracy": 0.23846973329782487,
      "num_tokens": 46221501.0,
      "step": 25125
    },
    {
      "entropy": 4.977916097640991,
      "epoch": 2.099064483795523,
      "grad_norm": 1.125,
      "learning_rate": 0.00045567202710859056,
      "loss": 4.6761,
      "mean_token_accuracy": 0.23025596290826797,
      "num_tokens": 46230056.0,
      "step": 25130
    },
    {
      "entropy": 5.022942924499512,
      "epoch": 2.099482124958236,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004556542827149926,
      "loss": 4.7041,
      "mean_token_accuracy": 0.22675396054983138,
      "num_tokens": 46239239.0,
      "step": 25135
    },
    {
      "entropy": 5.044459199905395,
      "epoch": 2.099899766120949,
      "grad_norm": 1.125,
      "learning_rate": 0.0004556365351588239,
      "loss": 4.7541,
      "mean_token_accuracy": 0.21661781668663024,
      "num_tokens": 46249331.0,
      "step": 25140
    },
    {
      "entropy": 5.036999368667603,
      "epoch": 2.100317407283662,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045561878444039526,
      "loss": 4.7192,
      "mean_token_accuracy": 0.22677402794361115,
      "num_tokens": 46258114.0,
      "step": 25145
    },
    {
      "entropy": 5.037876558303833,
      "epoch": 2.1007350484463747,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045560103056001733,
      "loss": 4.7072,
      "mean_token_accuracy": 0.22990042269229888,
      "num_tokens": 46266443.0,
      "step": 25150
    },
    {
      "entropy": 5.0332752704620365,
      "epoch": 2.101152689609088,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000455583273518001,
      "loss": 4.7479,
      "mean_token_accuracy": 0.22877189218997956,
      "num_tokens": 46275722.0,
      "step": 25155
    },
    {
      "entropy": 4.975849199295044,
      "epoch": 2.1015703307718008,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004555655133146571,
      "loss": 4.6189,
      "mean_token_accuracy": 0.23355795741081237,
      "num_tokens": 46283942.0,
      "step": 25160
    },
    {
      "entropy": 4.992610597610474,
      "epoch": 2.101987971934514,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004555477499502965,
      "loss": 4.6994,
      "mean_token_accuracy": 0.22804312258958817,
      "num_tokens": 46293162.0,
      "step": 25165
    },
    {
      "entropy": 5.023422145843506,
      "epoch": 2.102405613097227,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004555299834252303,
      "loss": 4.7837,
      "mean_token_accuracy": 0.225237175822258,
      "num_tokens": 46301616.0,
      "step": 25170
    },
    {
      "entropy": 5.012982177734375,
      "epoch": 2.1028232542599397,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004555122137397694,
      "loss": 4.7132,
      "mean_token_accuracy": 0.22150464802980424,
      "num_tokens": 46311021.0,
      "step": 25175
    },
    {
      "entropy": 5.036050081253052,
      "epoch": 2.103240895422653,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045549444089422486,
      "loss": 4.7209,
      "mean_token_accuracy": 0.22133426666259765,
      "num_tokens": 46320107.0,
      "step": 25180
    },
    {
      "entropy": 5.139594316482544,
      "epoch": 2.1036585365853657,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045547666488890794,
      "loss": 4.8779,
      "mean_token_accuracy": 0.21108482480049134,
      "num_tokens": 46329063.0,
      "step": 25185
    },
    {
      "entropy": 5.061287117004395,
      "epoch": 2.104076177748079,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004554588857241297,
      "loss": 4.7655,
      "mean_token_accuracy": 0.2180076628923416,
      "num_tokens": 46338228.0,
      "step": 25190
    },
    {
      "entropy": 5.031609344482422,
      "epoch": 2.104493818910792,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045544110340020155,
      "loss": 4.683,
      "mean_token_accuracy": 0.22300671339035033,
      "num_tokens": 46347318.0,
      "step": 25195
    },
    {
      "entropy": 5.047584533691406,
      "epoch": 2.104911460073505,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004554233179174346,
      "loss": 4.7547,
      "mean_token_accuracy": 0.21937695741653443,
      "num_tokens": 46357112.0,
      "step": 25200
    },
    {
      "entropy": 5.06604495048523,
      "epoch": 2.105329101236218,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004554055292761403,
      "loss": 4.8248,
      "mean_token_accuracy": 0.21858367025852204,
      "num_tokens": 46366244.0,
      "step": 25205
    },
    {
      "entropy": 5.092596435546875,
      "epoch": 2.1057467423989307,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004553877374766301,
      "loss": 4.7251,
      "mean_token_accuracy": 0.2169475793838501,
      "num_tokens": 46375319.0,
      "step": 25210
    },
    {
      "entropy": 5.059685087203979,
      "epoch": 2.106164383561644,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045536994251921535,
      "loss": 4.763,
      "mean_token_accuracy": 0.2171173632144928,
      "num_tokens": 46383870.0,
      "step": 25215
    },
    {
      "entropy": 5.080246448516846,
      "epoch": 2.1065820247243567,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004553521444042077,
      "loss": 4.7565,
      "mean_token_accuracy": 0.22442761510610582,
      "num_tokens": 46394104.0,
      "step": 25220
    },
    {
      "entropy": 5.034221124649048,
      "epoch": 2.10699966588707,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045533434313191866,
      "loss": 4.7237,
      "mean_token_accuracy": 0.2239132970571518,
      "num_tokens": 46403159.0,
      "step": 25225
    },
    {
      "entropy": 4.982892084121704,
      "epoch": 2.107417307049783,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045531653870265994,
      "loss": 4.7256,
      "mean_token_accuracy": 0.22744526267051696,
      "num_tokens": 46411806.0,
      "step": 25230
    },
    {
      "entropy": 5.027695417404175,
      "epoch": 2.1078349482124956,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004552987311167431,
      "loss": 4.6446,
      "mean_token_accuracy": 0.22974320948123933,
      "num_tokens": 46421156.0,
      "step": 25235
    },
    {
      "entropy": 5.061120986938477,
      "epoch": 2.108252589375209,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045528092037447994,
      "loss": 4.7214,
      "mean_token_accuracy": 0.2203056126832962,
      "num_tokens": 46430410.0,
      "step": 25240
    },
    {
      "entropy": 4.962077856063843,
      "epoch": 2.1086702305379217,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004552631064761822,
      "loss": 4.5919,
      "mean_token_accuracy": 0.23058743327856063,
      "num_tokens": 46438973.0,
      "step": 25245
    },
    {
      "entropy": 4.965929365158081,
      "epoch": 2.109087871700635,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004552452894221619,
      "loss": 4.7072,
      "mean_token_accuracy": 0.22096679508686065,
      "num_tokens": 46447851.0,
      "step": 25250
    },
    {
      "entropy": 5.0896343231201175,
      "epoch": 2.1095055128633478,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045522746921273084,
      "loss": 4.8478,
      "mean_token_accuracy": 0.2157930389046669,
      "num_tokens": 46457229.0,
      "step": 25255
    },
    {
      "entropy": 5.026849269866943,
      "epoch": 2.109923154026061,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045520964584820097,
      "loss": 4.6826,
      "mean_token_accuracy": 0.221237975358963,
      "num_tokens": 46466050.0,
      "step": 25260
    },
    {
      "entropy": 4.939265155792237,
      "epoch": 2.110340795188774,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004551918193288844,
      "loss": 4.6525,
      "mean_token_accuracy": 0.22865891605615615,
      "num_tokens": 46476685.0,
      "step": 25265
    },
    {
      "entropy": 5.0087892532348635,
      "epoch": 2.1107584363514866,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045517398965509297,
      "loss": 4.76,
      "mean_token_accuracy": 0.2182423561811447,
      "num_tokens": 46486427.0,
      "step": 25270
    },
    {
      "entropy": 5.114856004714966,
      "epoch": 2.1111760775142,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045515615682713913,
      "loss": 4.7802,
      "mean_token_accuracy": 0.2223707303404808,
      "num_tokens": 46495775.0,
      "step": 25275
    },
    {
      "entropy": 5.053496980667115,
      "epoch": 2.1115937186769127,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045513832084533484,
      "loss": 4.7168,
      "mean_token_accuracy": 0.2169124722480774,
      "num_tokens": 46504259.0,
      "step": 25280
    },
    {
      "entropy": 4.980295848846436,
      "epoch": 2.112011359839626,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004551204817099924,
      "loss": 4.7019,
      "mean_token_accuracy": 0.2297398641705513,
      "num_tokens": 46513365.0,
      "step": 25285
    },
    {
      "entropy": 5.081360101699829,
      "epoch": 2.1124290010023388,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045510263942142416,
      "loss": 4.7222,
      "mean_token_accuracy": 0.2279132515192032,
      "num_tokens": 46522756.0,
      "step": 25290
    },
    {
      "entropy": 5.016512107849121,
      "epoch": 2.1128466421650516,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004550847939799424,
      "loss": 4.7257,
      "mean_token_accuracy": 0.22278127074241638,
      "num_tokens": 46532700.0,
      "step": 25295
    },
    {
      "entropy": 5.129359531402588,
      "epoch": 2.113264283327765,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045506694538585956,
      "loss": 4.7465,
      "mean_token_accuracy": 0.22660114914178847,
      "num_tokens": 46542112.0,
      "step": 25300
    },
    {
      "entropy": 5.111936855316162,
      "epoch": 2.1136819244904776,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004550490936394881,
      "loss": 4.8281,
      "mean_token_accuracy": 0.21766289472579955,
      "num_tokens": 46550686.0,
      "step": 25305
    },
    {
      "entropy": 5.061140680313111,
      "epoch": 2.114099565653191,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004550312387411406,
      "loss": 4.8175,
      "mean_token_accuracy": 0.21993492990732194,
      "num_tokens": 46559861.0,
      "step": 25310
    },
    {
      "entropy": 5.0069256782531735,
      "epoch": 2.1145172068159037,
      "grad_norm": 1.125,
      "learning_rate": 0.0004550133806911295,
      "loss": 4.7062,
      "mean_token_accuracy": 0.2221523717045784,
      "num_tokens": 46569086.0,
      "step": 25315
    },
    {
      "entropy": 4.9739827632904055,
      "epoch": 2.114934847978617,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004549955194897674,
      "loss": 4.7176,
      "mean_token_accuracy": 0.22025089710950851,
      "num_tokens": 46578426.0,
      "step": 25320
    },
    {
      "entropy": 5.046280002593994,
      "epoch": 2.11535248914133,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045497765513736717,
      "loss": 4.7099,
      "mean_token_accuracy": 0.22239711731672288,
      "num_tokens": 46588632.0,
      "step": 25325
    },
    {
      "entropy": 5.01074366569519,
      "epoch": 2.1157701303040426,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045495978763424147,
      "loss": 4.7073,
      "mean_token_accuracy": 0.22887439876794816,
      "num_tokens": 46596585.0,
      "step": 25330
    },
    {
      "entropy": 5.113948678970337,
      "epoch": 2.116187771466756,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045494191698070295,
      "loss": 4.6923,
      "mean_token_accuracy": 0.22880735397338867,
      "num_tokens": 46605459.0,
      "step": 25335
    },
    {
      "entropy": 4.993125581741333,
      "epoch": 2.1166054126294687,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045492404317706473,
      "loss": 4.7508,
      "mean_token_accuracy": 0.22734970301389695,
      "num_tokens": 46614071.0,
      "step": 25340
    },
    {
      "entropy": 4.9590137004852295,
      "epoch": 2.117023053792182,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004549061662236394,
      "loss": 4.7451,
      "mean_token_accuracy": 0.22542113214731216,
      "num_tokens": 46622768.0,
      "step": 25345
    },
    {
      "entropy": 5.055737447738648,
      "epoch": 2.1174406949548947,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004548882861207402,
      "loss": 4.7531,
      "mean_token_accuracy": 0.21983543336391448,
      "num_tokens": 46632359.0,
      "step": 25350
    },
    {
      "entropy": 5.088540506362915,
      "epoch": 2.1178583361176075,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045487040286868,
      "loss": 4.7723,
      "mean_token_accuracy": 0.21867624819278716,
      "num_tokens": 46641583.0,
      "step": 25355
    },
    {
      "entropy": 5.089518976211548,
      "epoch": 2.118275977280321,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045485251646777187,
      "loss": 4.8126,
      "mean_token_accuracy": 0.21413386464118958,
      "num_tokens": 46651848.0,
      "step": 25360
    },
    {
      "entropy": 5.047911024093628,
      "epoch": 2.1186936184430336,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004548346269183289,
      "loss": 4.6741,
      "mean_token_accuracy": 0.22918954640626907,
      "num_tokens": 46660691.0,
      "step": 25365
    },
    {
      "entropy": 5.012687683105469,
      "epoch": 2.119111259605747,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045481673422066444,
      "loss": 4.7226,
      "mean_token_accuracy": 0.22155425548553467,
      "num_tokens": 46669656.0,
      "step": 25370
    },
    {
      "entropy": 5.037561130523682,
      "epoch": 2.1195289007684597,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004547988383750915,
      "loss": 4.7489,
      "mean_token_accuracy": 0.22227254509925842,
      "num_tokens": 46677906.0,
      "step": 25375
    },
    {
      "entropy": 4.982990360260009,
      "epoch": 2.119946541931173,
      "grad_norm": 1.125,
      "learning_rate": 0.0004547809393819235,
      "loss": 4.5557,
      "mean_token_accuracy": 0.24004893153905868,
      "num_tokens": 46686715.0,
      "step": 25380
    },
    {
      "entropy": 4.981684398651123,
      "epoch": 2.1203641830938857,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004547630372414737,
      "loss": 4.6908,
      "mean_token_accuracy": 0.2240722045302391,
      "num_tokens": 46695369.0,
      "step": 25385
    },
    {
      "entropy": 5.024152517318726,
      "epoch": 2.1207818242565986,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004547451319540557,
      "loss": 4.7191,
      "mean_token_accuracy": 0.2239597409963608,
      "num_tokens": 46705653.0,
      "step": 25390
    },
    {
      "entropy": 5.051331949234009,
      "epoch": 2.121199465419312,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004547272235199826,
      "loss": 4.7523,
      "mean_token_accuracy": 0.21861786395311356,
      "num_tokens": 46714886.0,
      "step": 25395
    },
    {
      "entropy": 5.005824136734009,
      "epoch": 2.1216171065820246,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004547093119395683,
      "loss": 4.7608,
      "mean_token_accuracy": 0.21582461893558502,
      "num_tokens": 46723837.0,
      "step": 25400
    },
    {
      "entropy": 5.038832712173462,
      "epoch": 2.122034747744738,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045469139721312607,
      "loss": 4.7858,
      "mean_token_accuracy": 0.22277941405773163,
      "num_tokens": 46734050.0,
      "step": 25405
    },
    {
      "entropy": 5.023769807815552,
      "epoch": 2.1224523889074507,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004546734793409697,
      "loss": 4.6662,
      "mean_token_accuracy": 0.22303376495838165,
      "num_tokens": 46743504.0,
      "step": 25410
    },
    {
      "entropy": 5.068182468414307,
      "epoch": 2.1228700300701635,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004546555583234128,
      "loss": 4.8084,
      "mean_token_accuracy": 0.21899864673614503,
      "num_tokens": 46751680.0,
      "step": 25415
    },
    {
      "entropy": 5.095439434051514,
      "epoch": 2.1232876712328768,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000454637634160769,
      "loss": 4.8103,
      "mean_token_accuracy": 0.21363236457109452,
      "num_tokens": 46761372.0,
      "step": 25420
    },
    {
      "entropy": 5.072255086898804,
      "epoch": 2.1237053123955896,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004546197068533523,
      "loss": 4.7161,
      "mean_token_accuracy": 0.22287018448114396,
      "num_tokens": 46771488.0,
      "step": 25425
    },
    {
      "entropy": 5.005947780609131,
      "epoch": 2.124122953558303,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004546017764014764,
      "loss": 4.6714,
      "mean_token_accuracy": 0.2169631078839302,
      "num_tokens": 46780676.0,
      "step": 25430
    },
    {
      "entropy": 5.020634698867798,
      "epoch": 2.1245405947210156,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004545838428054552,
      "loss": 4.7215,
      "mean_token_accuracy": 0.22238220125436783,
      "num_tokens": 46790129.0,
      "step": 25435
    },
    {
      "entropy": 5.102071762084961,
      "epoch": 2.124958235883729,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004545659060656027,
      "loss": 4.7819,
      "mean_token_accuracy": 0.22026288956403733,
      "num_tokens": 46799510.0,
      "step": 25440
    },
    {
      "entropy": 5.102435731887818,
      "epoch": 2.1253758770464417,
      "grad_norm": 1.125,
      "learning_rate": 0.0004545479661822328,
      "loss": 4.8229,
      "mean_token_accuracy": 0.22287532240152358,
      "num_tokens": 46808860.0,
      "step": 25445
    },
    {
      "entropy": 5.027861928939819,
      "epoch": 2.1257935182091545,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045453002315565965,
      "loss": 4.815,
      "mean_token_accuracy": 0.21839554607868195,
      "num_tokens": 46819648.0,
      "step": 25450
    },
    {
      "entropy": 5.067783546447754,
      "epoch": 2.1262111593718678,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004545120769861974,
      "loss": 4.712,
      "mean_token_accuracy": 0.2257791131734848,
      "num_tokens": 46827769.0,
      "step": 25455
    },
    {
      "entropy": 5.032239723205566,
      "epoch": 2.1266288005345806,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004544941276741601,
      "loss": 4.7505,
      "mean_token_accuracy": 0.23150816559791565,
      "num_tokens": 46837404.0,
      "step": 25460
    },
    {
      "entropy": 5.084584283828735,
      "epoch": 2.127046441697294,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045447617521986205,
      "loss": 4.8595,
      "mean_token_accuracy": 0.21560692191123962,
      "num_tokens": 46846419.0,
      "step": 25465
    },
    {
      "entropy": 4.977153491973877,
      "epoch": 2.1274640828600067,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004544582196236175,
      "loss": 4.6842,
      "mean_token_accuracy": 0.2227828949689865,
      "num_tokens": 46854648.0,
      "step": 25470
    },
    {
      "entropy": 5.090299081802368,
      "epoch": 2.1278817240227195,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004544402608857409,
      "loss": 4.7543,
      "mean_token_accuracy": 0.225632506608963,
      "num_tokens": 46864121.0,
      "step": 25475
    },
    {
      "entropy": 5.164941501617432,
      "epoch": 2.1282993651854327,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004544222990065464,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2112540990114212,
      "num_tokens": 46873157.0,
      "step": 25480
    },
    {
      "entropy": 5.061426496505737,
      "epoch": 2.1287170063481455,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045440433398634866,
      "loss": 4.6759,
      "mean_token_accuracy": 0.22830716669559478,
      "num_tokens": 46882050.0,
      "step": 25485
    },
    {
      "entropy": 4.994316577911377,
      "epoch": 2.129134647510859,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004543863658254622,
      "loss": 4.6971,
      "mean_token_accuracy": 0.2196570008993149,
      "num_tokens": 46891571.0,
      "step": 25490
    },
    {
      "entropy": 5.0749115467071535,
      "epoch": 2.1295522886735716,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004543683945242013,
      "loss": 4.8274,
      "mean_token_accuracy": 0.21338165998458863,
      "num_tokens": 46901437.0,
      "step": 25495
    },
    {
      "entropy": 5.105347967147827,
      "epoch": 2.129969929836285,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045435042008288093,
      "loss": 4.8392,
      "mean_token_accuracy": 0.22061317414045334,
      "num_tokens": 46909985.0,
      "step": 25500
    },
    {
      "entropy": 5.081341743469238,
      "epoch": 2.1303875709989977,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045433244250181535,
      "loss": 4.6871,
      "mean_token_accuracy": 0.22290479838848115,
      "num_tokens": 46919267.0,
      "step": 25505
    },
    {
      "entropy": 5.026331806182862,
      "epoch": 2.1308052121617105,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045431446178131966,
      "loss": 4.7748,
      "mean_token_accuracy": 0.21293215155601503,
      "num_tokens": 46928519.0,
      "step": 25510
    },
    {
      "entropy": 5.016300249099731,
      "epoch": 2.1312228533244237,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004542964779217085,
      "loss": 4.7642,
      "mean_token_accuracy": 0.22425772994756699,
      "num_tokens": 46937472.0,
      "step": 25515
    },
    {
      "entropy": 4.994348049163818,
      "epoch": 2.1316404944871366,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045427849092329655,
      "loss": 4.672,
      "mean_token_accuracy": 0.22262757867574692,
      "num_tokens": 46946607.0,
      "step": 25520
    },
    {
      "entropy": 4.9890645980834964,
      "epoch": 2.13205813564985,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045426050078639884,
      "loss": 4.6565,
      "mean_token_accuracy": 0.23301919996738435,
      "num_tokens": 46955254.0,
      "step": 25525
    },
    {
      "entropy": 5.016126585006714,
      "epoch": 2.1324757768125626,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004542425075113303,
      "loss": 4.7362,
      "mean_token_accuracy": 0.22269360572099686,
      "num_tokens": 46963982.0,
      "step": 25530
    },
    {
      "entropy": 5.054545545578003,
      "epoch": 2.132893417975276,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045422451109840604,
      "loss": 4.7001,
      "mean_token_accuracy": 0.22737773656845092,
      "num_tokens": 46972781.0,
      "step": 25535
    },
    {
      "entropy": 5.075618410110474,
      "epoch": 2.1333110591379887,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045420651154794086,
      "loss": 4.7036,
      "mean_token_accuracy": 0.22510286867618562,
      "num_tokens": 46982054.0,
      "step": 25540
    },
    {
      "entropy": 5.0440675735473635,
      "epoch": 2.1337287003007015,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045418850886025006,
      "loss": 4.7982,
      "mean_token_accuracy": 0.21655913293361664,
      "num_tokens": 46991800.0,
      "step": 25545
    },
    {
      "entropy": 5.0215826511383055,
      "epoch": 2.1341463414634148,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045417050303564867,
      "loss": 4.6735,
      "mean_token_accuracy": 0.2283475860953331,
      "num_tokens": 47000118.0,
      "step": 25550
    },
    {
      "entropy": 5.146001958847046,
      "epoch": 2.1345639826261276,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045415249407445196,
      "loss": 4.8537,
      "mean_token_accuracy": 0.20773697048425674,
      "num_tokens": 47009770.0,
      "step": 25555
    },
    {
      "entropy": 5.042645835876465,
      "epoch": 2.134981623788841,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004541344819769751,
      "loss": 4.7189,
      "mean_token_accuracy": 0.22472213208675385,
      "num_tokens": 47018050.0,
      "step": 25560
    },
    {
      "entropy": 4.973672485351562,
      "epoch": 2.1353992649515536,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045411646674353365,
      "loss": 4.7308,
      "mean_token_accuracy": 0.21954336613416672,
      "num_tokens": 47027092.0,
      "step": 25565
    },
    {
      "entropy": 5.036605501174927,
      "epoch": 2.1358169061142664,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045409844837444283,
      "loss": 4.727,
      "mean_token_accuracy": 0.22177117466926574,
      "num_tokens": 47036431.0,
      "step": 25570
    },
    {
      "entropy": 5.098110914230347,
      "epoch": 2.1362345472769797,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045408042687001806,
      "loss": 4.7603,
      "mean_token_accuracy": 0.2196684420108795,
      "num_tokens": 47045164.0,
      "step": 25575
    },
    {
      "entropy": 5.138502502441407,
      "epoch": 2.1366521884396925,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004540624022305749,
      "loss": 4.7188,
      "mean_token_accuracy": 0.21858872175216676,
      "num_tokens": 47054127.0,
      "step": 25580
    },
    {
      "entropy": 5.037451601028442,
      "epoch": 2.1370698296024058,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004540443744564288,
      "loss": 4.788,
      "mean_token_accuracy": 0.2168658047914505,
      "num_tokens": 47062504.0,
      "step": 25585
    },
    {
      "entropy": 5.067235517501831,
      "epoch": 2.1374874707651186,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004540263435478953,
      "loss": 4.8097,
      "mean_token_accuracy": 0.21432609111070633,
      "num_tokens": 47072507.0,
      "step": 25590
    },
    {
      "entropy": 5.003238534927368,
      "epoch": 2.1379051119278314,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045400830950529037,
      "loss": 4.7777,
      "mean_token_accuracy": 0.21877647936344147,
      "num_tokens": 47082249.0,
      "step": 25595
    },
    {
      "entropy": 5.007542657852173,
      "epoch": 2.1383227530905446,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004539902723289294,
      "loss": 4.7394,
      "mean_token_accuracy": 0.22507981359958648,
      "num_tokens": 47090878.0,
      "step": 25600
    },
    {
      "entropy": 5.110226726531982,
      "epoch": 2.1387403942532575,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045397223201912827,
      "loss": 4.7408,
      "mean_token_accuracy": 0.2247227296233177,
      "num_tokens": 47100852.0,
      "step": 25605
    },
    {
      "entropy": 5.035921049118042,
      "epoch": 2.1391580354159707,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004539541885762028,
      "loss": 4.6624,
      "mean_token_accuracy": 0.23449878990650178,
      "num_tokens": 47110037.0,
      "step": 25610
    },
    {
      "entropy": 4.9702249526977536,
      "epoch": 2.1395756765786835,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004539361420004689,
      "loss": 4.7118,
      "mean_token_accuracy": 0.2192573919892311,
      "num_tokens": 47119393.0,
      "step": 25615
    },
    {
      "entropy": 5.034157657623291,
      "epoch": 2.139993317741397,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045391809229224236,
      "loss": 4.7568,
      "mean_token_accuracy": 0.21957631707191466,
      "num_tokens": 47128270.0,
      "step": 25620
    },
    {
      "entropy": 5.014451169967652,
      "epoch": 2.1404109589041096,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004539000394518393,
      "loss": 4.6443,
      "mean_token_accuracy": 0.22912204265594482,
      "num_tokens": 47137104.0,
      "step": 25625
    },
    {
      "entropy": 4.9845202445983885,
      "epoch": 2.1408286000668224,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004538819834795757,
      "loss": 4.7845,
      "mean_token_accuracy": 0.21327209174633027,
      "num_tokens": 47145927.0,
      "step": 25630
    },
    {
      "entropy": 5.06554765701294,
      "epoch": 2.1412462412295357,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045386392437576765,
      "loss": 4.7211,
      "mean_token_accuracy": 0.2234214037656784,
      "num_tokens": 47154408.0,
      "step": 25635
    },
    {
      "entropy": 5.18594446182251,
      "epoch": 2.1416638823922485,
      "grad_norm": 1.125,
      "learning_rate": 0.00045384586214073134,
      "loss": 4.854,
      "mean_token_accuracy": 0.21011236160993577,
      "num_tokens": 47163928.0,
      "step": 25640
    },
    {
      "entropy": 5.025091886520386,
      "epoch": 2.1420815235549617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045382779677478294,
      "loss": 4.7045,
      "mean_token_accuracy": 0.22502790689468383,
      "num_tokens": 47173072.0,
      "step": 25645
    },
    {
      "entropy": 5.105824899673462,
      "epoch": 2.1424991647176745,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045380972827823866,
      "loss": 4.8011,
      "mean_token_accuracy": 0.21618063598871232,
      "num_tokens": 47182956.0,
      "step": 25650
    },
    {
      "entropy": 5.042319631576538,
      "epoch": 2.142916805880388,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004537916566514149,
      "loss": 4.8048,
      "mean_token_accuracy": 0.22782567292451858,
      "num_tokens": 47192083.0,
      "step": 25655
    },
    {
      "entropy": 5.004924631118774,
      "epoch": 2.1433344470431006,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004537735818946279,
      "loss": 4.6719,
      "mean_token_accuracy": 0.2265220195055008,
      "num_tokens": 47201639.0,
      "step": 25660
    },
    {
      "entropy": 5.063330507278442,
      "epoch": 2.1437520882058134,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004537555040081942,
      "loss": 4.7129,
      "mean_token_accuracy": 0.22333959341049195,
      "num_tokens": 47211522.0,
      "step": 25665
    },
    {
      "entropy": 5.10005464553833,
      "epoch": 2.1441697293685267,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045373742299243026,
      "loss": 4.7736,
      "mean_token_accuracy": 0.21850888878107072,
      "num_tokens": 47220788.0,
      "step": 25670
    },
    {
      "entropy": 5.0915628433227536,
      "epoch": 2.1445873705312395,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045371933884765246,
      "loss": 4.8813,
      "mean_token_accuracy": 0.20896528661251068,
      "num_tokens": 47230941.0,
      "step": 25675
    },
    {
      "entropy": 5.133160018920899,
      "epoch": 2.1450050116939527,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004537012515741777,
      "loss": 4.7607,
      "mean_token_accuracy": 0.22295261323451995,
      "num_tokens": 47240294.0,
      "step": 25680
    },
    {
      "entropy": 5.067159700393677,
      "epoch": 2.1454226528566656,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004536831611723223,
      "loss": 4.7744,
      "mean_token_accuracy": 0.22024628818035125,
      "num_tokens": 47248757.0,
      "step": 25685
    },
    {
      "entropy": 5.077360153198242,
      "epoch": 2.1458402940193784,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004536650676424032,
      "loss": 4.8003,
      "mean_token_accuracy": 0.20792748779058456,
      "num_tokens": 47258836.0,
      "step": 25690
    },
    {
      "entropy": 5.028138208389282,
      "epoch": 2.1462579351820916,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004536469709847369,
      "loss": 4.6522,
      "mean_token_accuracy": 0.2316773369908333,
      "num_tokens": 47267624.0,
      "step": 25695
    },
    {
      "entropy": 4.90849347114563,
      "epoch": 2.1466755763448044,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004536288711996404,
      "loss": 4.6059,
      "mean_token_accuracy": 0.22771304547786714,
      "num_tokens": 47277480.0,
      "step": 25700
    },
    {
      "entropy": 5.049386739730835,
      "epoch": 2.1470932175075177,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045361076828743054,
      "loss": 4.7751,
      "mean_token_accuracy": 0.2196110889315605,
      "num_tokens": 47286920.0,
      "step": 25705
    },
    {
      "entropy": 5.076989984512329,
      "epoch": 2.1475108586702305,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004535926622484242,
      "loss": 4.6781,
      "mean_token_accuracy": 0.23220635205507278,
      "num_tokens": 47295994.0,
      "step": 25710
    },
    {
      "entropy": 4.969162797927856,
      "epoch": 2.1479284998329433,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045357455308293816,
      "loss": 4.6204,
      "mean_token_accuracy": 0.23474421203136445,
      "num_tokens": 47304503.0,
      "step": 25715
    },
    {
      "entropy": 4.920914030075073,
      "epoch": 2.1483461409956566,
      "grad_norm": 1.25,
      "learning_rate": 0.0004535564407912898,
      "loss": 4.6222,
      "mean_token_accuracy": 0.23969502598047257,
      "num_tokens": 47312785.0,
      "step": 25720
    },
    {
      "entropy": 4.990327167510986,
      "epoch": 2.1487637821583694,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004535383253737959,
      "loss": 4.6694,
      "mean_token_accuracy": 0.22400972694158555,
      "num_tokens": 47322050.0,
      "step": 25725
    },
    {
      "entropy": 5.079989051818847,
      "epoch": 2.1491814233210826,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004535202068307737,
      "loss": 4.8009,
      "mean_token_accuracy": 0.21830984503030776,
      "num_tokens": 47331877.0,
      "step": 25730
    },
    {
      "entropy": 4.988127660751343,
      "epoch": 2.1495990644837955,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004535020851625405,
      "loss": 4.6787,
      "mean_token_accuracy": 0.2313017576932907,
      "num_tokens": 47340788.0,
      "step": 25735
    },
    {
      "entropy": 4.877593326568603,
      "epoch": 2.1500167056465087,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045348396036941336,
      "loss": 4.5341,
      "mean_token_accuracy": 0.2423727422952652,
      "num_tokens": 47349136.0,
      "step": 25740
    },
    {
      "entropy": 5.05623688697815,
      "epoch": 2.1504343468092215,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004534658324517097,
      "loss": 4.8598,
      "mean_token_accuracy": 0.2160092517733574,
      "num_tokens": 47358634.0,
      "step": 25745
    },
    {
      "entropy": 5.028234910964966,
      "epoch": 2.1508519879719343,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004534477014097468,
      "loss": 4.7014,
      "mean_token_accuracy": 0.2169150248169899,
      "num_tokens": 47368068.0,
      "step": 25750
    },
    {
      "entropy": 5.105899143218994,
      "epoch": 2.1512696291346476,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045342956724384205,
      "loss": 4.7545,
      "mean_token_accuracy": 0.22277440428733825,
      "num_tokens": 47377456.0,
      "step": 25755
    },
    {
      "entropy": 5.069699716567993,
      "epoch": 2.1516872702973604,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000453411429954313,
      "loss": 4.7449,
      "mean_token_accuracy": 0.2209816724061966,
      "num_tokens": 47386013.0,
      "step": 25760
    },
    {
      "entropy": 5.091969156265259,
      "epoch": 2.1521049114600737,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004533932895414771,
      "loss": 4.703,
      "mean_token_accuracy": 0.230768983066082,
      "num_tokens": 47394939.0,
      "step": 25765
    },
    {
      "entropy": 5.097317886352539,
      "epoch": 2.1525225526227865,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045337514600565196,
      "loss": 4.8476,
      "mean_token_accuracy": 0.2081403985619545,
      "num_tokens": 47404469.0,
      "step": 25770
    },
    {
      "entropy": 5.1118812084198,
      "epoch": 2.1529401937854997,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045335699934715513,
      "loss": 4.7905,
      "mean_token_accuracy": 0.21911909729242324,
      "num_tokens": 47414518.0,
      "step": 25775
    },
    {
      "entropy": 5.061270236968994,
      "epoch": 2.1533578349482125,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045333884956630437,
      "loss": 4.7452,
      "mean_token_accuracy": 0.2191556289792061,
      "num_tokens": 47423264.0,
      "step": 25780
    },
    {
      "entropy": 4.997159194946289,
      "epoch": 2.1537754761109253,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045332069666341733,
      "loss": 4.6523,
      "mean_token_accuracy": 0.2306860089302063,
      "num_tokens": 47431989.0,
      "step": 25785
    },
    {
      "entropy": 5.0001500129699705,
      "epoch": 2.1541931172736386,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004533025406388119,
      "loss": 4.6203,
      "mean_token_accuracy": 0.22617037892341613,
      "num_tokens": 47440911.0,
      "step": 25790
    },
    {
      "entropy": 4.978374004364014,
      "epoch": 2.1546107584363514,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004532843814928059,
      "loss": 4.7004,
      "mean_token_accuracy": 0.22791739404201508,
      "num_tokens": 47450667.0,
      "step": 25795
    },
    {
      "entropy": 5.104659700393677,
      "epoch": 2.1550283995990647,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004532662192257172,
      "loss": 4.8033,
      "mean_token_accuracy": 0.2163358137011528,
      "num_tokens": 47459950.0,
      "step": 25800
    },
    {
      "entropy": 5.06749005317688,
      "epoch": 2.1554460407617775,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004532480538378637,
      "loss": 4.8206,
      "mean_token_accuracy": 0.21759223490953444,
      "num_tokens": 47468770.0,
      "step": 25805
    },
    {
      "entropy": 4.987118673324585,
      "epoch": 2.1558636819244903,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045322988532956344,
      "loss": 4.6855,
      "mean_token_accuracy": 0.22813877165317537,
      "num_tokens": 47478862.0,
      "step": 25810
    },
    {
      "entropy": 5.033622932434082,
      "epoch": 2.1562813230872035,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045321171370113456,
      "loss": 4.6557,
      "mean_token_accuracy": 0.22455999851226807,
      "num_tokens": 47487674.0,
      "step": 25815
    },
    {
      "entropy": 4.984213018417359,
      "epoch": 2.1566989642499164,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004531935389528951,
      "loss": 4.6743,
      "mean_token_accuracy": 0.2265433207154274,
      "num_tokens": 47496127.0,
      "step": 25820
    },
    {
      "entropy": 5.0562128067016605,
      "epoch": 2.1571166054126296,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004531753610851633,
      "loss": 4.6886,
      "mean_token_accuracy": 0.22381995171308516,
      "num_tokens": 47506027.0,
      "step": 25825
    },
    {
      "entropy": 5.090599155426025,
      "epoch": 2.1575342465753424,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004531571800982573,
      "loss": 4.7064,
      "mean_token_accuracy": 0.2216132551431656,
      "num_tokens": 47516108.0,
      "step": 25830
    },
    {
      "entropy": 4.981888341903686,
      "epoch": 2.1579518877380552,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045313899599249536,
      "loss": 4.6866,
      "mean_token_accuracy": 0.2278309315443039,
      "num_tokens": 47524322.0,
      "step": 25835
    },
    {
      "entropy": 4.975980377197265,
      "epoch": 2.1583695289007685,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045312080876819597,
      "loss": 4.723,
      "mean_token_accuracy": 0.22355493754148484,
      "num_tokens": 47533718.0,
      "step": 25840
    },
    {
      "entropy": 5.065723991394043,
      "epoch": 2.1587871700634813,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004531026184256773,
      "loss": 4.7626,
      "mean_token_accuracy": 0.21711696684360504,
      "num_tokens": 47542369.0,
      "step": 25845
    },
    {
      "entropy": 5.097001791000366,
      "epoch": 2.1592048112261946,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000453084424965258,
      "loss": 4.71,
      "mean_token_accuracy": 0.21659397035837175,
      "num_tokens": 47551816.0,
      "step": 25850
    },
    {
      "entropy": 5.002888298034668,
      "epoch": 2.1596224523889074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045306622838725644,
      "loss": 4.725,
      "mean_token_accuracy": 0.2307705193758011,
      "num_tokens": 47561402.0,
      "step": 25855
    },
    {
      "entropy": 4.995731544494629,
      "epoch": 2.1600400935516206,
      "grad_norm": 1.125,
      "learning_rate": 0.0004530480286919912,
      "loss": 4.7318,
      "mean_token_accuracy": 0.22215882539749146,
      "num_tokens": 47570525.0,
      "step": 25860
    },
    {
      "entropy": 4.98984842300415,
      "epoch": 2.1604577347143334,
      "grad_norm": 1.078125,
      "learning_rate": 0.000453029825879781,
      "loss": 4.6481,
      "mean_token_accuracy": 0.22667524367570877,
      "num_tokens": 47579835.0,
      "step": 25865
    },
    {
      "entropy": 5.113577270507813,
      "epoch": 2.1608753758770463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045301161995094433,
      "loss": 4.7735,
      "mean_token_accuracy": 0.21908335387706757,
      "num_tokens": 47589627.0,
      "step": 25870
    },
    {
      "entropy": 5.090448474884033,
      "epoch": 2.1612930170397595,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045299341090579995,
      "loss": 4.7519,
      "mean_token_accuracy": 0.22245831191539764,
      "num_tokens": 47598006.0,
      "step": 25875
    },
    {
      "entropy": 4.929087591171265,
      "epoch": 2.1617106582024723,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004529751987446667,
      "loss": 4.6307,
      "mean_token_accuracy": 0.23614456057548522,
      "num_tokens": 47606455.0,
      "step": 25880
    },
    {
      "entropy": 5.054486894607544,
      "epoch": 2.1621282993651856,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004529569834678633,
      "loss": 4.8218,
      "mean_token_accuracy": 0.21566896438598632,
      "num_tokens": 47615697.0,
      "step": 25885
    },
    {
      "entropy": 5.176187324523926,
      "epoch": 2.1625459405278984,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004529387650757088,
      "loss": 4.7449,
      "mean_token_accuracy": 0.21743777096271516,
      "num_tokens": 47625499.0,
      "step": 25890
    },
    {
      "entropy": 5.039952230453491,
      "epoch": 2.1629635816906116,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045292054356852195,
      "loss": 4.6611,
      "mean_token_accuracy": 0.22504417300224305,
      "num_tokens": 47634627.0,
      "step": 25895
    },
    {
      "entropy": 4.958221817016602,
      "epoch": 2.1633812228533245,
      "grad_norm": 1.125,
      "learning_rate": 0.0004529023189466219,
      "loss": 4.6898,
      "mean_token_accuracy": 0.23849308341741562,
      "num_tokens": 47642628.0,
      "step": 25900
    },
    {
      "entropy": 4.951693677902222,
      "epoch": 2.1637988640160373,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045288409121032756,
      "loss": 4.678,
      "mean_token_accuracy": 0.2203935220837593,
      "num_tokens": 47651654.0,
      "step": 25905
    },
    {
      "entropy": 5.049699926376343,
      "epoch": 2.1642165051787505,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045286586035995813,
      "loss": 4.6571,
      "mean_token_accuracy": 0.2248750150203705,
      "num_tokens": 47661092.0,
      "step": 25910
    },
    {
      "entropy": 4.997106266021729,
      "epoch": 2.1646341463414633,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045284762639583265,
      "loss": 4.6984,
      "mean_token_accuracy": 0.22213085740804672,
      "num_tokens": 47670141.0,
      "step": 25915
    },
    {
      "entropy": 5.025136566162109,
      "epoch": 2.1650517875041766,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045282938931827046,
      "loss": 4.7122,
      "mean_token_accuracy": 0.226036337018013,
      "num_tokens": 47679068.0,
      "step": 25920
    },
    {
      "entropy": 5.0624174118042,
      "epoch": 2.1654694286668894,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004528111491275907,
      "loss": 4.6673,
      "mean_token_accuracy": 0.22605119049549102,
      "num_tokens": 47687337.0,
      "step": 25925
    },
    {
      "entropy": 4.986572647094727,
      "epoch": 2.165887069829602,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004527929058241128,
      "loss": 4.7249,
      "mean_token_accuracy": 0.21652610898017882,
      "num_tokens": 47696938.0,
      "step": 25930
    },
    {
      "entropy": 5.0484827041625975,
      "epoch": 2.1663047109923155,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045277465940815603,
      "loss": 4.8134,
      "mean_token_accuracy": 0.215667225420475,
      "num_tokens": 47707125.0,
      "step": 25935
    },
    {
      "entropy": 5.045136213302612,
      "epoch": 2.1667223521550283,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004527564098800399,
      "loss": 4.7495,
      "mean_token_accuracy": 0.21652752459049224,
      "num_tokens": 47716958.0,
      "step": 25940
    },
    {
      "entropy": 4.98993558883667,
      "epoch": 2.1671399933177415,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004527381572400838,
      "loss": 4.6552,
      "mean_token_accuracy": 0.22682733684778214,
      "num_tokens": 47726177.0,
      "step": 25945
    },
    {
      "entropy": 5.087247514724732,
      "epoch": 2.1675576344804544,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045271990148860737,
      "loss": 4.7273,
      "mean_token_accuracy": 0.22429073005914688,
      "num_tokens": 47734103.0,
      "step": 25950
    },
    {
      "entropy": 4.9990159034729,
      "epoch": 2.167975275643167,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045270164262593013,
      "loss": 4.644,
      "mean_token_accuracy": 0.23655765503644943,
      "num_tokens": 47743058.0,
      "step": 25955
    },
    {
      "entropy": 5.0365297317504885,
      "epoch": 2.1683929168058804,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004526833806523717,
      "loss": 4.7933,
      "mean_token_accuracy": 0.21666575074195862,
      "num_tokens": 47753031.0,
      "step": 25960
    },
    {
      "entropy": 5.004342079162598,
      "epoch": 2.1688105579685932,
      "grad_norm": 1.125,
      "learning_rate": 0.00045266511556825184,
      "loss": 4.6899,
      "mean_token_accuracy": 0.2163880690932274,
      "num_tokens": 47761403.0,
      "step": 25965
    },
    {
      "entropy": 5.098011589050293,
      "epoch": 2.1692281991313065,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045264684737389026,
      "loss": 4.8054,
      "mean_token_accuracy": 0.21810798048973085,
      "num_tokens": 47770445.0,
      "step": 25970
    },
    {
      "entropy": 5.158545875549317,
      "epoch": 2.1696458402940193,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045262857606960684,
      "loss": 4.8666,
      "mean_token_accuracy": 0.21248743832111358,
      "num_tokens": 47779497.0,
      "step": 25975
    },
    {
      "entropy": 5.09198899269104,
      "epoch": 2.1700634814567326,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045261030165572134,
      "loss": 4.6968,
      "mean_token_accuracy": 0.22365068942308425,
      "num_tokens": 47788240.0,
      "step": 25980
    },
    {
      "entropy": 5.084444618225097,
      "epoch": 2.1704811226194454,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004525920241325537,
      "loss": 4.7759,
      "mean_token_accuracy": 0.2205178290605545,
      "num_tokens": 47797057.0,
      "step": 25985
    },
    {
      "entropy": 5.031615257263184,
      "epoch": 2.170898763782158,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004525737435004239,
      "loss": 4.7347,
      "mean_token_accuracy": 0.2151290714740753,
      "num_tokens": 47807327.0,
      "step": 25990
    },
    {
      "entropy": 5.092732524871826,
      "epoch": 2.1713164049448714,
      "grad_norm": 1.046875,
      "learning_rate": 0.000452555459759652,
      "loss": 4.7516,
      "mean_token_accuracy": 0.21842468827962874,
      "num_tokens": 47816938.0,
      "step": 25995
    },
    {
      "entropy": 5.063276243209839,
      "epoch": 2.1717340461075842,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004525371729105581,
      "loss": 4.7435,
      "mean_token_accuracy": 0.21239256411790847,
      "num_tokens": 47825956.0,
      "step": 26000
    },
    {
      "entropy": 5.059375953674317,
      "epoch": 2.1721516872702975,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004525188829534622,
      "loss": 4.802,
      "mean_token_accuracy": 0.2213391050696373,
      "num_tokens": 47836623.0,
      "step": 26005
    },
    {
      "entropy": 5.132289218902588,
      "epoch": 2.1725693284330103,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004525005898886847,
      "loss": 4.7501,
      "mean_token_accuracy": 0.2157948449254036,
      "num_tokens": 47844679.0,
      "step": 26010
    },
    {
      "entropy": 5.076801347732544,
      "epoch": 2.1729869695957236,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004524822937165456,
      "loss": 4.7083,
      "mean_token_accuracy": 0.22190575301647186,
      "num_tokens": 47853933.0,
      "step": 26015
    },
    {
      "entropy": 5.045118188858032,
      "epoch": 2.1734046107584364,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045246399443736536,
      "loss": 4.7916,
      "mean_token_accuracy": 0.21871333867311477,
      "num_tokens": 47863608.0,
      "step": 26020
    },
    {
      "entropy": 4.971820116043091,
      "epoch": 2.173822251921149,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045244569205146425,
      "loss": 4.6314,
      "mean_token_accuracy": 0.23410231620073318,
      "num_tokens": 47872275.0,
      "step": 26025
    },
    {
      "entropy": 5.148814392089844,
      "epoch": 2.1742398930838625,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004524273865591627,
      "loss": 4.849,
      "mean_token_accuracy": 0.215231853723526,
      "num_tokens": 47882046.0,
      "step": 26030
    },
    {
      "entropy": 4.9654457569122314,
      "epoch": 2.1746575342465753,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045240907796078125,
      "loss": 4.6911,
      "mean_token_accuracy": 0.2305859476327896,
      "num_tokens": 47891154.0,
      "step": 26035
    },
    {
      "entropy": 5.111255550384522,
      "epoch": 2.1750751754092885,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004523907662566403,
      "loss": 4.7888,
      "mean_token_accuracy": 0.21194939613342284,
      "num_tokens": 47901052.0,
      "step": 26040
    },
    {
      "entropy": 5.155873966217041,
      "epoch": 2.1754928165720013,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045237245144706044,
      "loss": 4.7427,
      "mean_token_accuracy": 0.211312435567379,
      "num_tokens": 47910370.0,
      "step": 26045
    },
    {
      "entropy": 4.9619184017181395,
      "epoch": 2.175910457734714,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004523541335323624,
      "loss": 4.6153,
      "mean_token_accuracy": 0.233388315141201,
      "num_tokens": 47920102.0,
      "step": 26050
    },
    {
      "entropy": 4.967746543884277,
      "epoch": 2.1763280988974274,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004523358125128667,
      "loss": 4.6778,
      "mean_token_accuracy": 0.21954386979341506,
      "num_tokens": 47929338.0,
      "step": 26055
    },
    {
      "entropy": 5.1058595180511475,
      "epoch": 2.17674574006014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004523174883888941,
      "loss": 4.7318,
      "mean_token_accuracy": 0.22191091775894164,
      "num_tokens": 47937718.0,
      "step": 26060
    },
    {
      "entropy": 5.048461151123047,
      "epoch": 2.1771633812228535,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004522991611607655,
      "loss": 4.6654,
      "mean_token_accuracy": 0.2276118814945221,
      "num_tokens": 47946921.0,
      "step": 26065
    },
    {
      "entropy": 4.925285196304321,
      "epoch": 2.1775810223855663,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045228083082880164,
      "loss": 4.6717,
      "mean_token_accuracy": 0.23225207477808,
      "num_tokens": 47956680.0,
      "step": 26070
    },
    {
      "entropy": 5.045437145233154,
      "epoch": 2.177998663548279,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004522624973933234,
      "loss": 4.6784,
      "mean_token_accuracy": 0.22056327909231185,
      "num_tokens": 47965425.0,
      "step": 26075
    },
    {
      "entropy": 5.130672359466553,
      "epoch": 2.1784163047109923,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045224416085465195,
      "loss": 4.7941,
      "mean_token_accuracy": 0.2163210168480873,
      "num_tokens": 47974756.0,
      "step": 26080
    },
    {
      "entropy": 5.056910371780395,
      "epoch": 2.178833945873705,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045222582121310793,
      "loss": 4.7134,
      "mean_token_accuracy": 0.21759443134069442,
      "num_tokens": 47983896.0,
      "step": 26085
    },
    {
      "entropy": 4.968153429031372,
      "epoch": 2.1792515870364184,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045220747846901266,
      "loss": 4.6134,
      "mean_token_accuracy": 0.2303548589348793,
      "num_tokens": 47993987.0,
      "step": 26090
    },
    {
      "entropy": 4.970527839660645,
      "epoch": 2.1796692281991312,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045218913262268717,
      "loss": 4.6721,
      "mean_token_accuracy": 0.22954030930995942,
      "num_tokens": 48003339.0,
      "step": 26095
    },
    {
      "entropy": 5.029125833511353,
      "epoch": 2.1800868693618445,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045217078367445264,
      "loss": 4.7382,
      "mean_token_accuracy": 0.22305805385112762,
      "num_tokens": 48012679.0,
      "step": 26100
    },
    {
      "entropy": 5.037301445007325,
      "epoch": 2.1805045105245573,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045215243162463023,
      "loss": 4.7222,
      "mean_token_accuracy": 0.22438310831785202,
      "num_tokens": 48022573.0,
      "step": 26105
    },
    {
      "entropy": 5.010870409011841,
      "epoch": 2.18092215168727,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004521340764735413,
      "loss": 4.6354,
      "mean_token_accuracy": 0.22964758425951004,
      "num_tokens": 48030834.0,
      "step": 26110
    },
    {
      "entropy": 5.036025285720825,
      "epoch": 2.1813397928499834,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004521157182215071,
      "loss": 4.8063,
      "mean_token_accuracy": 0.21278886497020721,
      "num_tokens": 48040159.0,
      "step": 26115
    },
    {
      "entropy": 4.947514772415161,
      "epoch": 2.181757434012696,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045209735686884913,
      "loss": 4.6402,
      "mean_token_accuracy": 0.2257721707224846,
      "num_tokens": 48049478.0,
      "step": 26120
    },
    {
      "entropy": 5.061100912094116,
      "epoch": 2.1821750751754094,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045207899241588874,
      "loss": 4.7464,
      "mean_token_accuracy": 0.21966032385826112,
      "num_tokens": 48059197.0,
      "step": 26125
    },
    {
      "entropy": 5.036784076690674,
      "epoch": 2.1825927163381222,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004520606248629474,
      "loss": 4.7393,
      "mean_token_accuracy": 0.22168187350034713,
      "num_tokens": 48067794.0,
      "step": 26130
    },
    {
      "entropy": 5.059645891189575,
      "epoch": 2.1830103575008355,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004520422542103467,
      "loss": 4.7306,
      "mean_token_accuracy": 0.21913575977087021,
      "num_tokens": 48077224.0,
      "step": 26135
    },
    {
      "entropy": 5.072814846038819,
      "epoch": 2.1834279986635483,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004520238804584083,
      "loss": 4.7405,
      "mean_token_accuracy": 0.22666552513837815,
      "num_tokens": 48087311.0,
      "step": 26140
    },
    {
      "entropy": 4.982792377471924,
      "epoch": 2.183845639826261,
      "grad_norm": 1.125,
      "learning_rate": 0.0004520055036074538,
      "loss": 4.649,
      "mean_token_accuracy": 0.2240900844335556,
      "num_tokens": 48095803.0,
      "step": 26145
    },
    {
      "entropy": 5.101936388015747,
      "epoch": 2.1842632809889744,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004519871236578048,
      "loss": 4.8109,
      "mean_token_accuracy": 0.2198093369603157,
      "num_tokens": 48105281.0,
      "step": 26150
    },
    {
      "entropy": 5.0367717266082765,
      "epoch": 2.184680922151687,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004519687406097832,
      "loss": 4.7124,
      "mean_token_accuracy": 0.21956146210432054,
      "num_tokens": 48114835.0,
      "step": 26155
    },
    {
      "entropy": 5.061205911636352,
      "epoch": 2.1850985633144004,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004519503544637107,
      "loss": 4.7645,
      "mean_token_accuracy": 0.22384793013334275,
      "num_tokens": 48123194.0,
      "step": 26160
    },
    {
      "entropy": 4.966100120544434,
      "epoch": 2.1855162044771133,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045193196521990935,
      "loss": 4.5725,
      "mean_token_accuracy": 0.2335738867521286,
      "num_tokens": 48132286.0,
      "step": 26165
    },
    {
      "entropy": 5.034630060195923,
      "epoch": 2.185933845639826,
      "grad_norm": 1.140625,
      "learning_rate": 0.000451913572878701,
      "loss": 4.7619,
      "mean_token_accuracy": 0.22956983596086503,
      "num_tokens": 48141166.0,
      "step": 26170
    },
    {
      "entropy": 5.03041706085205,
      "epoch": 2.1863514868025393,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004518951774404076,
      "loss": 4.668,
      "mean_token_accuracy": 0.226115782558918,
      "num_tokens": 48150809.0,
      "step": 26175
    },
    {
      "entropy": 5.156037664413452,
      "epoch": 2.186769127965252,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004518767789053511,
      "loss": 4.8806,
      "mean_token_accuracy": 0.20850945711135865,
      "num_tokens": 48159786.0,
      "step": 26180
    },
    {
      "entropy": 4.9849272727966305,
      "epoch": 2.1871867691279654,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045185837727385373,
      "loss": 4.7076,
      "mean_token_accuracy": 0.22940724492073059,
      "num_tokens": 48169543.0,
      "step": 26185
    },
    {
      "entropy": 5.0291272640228275,
      "epoch": 2.187604410290678,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004518399725462376,
      "loss": 4.7039,
      "mean_token_accuracy": 0.22716838568449021,
      "num_tokens": 48178282.0,
      "step": 26190
    },
    {
      "entropy": 5.142953872680664,
      "epoch": 2.188022051453391,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004518215647228248,
      "loss": 4.8367,
      "mean_token_accuracy": 0.21483120769262315,
      "num_tokens": 48188673.0,
      "step": 26195
    },
    {
      "entropy": 5.11625566482544,
      "epoch": 2.1884396926161043,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045180315380393776,
      "loss": 4.7689,
      "mean_token_accuracy": 0.21550682932138443,
      "num_tokens": 48198596.0,
      "step": 26200
    },
    {
      "entropy": 5.021616506576538,
      "epoch": 2.188857333778817,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004517847397898987,
      "loss": 4.6763,
      "mean_token_accuracy": 0.22570033073425294,
      "num_tokens": 48207585.0,
      "step": 26205
    },
    {
      "entropy": 5.071789789199829,
      "epoch": 2.1892749749415303,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045176632268102994,
      "loss": 4.8337,
      "mean_token_accuracy": 0.21204297840595246,
      "num_tokens": 48217804.0,
      "step": 26210
    },
    {
      "entropy": 5.065474367141723,
      "epoch": 2.189692616104243,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000451747902477654,
      "loss": 4.7455,
      "mean_token_accuracy": 0.22408876717090606,
      "num_tokens": 48228363.0,
      "step": 26215
    },
    {
      "entropy": 5.100418519973755,
      "epoch": 2.1901102572669564,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045172947918009317,
      "loss": 4.6852,
      "mean_token_accuracy": 0.22527474761009217,
      "num_tokens": 48237965.0,
      "step": 26220
    },
    {
      "entropy": 4.95916166305542,
      "epoch": 2.190527898429669,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004517110527886701,
      "loss": 4.6859,
      "mean_token_accuracy": 0.22253114283084868,
      "num_tokens": 48247455.0,
      "step": 26225
    },
    {
      "entropy": 5.0559412956237795,
      "epoch": 2.190945539592382,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004516926233037074,
      "loss": 4.8309,
      "mean_token_accuracy": 0.21111894994974137,
      "num_tokens": 48256543.0,
      "step": 26230
    },
    {
      "entropy": 5.073107004165649,
      "epoch": 2.1913631807550953,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004516741907255276,
      "loss": 4.7019,
      "mean_token_accuracy": 0.2260959342122078,
      "num_tokens": 48265632.0,
      "step": 26235
    },
    {
      "entropy": 5.1101585865020756,
      "epoch": 2.191780821917808,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004516557550544534,
      "loss": 4.7557,
      "mean_token_accuracy": 0.22221297174692153,
      "num_tokens": 48274877.0,
      "step": 26240
    },
    {
      "entropy": 5.074300909042359,
      "epoch": 2.1921984630805214,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004516373162908077,
      "loss": 4.7676,
      "mean_token_accuracy": 0.21734315156936646,
      "num_tokens": 48284411.0,
      "step": 26245
    },
    {
      "entropy": 4.916233730316162,
      "epoch": 2.192616104243234,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045161887443491297,
      "loss": 4.5561,
      "mean_token_accuracy": 0.23294610381126404,
      "num_tokens": 48293128.0,
      "step": 26250
    },
    {
      "entropy": 4.960615158081055,
      "epoch": 2.1930337454059474,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045160042948709235,
      "loss": 4.7742,
      "mean_token_accuracy": 0.2238401636481285,
      "num_tokens": 48302751.0,
      "step": 26255
    },
    {
      "entropy": 5.05275297164917,
      "epoch": 2.1934513865686602,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045158198144766863,
      "loss": 4.7082,
      "mean_token_accuracy": 0.22987052351236342,
      "num_tokens": 48312119.0,
      "step": 26260
    },
    {
      "entropy": 5.021503067016601,
      "epoch": 2.193869027731373,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004515635303169648,
      "loss": 4.6968,
      "mean_token_accuracy": 0.22254738956689835,
      "num_tokens": 48322346.0,
      "step": 26265
    },
    {
      "entropy": 5.032074451446533,
      "epoch": 2.1942866688940863,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045154507609530375,
      "loss": 4.6797,
      "mean_token_accuracy": 0.22513826191425323,
      "num_tokens": 48330852.0,
      "step": 26270
    },
    {
      "entropy": 4.992886066436768,
      "epoch": 2.194704310056799,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004515266187830087,
      "loss": 4.6815,
      "mean_token_accuracy": 0.2291284218430519,
      "num_tokens": 48339777.0,
      "step": 26275
    },
    {
      "entropy": 5.057147979736328,
      "epoch": 2.1951219512195124,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045150815838040275,
      "loss": 4.7747,
      "mean_token_accuracy": 0.21995357871055604,
      "num_tokens": 48349096.0,
      "step": 26280
    },
    {
      "entropy": 4.975469493865967,
      "epoch": 2.195539592382225,
      "grad_norm": 1.125,
      "learning_rate": 0.00045148969488780893,
      "loss": 4.6472,
      "mean_token_accuracy": 0.2335678294301033,
      "num_tokens": 48357683.0,
      "step": 26285
    },
    {
      "entropy": 4.98939528465271,
      "epoch": 2.195957233544938,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004514712283055506,
      "loss": 4.6562,
      "mean_token_accuracy": 0.22841884344816207,
      "num_tokens": 48365998.0,
      "step": 26290
    },
    {
      "entropy": 5.103862476348877,
      "epoch": 2.1963748747076512,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004514527586339511,
      "loss": 4.7335,
      "mean_token_accuracy": 0.22874172478914262,
      "num_tokens": 48374439.0,
      "step": 26295
    },
    {
      "entropy": 5.057086038589477,
      "epoch": 2.196792515870364,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045143428587333354,
      "loss": 4.7844,
      "mean_token_accuracy": 0.21651051491498946,
      "num_tokens": 48384219.0,
      "step": 26300
    },
    {
      "entropy": 5.031060838699341,
      "epoch": 2.1972101570330773,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004514158100240214,
      "loss": 4.6848,
      "mean_token_accuracy": 0.22773373872041702,
      "num_tokens": 48393513.0,
      "step": 26305
    },
    {
      "entropy": 4.952331972122193,
      "epoch": 2.19762779819579,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004513973310863383,
      "loss": 4.6513,
      "mean_token_accuracy": 0.22852333188056945,
      "num_tokens": 48402221.0,
      "step": 26310
    },
    {
      "entropy": 5.0974287509918215,
      "epoch": 2.198045439358503,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045137884906060747,
      "loss": 4.826,
      "mean_token_accuracy": 0.2187034770846367,
      "num_tokens": 48412240.0,
      "step": 26315
    },
    {
      "entropy": 5.032522249221802,
      "epoch": 2.198463080521216,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004513603639471526,
      "loss": 4.7503,
      "mean_token_accuracy": 0.21727099269628525,
      "num_tokens": 48420691.0,
      "step": 26320
    },
    {
      "entropy": 5.078086185455322,
      "epoch": 2.198880721683929,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004513418757462973,
      "loss": 4.8129,
      "mean_token_accuracy": 0.20547639280557634,
      "num_tokens": 48430330.0,
      "step": 26325
    },
    {
      "entropy": 4.942305088043213,
      "epoch": 2.1992983628466423,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045132338445836516,
      "loss": 4.6304,
      "mean_token_accuracy": 0.2238093361258507,
      "num_tokens": 48439729.0,
      "step": 26330
    },
    {
      "entropy": 5.032222557067871,
      "epoch": 2.199716004009355,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045130489008367997,
      "loss": 4.755,
      "mean_token_accuracy": 0.21691639721393585,
      "num_tokens": 48449011.0,
      "step": 26335
    },
    {
      "entropy": 5.095153665542602,
      "epoch": 2.2001336451720683,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004512863926225655,
      "loss": 4.7499,
      "mean_token_accuracy": 0.21550006717443465,
      "num_tokens": 48458164.0,
      "step": 26340
    },
    {
      "entropy": 4.975751161575317,
      "epoch": 2.200551286334781,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045126789207534544,
      "loss": 4.6473,
      "mean_token_accuracy": 0.22457277476787568,
      "num_tokens": 48467069.0,
      "step": 26345
    },
    {
      "entropy": 5.0904261589050295,
      "epoch": 2.200968927497494,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045124938844234386,
      "loss": 4.8506,
      "mean_token_accuracy": 0.21530150324106218,
      "num_tokens": 48475876.0,
      "step": 26350
    },
    {
      "entropy": 5.01316614151001,
      "epoch": 2.201386568660207,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045123088172388455,
      "loss": 4.6043,
      "mean_token_accuracy": 0.23807281106710435,
      "num_tokens": 48484282.0,
      "step": 26355
    },
    {
      "entropy": 5.00877833366394,
      "epoch": 2.20180420982292,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045121237192029154,
      "loss": 4.7148,
      "mean_token_accuracy": 0.22335435897111894,
      "num_tokens": 48494007.0,
      "step": 26360
    },
    {
      "entropy": 4.98975191116333,
      "epoch": 2.2022218509856333,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045119385903188877,
      "loss": 4.6416,
      "mean_token_accuracy": 0.2185087114572525,
      "num_tokens": 48503402.0,
      "step": 26365
    },
    {
      "entropy": 5.015630626678467,
      "epoch": 2.202639492148346,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004511753430590004,
      "loss": 4.6762,
      "mean_token_accuracy": 0.22858438044786453,
      "num_tokens": 48511547.0,
      "step": 26370
    },
    {
      "entropy": 4.887270975112915,
      "epoch": 2.2030571333110593,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004511568240019507,
      "loss": 4.5422,
      "mean_token_accuracy": 0.23761265277862548,
      "num_tokens": 48520064.0,
      "step": 26375
    },
    {
      "entropy": 5.01862530708313,
      "epoch": 2.203474774473772,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045113830186106367,
      "loss": 4.768,
      "mean_token_accuracy": 0.21765480786561966,
      "num_tokens": 48529506.0,
      "step": 26380
    },
    {
      "entropy": 5.02915472984314,
      "epoch": 2.203892415636485,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004511197766366637,
      "loss": 4.6878,
      "mean_token_accuracy": 0.23018942773342133,
      "num_tokens": 48538601.0,
      "step": 26385
    },
    {
      "entropy": 5.071386289596558,
      "epoch": 2.2043100567991982,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045110124832907505,
      "loss": 4.7972,
      "mean_token_accuracy": 0.2127843976020813,
      "num_tokens": 48547533.0,
      "step": 26390
    },
    {
      "entropy": 5.111337947845459,
      "epoch": 2.204727697961911,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004510827169386221,
      "loss": 4.822,
      "mean_token_accuracy": 0.220682193338871,
      "num_tokens": 48556694.0,
      "step": 26395
    },
    {
      "entropy": 5.012409114837647,
      "epoch": 2.2051453391246243,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004510641824656292,
      "loss": 4.7446,
      "mean_token_accuracy": 0.23151030391454697,
      "num_tokens": 48565200.0,
      "step": 26400
    },
    {
      "entropy": 5.086540937423706,
      "epoch": 2.205562980287337,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004510456449104208,
      "loss": 4.7653,
      "mean_token_accuracy": 0.21962059736251832,
      "num_tokens": 48574112.0,
      "step": 26405
    },
    {
      "entropy": 5.105652523040772,
      "epoch": 2.20598062145005,
      "grad_norm": 1.125,
      "learning_rate": 0.0004510271042733215,
      "loss": 4.7657,
      "mean_token_accuracy": 0.2240808129310608,
      "num_tokens": 48582935.0,
      "step": 26410
    },
    {
      "entropy": 5.08125057220459,
      "epoch": 2.206398262612763,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004510085605546559,
      "loss": 4.7913,
      "mean_token_accuracy": 0.21734238117933274,
      "num_tokens": 48593275.0,
      "step": 26415
    },
    {
      "entropy": 5.010436773300171,
      "epoch": 2.206815903775476,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045099001375474855,
      "loss": 4.6704,
      "mean_token_accuracy": 0.22910652309656143,
      "num_tokens": 48602458.0,
      "step": 26420
    },
    {
      "entropy": 5.04610390663147,
      "epoch": 2.2072335449381892,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045097146387392423,
      "loss": 4.7366,
      "mean_token_accuracy": 0.22067468166351317,
      "num_tokens": 48610982.0,
      "step": 26425
    },
    {
      "entropy": 5.1157301902771,
      "epoch": 2.207651186100902,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045095291091250755,
      "loss": 4.751,
      "mean_token_accuracy": 0.2180635392665863,
      "num_tokens": 48620268.0,
      "step": 26430
    },
    {
      "entropy": 5.0996113300323485,
      "epoch": 2.2080688272636153,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004509343548708235,
      "loss": 4.7933,
      "mean_token_accuracy": 0.2134250670671463,
      "num_tokens": 48629102.0,
      "step": 26435
    },
    {
      "entropy": 5.133145380020141,
      "epoch": 2.208486468426328,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004509157957491966,
      "loss": 4.8195,
      "mean_token_accuracy": 0.2162853315472603,
      "num_tokens": 48638337.0,
      "step": 26440
    },
    {
      "entropy": 5.0072849750518795,
      "epoch": 2.208904109589041,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045089723354795204,
      "loss": 4.6804,
      "mean_token_accuracy": 0.22671489417552948,
      "num_tokens": 48647087.0,
      "step": 26445
    },
    {
      "entropy": 5.060801887512207,
      "epoch": 2.209321750751754,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004508786682674147,
      "loss": 4.8254,
      "mean_token_accuracy": 0.20688123404979705,
      "num_tokens": 48657373.0,
      "step": 26450
    },
    {
      "entropy": 5.097093820571899,
      "epoch": 2.209739391914467,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045086009990790956,
      "loss": 4.8133,
      "mean_token_accuracy": 0.21211995631456376,
      "num_tokens": 48668309.0,
      "step": 26455
    },
    {
      "entropy": 5.0695130825042725,
      "epoch": 2.2101570330771803,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045084152846976166,
      "loss": 4.7476,
      "mean_token_accuracy": 0.22317629009485246,
      "num_tokens": 48677570.0,
      "step": 26460
    },
    {
      "entropy": 4.996308040618897,
      "epoch": 2.210574674239893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045082295395329616,
      "loss": 4.7429,
      "mean_token_accuracy": 0.2217785581946373,
      "num_tokens": 48686140.0,
      "step": 26465
    },
    {
      "entropy": 4.937915182113647,
      "epoch": 2.210992315402606,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004508043763588382,
      "loss": 4.6229,
      "mean_token_accuracy": 0.22806940078735352,
      "num_tokens": 48694468.0,
      "step": 26470
    },
    {
      "entropy": 5.154194116592407,
      "epoch": 2.211409956565319,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045078579568671307,
      "loss": 4.772,
      "mean_token_accuracy": 0.21484399884939193,
      "num_tokens": 48703242.0,
      "step": 26475
    },
    {
      "entropy": 4.990377616882324,
      "epoch": 2.211827597728032,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004507672119372459,
      "loss": 4.6783,
      "mean_token_accuracy": 0.22333232015371324,
      "num_tokens": 48711895.0,
      "step": 26480
    },
    {
      "entropy": 5.011617755889892,
      "epoch": 2.212245238890745,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045074862511076215,
      "loss": 4.6704,
      "mean_token_accuracy": 0.23220356851816176,
      "num_tokens": 48720489.0,
      "step": 26485
    },
    {
      "entropy": 5.031261348724366,
      "epoch": 2.212662880053458,
      "grad_norm": 1.125,
      "learning_rate": 0.0004507300352075872,
      "loss": 4.7333,
      "mean_token_accuracy": 0.22122908085584642,
      "num_tokens": 48730558.0,
      "step": 26490
    },
    {
      "entropy": 5.036866188049316,
      "epoch": 2.2130805212161713,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045071144222804646,
      "loss": 4.6169,
      "mean_token_accuracy": 0.2365153521299362,
      "num_tokens": 48739188.0,
      "step": 26495
    },
    {
      "entropy": 5.024508905410767,
      "epoch": 2.213498162378884,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004506928461724654,
      "loss": 4.7213,
      "mean_token_accuracy": 0.22319550961256027,
      "num_tokens": 48747563.0,
      "step": 26500
    },
    {
      "entropy": 5.03810305595398,
      "epoch": 2.213915803541597,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045067424704116956,
      "loss": 4.7844,
      "mean_token_accuracy": 0.2207907110452652,
      "num_tokens": 48756473.0,
      "step": 26505
    },
    {
      "entropy": 5.006614923477173,
      "epoch": 2.21433344470431,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004506556448344846,
      "loss": 4.7065,
      "mean_token_accuracy": 0.2231632351875305,
      "num_tokens": 48766053.0,
      "step": 26510
    },
    {
      "entropy": 5.034424734115601,
      "epoch": 2.214751085867023,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004506370395527362,
      "loss": 4.6894,
      "mean_token_accuracy": 0.22495282143354417,
      "num_tokens": 48774574.0,
      "step": 26515
    },
    {
      "entropy": 5.060122537612915,
      "epoch": 2.215168727029736,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045061843119624984,
      "loss": 4.7686,
      "mean_token_accuracy": 0.21313148736953735,
      "num_tokens": 48783309.0,
      "step": 26520
    },
    {
      "entropy": 5.067874383926392,
      "epoch": 2.215586368192449,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004505998197653516,
      "loss": 4.7774,
      "mean_token_accuracy": 0.2150296077132225,
      "num_tokens": 48791857.0,
      "step": 26525
    },
    {
      "entropy": 5.117160797119141,
      "epoch": 2.216004009355162,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004505812052603671,
      "loss": 4.7454,
      "mean_token_accuracy": 0.2180093541741371,
      "num_tokens": 48801853.0,
      "step": 26530
    },
    {
      "entropy": 5.167975950241089,
      "epoch": 2.216421650517875,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004505625876816222,
      "loss": 4.8441,
      "mean_token_accuracy": 0.21388857513666154,
      "num_tokens": 48811721.0,
      "step": 26535
    },
    {
      "entropy": 5.07542052268982,
      "epoch": 2.216839291680588,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045054396702944293,
      "loss": 4.7622,
      "mean_token_accuracy": 0.21698858886957167,
      "num_tokens": 48820626.0,
      "step": 26540
    },
    {
      "entropy": 5.043971347808838,
      "epoch": 2.217256932843301,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045052534330415525,
      "loss": 4.781,
      "mean_token_accuracy": 0.21942430585622788,
      "num_tokens": 48829504.0,
      "step": 26545
    },
    {
      "entropy": 5.048340749740601,
      "epoch": 2.217674574006014,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045050671650608513,
      "loss": 4.7483,
      "mean_token_accuracy": 0.21735415756702423,
      "num_tokens": 48838391.0,
      "step": 26550
    },
    {
      "entropy": 5.098257160186767,
      "epoch": 2.2180922151687272,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004504880866355587,
      "loss": 4.7399,
      "mean_token_accuracy": 0.2235512912273407,
      "num_tokens": 48848238.0,
      "step": 26555
    },
    {
      "entropy": 5.037132596969604,
      "epoch": 2.21850985633144,
      "grad_norm": 1.125,
      "learning_rate": 0.000450469453692902,
      "loss": 4.7056,
      "mean_token_accuracy": 0.2276079997420311,
      "num_tokens": 48857177.0,
      "step": 26560
    },
    {
      "entropy": 4.93217101097107,
      "epoch": 2.218927497494153,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045045081767844137,
      "loss": 4.609,
      "mean_token_accuracy": 0.236398383975029,
      "num_tokens": 48866952.0,
      "step": 26565
    },
    {
      "entropy": 5.0768674373626705,
      "epoch": 2.219345138656866,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000450432178592503,
      "loss": 4.8277,
      "mean_token_accuracy": 0.21737348586320876,
      "num_tokens": 48876488.0,
      "step": 26570
    },
    {
      "entropy": 5.168977308273315,
      "epoch": 2.219762779819579,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004504135364354131,
      "loss": 4.8135,
      "mean_token_accuracy": 0.21764546036720275,
      "num_tokens": 48885281.0,
      "step": 26575
    },
    {
      "entropy": 5.118706846237183,
      "epoch": 2.220180420982292,
      "grad_norm": 1.125,
      "learning_rate": 0.00045039489120749816,
      "loss": 4.7507,
      "mean_token_accuracy": 0.2287771448493004,
      "num_tokens": 48894333.0,
      "step": 26580
    },
    {
      "entropy": 4.979081106185913,
      "epoch": 2.220598062145005,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004503762429090845,
      "loss": 4.6633,
      "mean_token_accuracy": 0.2225970059633255,
      "num_tokens": 48903322.0,
      "step": 26585
    },
    {
      "entropy": 4.994715690612793,
      "epoch": 2.221015703307718,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004503575915404986,
      "loss": 4.6377,
      "mean_token_accuracy": 0.23416208177804948,
      "num_tokens": 48911788.0,
      "step": 26590
    },
    {
      "entropy": 5.0029703140258786,
      "epoch": 2.221433344470431,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000450338937102067,
      "loss": 4.726,
      "mean_token_accuracy": 0.23286814689636232,
      "num_tokens": 48920122.0,
      "step": 26595
    },
    {
      "entropy": 5.0903808116912845,
      "epoch": 2.221850985633144,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045032027959411624,
      "loss": 4.7408,
      "mean_token_accuracy": 0.22939995676279068,
      "num_tokens": 48928556.0,
      "step": 26600
    },
    {
      "entropy": 5.1344019889831545,
      "epoch": 2.222268626795857,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000450301619016973,
      "loss": 4.7994,
      "mean_token_accuracy": 0.21229207664728164,
      "num_tokens": 48938043.0,
      "step": 26605
    },
    {
      "entropy": 5.088619709014893,
      "epoch": 2.22268626795857,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045028295537096385,
      "loss": 4.7992,
      "mean_token_accuracy": 0.22077543586492537,
      "num_tokens": 48948032.0,
      "step": 26610
    },
    {
      "entropy": 5.006795501708984,
      "epoch": 2.223103909121283,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045026428865641563,
      "loss": 4.6716,
      "mean_token_accuracy": 0.2264975741505623,
      "num_tokens": 48956871.0,
      "step": 26615
    },
    {
      "entropy": 5.013470411300659,
      "epoch": 2.223521550283996,
      "grad_norm": 1.1875,
      "learning_rate": 0.000450245618873655,
      "loss": 4.6702,
      "mean_token_accuracy": 0.22588450312614441,
      "num_tokens": 48965071.0,
      "step": 26620
    },
    {
      "entropy": 5.096710968017578,
      "epoch": 2.223939191446709,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004502269460230089,
      "loss": 4.7773,
      "mean_token_accuracy": 0.21894364655017853,
      "num_tokens": 48973604.0,
      "step": 26625
    },
    {
      "entropy": 5.045039701461792,
      "epoch": 2.224356832609422,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004502082701048042,
      "loss": 4.6716,
      "mean_token_accuracy": 0.23182469457387925,
      "num_tokens": 48981820.0,
      "step": 26630
    },
    {
      "entropy": 5.063095140457153,
      "epoch": 2.224774473772135,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045018959111936785,
      "loss": 4.7752,
      "mean_token_accuracy": 0.22348157316446304,
      "num_tokens": 48991012.0,
      "step": 26635
    },
    {
      "entropy": 5.013464832305909,
      "epoch": 2.225192114934848,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004501709090670268,
      "loss": 4.7797,
      "mean_token_accuracy": 0.2336287170648575,
      "num_tokens": 48999561.0,
      "step": 26640
    },
    {
      "entropy": 4.986230802536011,
      "epoch": 2.225609756097561,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004501522239481081,
      "loss": 4.6708,
      "mean_token_accuracy": 0.228615503013134,
      "num_tokens": 49008906.0,
      "step": 26645
    },
    {
      "entropy": 5.077547979354859,
      "epoch": 2.2260273972602738,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045013353576293897,
      "loss": 4.7682,
      "mean_token_accuracy": 0.21561272889375688,
      "num_tokens": 49018344.0,
      "step": 26650
    },
    {
      "entropy": 5.095338296890259,
      "epoch": 2.226445038422987,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045011484451184646,
      "loss": 4.821,
      "mean_token_accuracy": 0.2198611468076706,
      "num_tokens": 49028183.0,
      "step": 26655
    },
    {
      "entropy": 4.911173582077026,
      "epoch": 2.2268626795857,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045009615019515783,
      "loss": 4.6003,
      "mean_token_accuracy": 0.23434397727251052,
      "num_tokens": 49037208.0,
      "step": 26660
    },
    {
      "entropy": 5.076073217391968,
      "epoch": 2.227280320748413,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045007745281320035,
      "loss": 4.7397,
      "mean_token_accuracy": 0.22487227022647857,
      "num_tokens": 49045807.0,
      "step": 26665
    },
    {
      "entropy": 5.012172698974609,
      "epoch": 2.227697961911126,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045005875236630124,
      "loss": 4.6298,
      "mean_token_accuracy": 0.22714647203683852,
      "num_tokens": 49055133.0,
      "step": 26670
    },
    {
      "entropy": 4.931532001495361,
      "epoch": 2.228115603073839,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045004004885478797,
      "loss": 4.615,
      "mean_token_accuracy": 0.22566746473312377,
      "num_tokens": 49063953.0,
      "step": 26675
    },
    {
      "entropy": 5.078385639190674,
      "epoch": 2.228533244236552,
      "grad_norm": 1.171875,
      "learning_rate": 0.000450021342278988,
      "loss": 4.8059,
      "mean_token_accuracy": 0.21182527244091034,
      "num_tokens": 49073087.0,
      "step": 26680
    },
    {
      "entropy": 5.041693830490113,
      "epoch": 2.228950885399265,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004500026326392287,
      "loss": 4.6442,
      "mean_token_accuracy": 0.23279110491275787,
      "num_tokens": 49081537.0,
      "step": 26685
    },
    {
      "entropy": 5.001098346710205,
      "epoch": 2.229368526561978,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044998391993583773,
      "loss": 4.7749,
      "mean_token_accuracy": 0.21448262333869933,
      "num_tokens": 49092081.0,
      "step": 26690
    },
    {
      "entropy": 4.9406272888183596,
      "epoch": 2.229786167724691,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004499652041691425,
      "loss": 4.6363,
      "mean_token_accuracy": 0.231331504881382,
      "num_tokens": 49100669.0,
      "step": 26695
    },
    {
      "entropy": 5.070400953292847,
      "epoch": 2.230203808887404,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004499464853394709,
      "loss": 4.6609,
      "mean_token_accuracy": 0.21999761760234832,
      "num_tokens": 49109536.0,
      "step": 26700
    },
    {
      "entropy": 5.04000506401062,
      "epoch": 2.230621450050117,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044992776344715036,
      "loss": 4.7134,
      "mean_token_accuracy": 0.2246331036090851,
      "num_tokens": 49119267.0,
      "step": 26705
    },
    {
      "entropy": 5.003693246841431,
      "epoch": 2.23103909121283,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004499090384925088,
      "loss": 4.7196,
      "mean_token_accuracy": 0.224245822429657,
      "num_tokens": 49127570.0,
      "step": 26710
    },
    {
      "entropy": 5.0083612442016605,
      "epoch": 2.231456732375543,
      "grad_norm": 1.125,
      "learning_rate": 0.00044989031047587406,
      "loss": 4.703,
      "mean_token_accuracy": 0.2203958049416542,
      "num_tokens": 49136919.0,
      "step": 26715
    },
    {
      "entropy": 5.028084325790405,
      "epoch": 2.231874373538256,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004498715793975738,
      "loss": 4.7551,
      "mean_token_accuracy": 0.2232462629675865,
      "num_tokens": 49146779.0,
      "step": 26720
    },
    {
      "entropy": 4.97459716796875,
      "epoch": 2.232292014700969,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044985284525793617,
      "loss": 4.6722,
      "mean_token_accuracy": 0.23181024938821793,
      "num_tokens": 49154618.0,
      "step": 26725
    },
    {
      "entropy": 5.027267599105835,
      "epoch": 2.232709655863682,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004498341080572889,
      "loss": 4.7168,
      "mean_token_accuracy": 0.22098055630922317,
      "num_tokens": 49163837.0,
      "step": 26730
    },
    {
      "entropy": 5.115616703033448,
      "epoch": 2.233127297026395,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044981536779596025,
      "loss": 4.8225,
      "mean_token_accuracy": 0.22193877547979354,
      "num_tokens": 49172891.0,
      "step": 26735
    },
    {
      "entropy": 5.040465688705444,
      "epoch": 2.233544938189108,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004497966244742781,
      "loss": 4.6577,
      "mean_token_accuracy": 0.23070076555013658,
      "num_tokens": 49181894.0,
      "step": 26740
    },
    {
      "entropy": 5.082738304138184,
      "epoch": 2.2339625793518207,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004497778780925706,
      "loss": 4.7933,
      "mean_token_accuracy": 0.2307378277182579,
      "num_tokens": 49191231.0,
      "step": 26745
    },
    {
      "entropy": 5.061469554901123,
      "epoch": 2.234380220514534,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044975912865116604,
      "loss": 4.7667,
      "mean_token_accuracy": 0.2214697018265724,
      "num_tokens": 49199761.0,
      "step": 26750
    },
    {
      "entropy": 5.142849636077881,
      "epoch": 2.234797861677247,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044974037615039253,
      "loss": 4.8538,
      "mean_token_accuracy": 0.2096710667014122,
      "num_tokens": 49209112.0,
      "step": 26755
    },
    {
      "entropy": 5.0362261772155765,
      "epoch": 2.23521550283996,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044972162059057837,
      "loss": 4.7282,
      "mean_token_accuracy": 0.22243111580610275,
      "num_tokens": 49218417.0,
      "step": 26760
    },
    {
      "entropy": 5.1099687099456785,
      "epoch": 2.235633144002673,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044970286197205196,
      "loss": 4.7754,
      "mean_token_accuracy": 0.21840943694114684,
      "num_tokens": 49227745.0,
      "step": 26765
    },
    {
      "entropy": 5.055571365356445,
      "epoch": 2.2360507851653857,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004496841002951416,
      "loss": 4.7548,
      "mean_token_accuracy": 0.220726178586483,
      "num_tokens": 49237606.0,
      "step": 26770
    },
    {
      "entropy": 5.001522588729858,
      "epoch": 2.236468426328099,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004496653355601759,
      "loss": 4.7592,
      "mean_token_accuracy": 0.21434602290391921,
      "num_tokens": 49246654.0,
      "step": 26775
    },
    {
      "entropy": 5.107018709182739,
      "epoch": 2.2368860674908118,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004496465677674832,
      "loss": 4.818,
      "mean_token_accuracy": 0.2137664183974266,
      "num_tokens": 49255613.0,
      "step": 26780
    },
    {
      "entropy": 5.07185320854187,
      "epoch": 2.237303708653525,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004496277969173921,
      "loss": 4.6707,
      "mean_token_accuracy": 0.22600764781236649,
      "num_tokens": 49264400.0,
      "step": 26785
    },
    {
      "entropy": 5.140026807785034,
      "epoch": 2.237721349816238,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044960902301023126,
      "loss": 4.8159,
      "mean_token_accuracy": 0.22452933490276336,
      "num_tokens": 49273879.0,
      "step": 26790
    },
    {
      "entropy": 4.964891338348389,
      "epoch": 2.238138990978951,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004495902460463293,
      "loss": 4.679,
      "mean_token_accuracy": 0.23592304438352585,
      "num_tokens": 49282739.0,
      "step": 26795
    },
    {
      "entropy": 4.958805799484253,
      "epoch": 2.238556632141664,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004495714660260149,
      "loss": 4.6042,
      "mean_token_accuracy": 0.23336805105209352,
      "num_tokens": 49291060.0,
      "step": 26800
    },
    {
      "entropy": 4.9926167011260985,
      "epoch": 2.2389742733043767,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044955268294961687,
      "loss": 4.6744,
      "mean_token_accuracy": 0.23591246902942659,
      "num_tokens": 49299684.0,
      "step": 26805
    },
    {
      "entropy": 4.8779487133026125,
      "epoch": 2.23939191446709,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044953389681746405,
      "loss": 4.6173,
      "mean_token_accuracy": 0.22922206819057464,
      "num_tokens": 49309108.0,
      "step": 26810
    },
    {
      "entropy": 5.070571851730347,
      "epoch": 2.2398095556298028,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004495151076298852,
      "loss": 4.7272,
      "mean_token_accuracy": 0.22255444377660752,
      "num_tokens": 49318240.0,
      "step": 26815
    },
    {
      "entropy": 5.072149467468262,
      "epoch": 2.240227196792516,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004494963153872093,
      "loss": 4.7583,
      "mean_token_accuracy": 0.21934670209884644,
      "num_tokens": 49327157.0,
      "step": 26820
    },
    {
      "entropy": 4.98439211845398,
      "epoch": 2.240644837955229,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044947752008976546,
      "loss": 4.7426,
      "mean_token_accuracy": 0.21892323344945908,
      "num_tokens": 49336168.0,
      "step": 26825
    },
    {
      "entropy": 5.038888454437256,
      "epoch": 2.241062479117942,
      "grad_norm": 1.125,
      "learning_rate": 0.0004494587217378826,
      "loss": 4.7365,
      "mean_token_accuracy": 0.22536407113075257,
      "num_tokens": 49345921.0,
      "step": 26830
    },
    {
      "entropy": 4.955388784408569,
      "epoch": 2.241480120280655,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044943992033188974,
      "loss": 4.5363,
      "mean_token_accuracy": 0.2417470097541809,
      "num_tokens": 49354166.0,
      "step": 26835
    },
    {
      "entropy": 5.061908006668091,
      "epoch": 2.2418977614433677,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004494211158721162,
      "loss": 4.7919,
      "mean_token_accuracy": 0.22336441576480864,
      "num_tokens": 49363377.0,
      "step": 26840
    },
    {
      "entropy": 5.096724224090576,
      "epoch": 2.242315402606081,
      "grad_norm": 1.078125,
      "learning_rate": 0.000449402308358891,
      "loss": 4.847,
      "mean_token_accuracy": 0.21382818073034288,
      "num_tokens": 49372717.0,
      "step": 26845
    },
    {
      "entropy": 5.163710117340088,
      "epoch": 2.242733043768794,
      "grad_norm": 1.125,
      "learning_rate": 0.0004493834977925435,
      "loss": 4.7847,
      "mean_token_accuracy": 0.21373616009950638,
      "num_tokens": 49382789.0,
      "step": 26850
    },
    {
      "entropy": 5.043903970718384,
      "epoch": 2.243150684931507,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044936468417340294,
      "loss": 4.7359,
      "mean_token_accuracy": 0.2241788387298584,
      "num_tokens": 49393059.0,
      "step": 26855
    },
    {
      "entropy": 5.068076181411743,
      "epoch": 2.24356832609422,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004493458675017987,
      "loss": 4.7538,
      "mean_token_accuracy": 0.2222486823797226,
      "num_tokens": 49401669.0,
      "step": 26860
    },
    {
      "entropy": 5.223622274398804,
      "epoch": 2.2439859672569327,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004493270477780602,
      "loss": 4.9679,
      "mean_token_accuracy": 0.21267763078212737,
      "num_tokens": 49411896.0,
      "step": 26865
    },
    {
      "entropy": 5.0115594387054445,
      "epoch": 2.244403608419646,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044930822500251693,
      "loss": 4.7135,
      "mean_token_accuracy": 0.22024763524532318,
      "num_tokens": 49420055.0,
      "step": 26870
    },
    {
      "entropy": 5.016980886459351,
      "epoch": 2.2448212495823587,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044928939917549833,
      "loss": 4.725,
      "mean_token_accuracy": 0.22394467294216155,
      "num_tokens": 49428985.0,
      "step": 26875
    },
    {
      "entropy": 5.073509740829468,
      "epoch": 2.245238890745072,
      "grad_norm": 1.140625,
      "learning_rate": 0.000449270570297334,
      "loss": 4.798,
      "mean_token_accuracy": 0.22702440470457078,
      "num_tokens": 49438383.0,
      "step": 26880
    },
    {
      "entropy": 5.171377086639405,
      "epoch": 2.245656531907785,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004492517383683536,
      "loss": 4.8655,
      "mean_token_accuracy": 0.21096551269292832,
      "num_tokens": 49447920.0,
      "step": 26885
    },
    {
      "entropy": 5.09482741355896,
      "epoch": 2.2460741730704976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044923290338888674,
      "loss": 4.7621,
      "mean_token_accuracy": 0.2193629577755928,
      "num_tokens": 49457073.0,
      "step": 26890
    },
    {
      "entropy": 5.000504684448242,
      "epoch": 2.246491814233211,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004492140653592632,
      "loss": 4.6734,
      "mean_token_accuracy": 0.2285323515534401,
      "num_tokens": 49465995.0,
      "step": 26895
    },
    {
      "entropy": 5.022324419021606,
      "epoch": 2.2469094553959237,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044919522427981276,
      "loss": 4.719,
      "mean_token_accuracy": 0.2241417035460472,
      "num_tokens": 49474823.0,
      "step": 26900
    },
    {
      "entropy": 5.003506326675415,
      "epoch": 2.247327096558637,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004491763801508652,
      "loss": 4.6671,
      "mean_token_accuracy": 0.22931479811668395,
      "num_tokens": 49483462.0,
      "step": 26905
    },
    {
      "entropy": 5.069011497497558,
      "epoch": 2.2477447377213498,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044915753297275046,
      "loss": 4.7602,
      "mean_token_accuracy": 0.2197437345981598,
      "num_tokens": 49492652.0,
      "step": 26910
    },
    {
      "entropy": 5.0563061237335205,
      "epoch": 2.248162378884063,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044913868274579856,
      "loss": 4.7666,
      "mean_token_accuracy": 0.21787910908460617,
      "num_tokens": 49501193.0,
      "step": 26915
    },
    {
      "entropy": 5.089885520935058,
      "epoch": 2.248580020046776,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044911982947033925,
      "loss": 4.8006,
      "mean_token_accuracy": 0.22441229671239854,
      "num_tokens": 49511543.0,
      "step": 26920
    },
    {
      "entropy": 5.004009437561035,
      "epoch": 2.2489976612094886,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004491009731467028,
      "loss": 4.6336,
      "mean_token_accuracy": 0.23296545594930648,
      "num_tokens": 49520743.0,
      "step": 26925
    },
    {
      "entropy": 5.040612602233887,
      "epoch": 2.249415302372202,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004490821137752193,
      "loss": 4.7006,
      "mean_token_accuracy": 0.22305691689252855,
      "num_tokens": 49529196.0,
      "step": 26930
    },
    {
      "entropy": 4.973658275604248,
      "epoch": 2.2498329435349147,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004490632513562188,
      "loss": 4.6058,
      "mean_token_accuracy": 0.23771993070840836,
      "num_tokens": 49538019.0,
      "step": 26935
    },
    {
      "entropy": 4.975428915023803,
      "epoch": 2.250250584697628,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004490443858900316,
      "loss": 4.6707,
      "mean_token_accuracy": 0.23150922060012818,
      "num_tokens": 49546646.0,
      "step": 26940
    },
    {
      "entropy": 5.0850317001342775,
      "epoch": 2.2506682258603408,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044902551737698785,
      "loss": 4.8449,
      "mean_token_accuracy": 0.21124834269285203,
      "num_tokens": 49556042.0,
      "step": 26945
    },
    {
      "entropy": 5.034253311157227,
      "epoch": 2.251085867023054,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044900664581741805,
      "loss": 4.7625,
      "mean_token_accuracy": 0.21938493251800537,
      "num_tokens": 49565043.0,
      "step": 26950
    },
    {
      "entropy": 5.012296724319458,
      "epoch": 2.251503508185767,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004489877712116524,
      "loss": 4.6871,
      "mean_token_accuracy": 0.2234481006860733,
      "num_tokens": 49575195.0,
      "step": 26955
    },
    {
      "entropy": 5.032193422317505,
      "epoch": 2.2519211493484796,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004489688935600214,
      "loss": 4.7516,
      "mean_token_accuracy": 0.21957076638936995,
      "num_tokens": 49584007.0,
      "step": 26960
    },
    {
      "entropy": 5.053209686279297,
      "epoch": 2.252338790511193,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044895001286285557,
      "loss": 4.678,
      "mean_token_accuracy": 0.23265742659568786,
      "num_tokens": 49592859.0,
      "step": 26965
    },
    {
      "entropy": 4.9935187816619875,
      "epoch": 2.2527564316739057,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044893112912048526,
      "loss": 4.6603,
      "mean_token_accuracy": 0.23290526121854782,
      "num_tokens": 49602846.0,
      "step": 26970
    },
    {
      "entropy": 5.01421046257019,
      "epoch": 2.253174072836619,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004489122423332413,
      "loss": 4.7123,
      "mean_token_accuracy": 0.22299446761608124,
      "num_tokens": 49612195.0,
      "step": 26975
    },
    {
      "entropy": 5.073737955093383,
      "epoch": 2.253591713999332,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044889335250145415,
      "loss": 4.753,
      "mean_token_accuracy": 0.2182840034365654,
      "num_tokens": 49621831.0,
      "step": 26980
    },
    {
      "entropy": 5.0658575057983395,
      "epoch": 2.2540093551620446,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004488744596254545,
      "loss": 4.746,
      "mean_token_accuracy": 0.22533396780490875,
      "num_tokens": 49631610.0,
      "step": 26985
    },
    {
      "entropy": 5.036496305465699,
      "epoch": 2.254426996324758,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044885556370557317,
      "loss": 4.7249,
      "mean_token_accuracy": 0.21892993301153182,
      "num_tokens": 49641215.0,
      "step": 26990
    },
    {
      "entropy": 5.05700798034668,
      "epoch": 2.2548446374874707,
      "grad_norm": 1.109375,
      "learning_rate": 0.000448836664742141,
      "loss": 4.7009,
      "mean_token_accuracy": 0.22471926212310792,
      "num_tokens": 49649820.0,
      "step": 26995
    },
    {
      "entropy": 4.943443393707275,
      "epoch": 2.255262278650184,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004488177627354887,
      "loss": 4.644,
      "mean_token_accuracy": 0.2320872187614441,
      "num_tokens": 49659269.0,
      "step": 27000
    },
    {
      "epoch": 2.255262278650184,
      "eval_entropy": 4.751085709172067,
      "eval_loss": 4.788825035095215,
      "eval_mean_token_accuracy": 0.2310622974258059,
      "eval_num_tokens": 49659269.0,
      "eval_runtime": 27.6565,
      "eval_samples_per_second": 1365.429,
      "eval_steps_per_second": 170.701,
      "step": 27000
    },
    {
      "entropy": 4.935283613204956,
      "epoch": 2.2556799198128967,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044879885768594726,
      "loss": 4.6647,
      "mean_token_accuracy": 0.22868740409612656,
      "num_tokens": 49668156.0,
      "step": 27005
    },
    {
      "entropy": 5.0906907081604,
      "epoch": 2.2560975609756095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044877994959384757,
      "loss": 4.7354,
      "mean_token_accuracy": 0.22248946726322175,
      "num_tokens": 49676860.0,
      "step": 27010
    },
    {
      "entropy": 5.104764127731324,
      "epoch": 2.256515202138323,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004487610384595207,
      "loss": 4.8193,
      "mean_token_accuracy": 0.21666548401117325,
      "num_tokens": 49687121.0,
      "step": 27015
    },
    {
      "entropy": 5.079688215255738,
      "epoch": 2.2569328433010356,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004487421242832977,
      "loss": 4.6776,
      "mean_token_accuracy": 0.22347214967012405,
      "num_tokens": 49696150.0,
      "step": 27020
    },
    {
      "entropy": 5.088489580154419,
      "epoch": 2.257350484463749,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004487232070655097,
      "loss": 4.8158,
      "mean_token_accuracy": 0.21435222178697586,
      "num_tokens": 49705082.0,
      "step": 27025
    },
    {
      "entropy": 4.986320877075196,
      "epoch": 2.2577681256264617,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004487042868064878,
      "loss": 4.6447,
      "mean_token_accuracy": 0.23769346624612808,
      "num_tokens": 49714220.0,
      "step": 27030
    },
    {
      "entropy": 5.107826614379883,
      "epoch": 2.258185766789175,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004486853635065634,
      "loss": 4.6859,
      "mean_token_accuracy": 0.22413790673017503,
      "num_tokens": 49724210.0,
      "step": 27035
    },
    {
      "entropy": 5.0409503936767575,
      "epoch": 2.2586034079518877,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044866643716606747,
      "loss": 4.7871,
      "mean_token_accuracy": 0.2180727705359459,
      "num_tokens": 49734305.0,
      "step": 27040
    },
    {
      "entropy": 5.074439764022827,
      "epoch": 2.2590210491146006,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004486475077853316,
      "loss": 4.6703,
      "mean_token_accuracy": 0.22597847729921341,
      "num_tokens": 49743808.0,
      "step": 27045
    },
    {
      "entropy": 4.9884261131286625,
      "epoch": 2.259438690277314,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044862857536468717,
      "loss": 4.6458,
      "mean_token_accuracy": 0.22422344982624054,
      "num_tokens": 49752603.0,
      "step": 27050
    },
    {
      "entropy": 5.0304395198822025,
      "epoch": 2.2598563314400266,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004486096399044654,
      "loss": 4.741,
      "mean_token_accuracy": 0.21290853172540664,
      "num_tokens": 49761753.0,
      "step": 27055
    },
    {
      "entropy": 5.028843355178833,
      "epoch": 2.26027397260274,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000448590701404998,
      "loss": 4.7735,
      "mean_token_accuracy": 0.22075500190258027,
      "num_tokens": 49770324.0,
      "step": 27060
    },
    {
      "entropy": 4.998972225189209,
      "epoch": 2.2606916137654527,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044857175986661634,
      "loss": 4.758,
      "mean_token_accuracy": 0.22141824662685394,
      "num_tokens": 49781004.0,
      "step": 27065
    },
    {
      "entropy": 5.114753484725952,
      "epoch": 2.261109254928166,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004485528152896522,
      "loss": 4.7959,
      "mean_token_accuracy": 0.21507514715194703,
      "num_tokens": 49790013.0,
      "step": 27070
    },
    {
      "entropy": 5.094102191925049,
      "epoch": 2.2615268960908788,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044853386767443706,
      "loss": 4.7012,
      "mean_token_accuracy": 0.2247815176844597,
      "num_tokens": 49798765.0,
      "step": 27075
    },
    {
      "entropy": 5.072072076797485,
      "epoch": 2.2619445372535916,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044851491702130265,
      "loss": 4.9006,
      "mean_token_accuracy": 0.21790654957294464,
      "num_tokens": 49808889.0,
      "step": 27080
    },
    {
      "entropy": 5.044763374328613,
      "epoch": 2.262362178416305,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044849596333058087,
      "loss": 4.6954,
      "mean_token_accuracy": 0.22588675916194917,
      "num_tokens": 49818196.0,
      "step": 27085
    },
    {
      "entropy": 4.922862339019775,
      "epoch": 2.2627798195790176,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044847700660260325,
      "loss": 4.6194,
      "mean_token_accuracy": 0.23533264845609664,
      "num_tokens": 49827272.0,
      "step": 27090
    },
    {
      "entropy": 5.011947250366211,
      "epoch": 2.263197460741731,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004484580468377019,
      "loss": 4.7139,
      "mean_token_accuracy": 0.22166894376277924,
      "num_tokens": 49836651.0,
      "step": 27095
    },
    {
      "entropy": 5.053423929214477,
      "epoch": 2.2636151019044437,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004484390840362086,
      "loss": 4.7348,
      "mean_token_accuracy": 0.2284197598695755,
      "num_tokens": 49845677.0,
      "step": 27100
    },
    {
      "entropy": 5.064637041091919,
      "epoch": 2.2640327430671565,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044842011819845546,
      "loss": 4.6892,
      "mean_token_accuracy": 0.22579249739646912,
      "num_tokens": 49855102.0,
      "step": 27105
    },
    {
      "entropy": 5.03024206161499,
      "epoch": 2.2644503842298698,
      "grad_norm": 1.125,
      "learning_rate": 0.00044840114932477435,
      "loss": 4.7162,
      "mean_token_accuracy": 0.22435843348503112,
      "num_tokens": 49864153.0,
      "step": 27110
    },
    {
      "entropy": 4.982381105422974,
      "epoch": 2.2648680253925826,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004483821774154975,
      "loss": 4.7939,
      "mean_token_accuracy": 0.22314644306898118,
      "num_tokens": 49873382.0,
      "step": 27115
    },
    {
      "entropy": 5.172749328613281,
      "epoch": 2.265285666555296,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044836320247095674,
      "loss": 4.7671,
      "mean_token_accuracy": 0.21799876689910888,
      "num_tokens": 49883283.0,
      "step": 27120
    },
    {
      "entropy": 5.048569250106811,
      "epoch": 2.2657033077180087,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004483442244914846,
      "loss": 4.7153,
      "mean_token_accuracy": 0.22639400511980057,
      "num_tokens": 49891994.0,
      "step": 27125
    },
    {
      "entropy": 5.033934545516968,
      "epoch": 2.2661209488807215,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004483252434774131,
      "loss": 4.6881,
      "mean_token_accuracy": 0.22265065908432008,
      "num_tokens": 49900465.0,
      "step": 27130
    },
    {
      "entropy": 5.01867094039917,
      "epoch": 2.2665385900434347,
      "grad_norm": 1.125,
      "learning_rate": 0.0004483062594290746,
      "loss": 4.6672,
      "mean_token_accuracy": 0.22460479885339737,
      "num_tokens": 49909108.0,
      "step": 27135
    },
    {
      "entropy": 4.981398582458496,
      "epoch": 2.2669562312061475,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004482872723468014,
      "loss": 4.652,
      "mean_token_accuracy": 0.2269057422876358,
      "num_tokens": 49918681.0,
      "step": 27140
    },
    {
      "entropy": 4.988452529907226,
      "epoch": 2.267373872368861,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044826828223092584,
      "loss": 4.7033,
      "mean_token_accuracy": 0.21829135566949845,
      "num_tokens": 49927080.0,
      "step": 27145
    },
    {
      "entropy": 5.065100812911988,
      "epoch": 2.2677915135315736,
      "grad_norm": 1.125,
      "learning_rate": 0.0004482492890817806,
      "loss": 4.752,
      "mean_token_accuracy": 0.22468229830265046,
      "num_tokens": 49935722.0,
      "step": 27150
    },
    {
      "entropy": 5.152456378936767,
      "epoch": 2.268209154694287,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044823029289969786,
      "loss": 4.8102,
      "mean_token_accuracy": 0.21785763800144195,
      "num_tokens": 49945063.0,
      "step": 27155
    },
    {
      "entropy": 5.063494539260864,
      "epoch": 2.2686267958569997,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004482112936850104,
      "loss": 4.7984,
      "mean_token_accuracy": 0.2154679387807846,
      "num_tokens": 49954953.0,
      "step": 27160
    },
    {
      "entropy": 5.00396842956543,
      "epoch": 2.2690444370197125,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044819229143805076,
      "loss": 4.6359,
      "mean_token_accuracy": 0.22669153660535812,
      "num_tokens": 49964293.0,
      "step": 27165
    },
    {
      "entropy": 4.9538891315460205,
      "epoch": 2.2694620781824257,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004481732861591516,
      "loss": 4.6846,
      "mean_token_accuracy": 0.22936004996299744,
      "num_tokens": 49973199.0,
      "step": 27170
    },
    {
      "entropy": 5.134088563919067,
      "epoch": 2.2698797193451385,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004481542778486456,
      "loss": 4.7302,
      "mean_token_accuracy": 0.22319511622190474,
      "num_tokens": 49981676.0,
      "step": 27175
    },
    {
      "entropy": 5.057632064819336,
      "epoch": 2.270297360507852,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004481352665068654,
      "loss": 4.7592,
      "mean_token_accuracy": 0.22076866626739503,
      "num_tokens": 49990930.0,
      "step": 27180
    },
    {
      "entropy": 5.002482891082764,
      "epoch": 2.2707150016705646,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044811625213414416,
      "loss": 4.6783,
      "mean_token_accuracy": 0.23057236075401305,
      "num_tokens": 50000201.0,
      "step": 27185
    },
    {
      "entropy": 4.974889612197876,
      "epoch": 2.271132642833278,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004480972347308145,
      "loss": 4.6266,
      "mean_token_accuracy": 0.23941368758678436,
      "num_tokens": 50008746.0,
      "step": 27190
    },
    {
      "entropy": 4.872776985168457,
      "epoch": 2.2715502839959907,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004480782142972093,
      "loss": 4.5147,
      "mean_token_accuracy": 0.2364830881357193,
      "num_tokens": 50016902.0,
      "step": 27195
    },
    {
      "entropy": 5.027250194549561,
      "epoch": 2.2719679251587035,
      "grad_norm": 1.125,
      "learning_rate": 0.00044805919083366175,
      "loss": 4.7837,
      "mean_token_accuracy": 0.22222827970981598,
      "num_tokens": 50026152.0,
      "step": 27200
    },
    {
      "entropy": 5.107401943206787,
      "epoch": 2.2723855663214167,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044804016434050466,
      "loss": 4.702,
      "mean_token_accuracy": 0.21603283882141114,
      "num_tokens": 50035141.0,
      "step": 27205
    },
    {
      "entropy": 5.047293472290039,
      "epoch": 2.2728032074841296,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004480211348180712,
      "loss": 4.7349,
      "mean_token_accuracy": 0.2153696149587631,
      "num_tokens": 50044883.0,
      "step": 27210
    },
    {
      "entropy": 5.0561247825622555,
      "epoch": 2.273220848646843,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044800210226669454,
      "loss": 4.7159,
      "mean_token_accuracy": 0.21957826018333435,
      "num_tokens": 50053695.0,
      "step": 27215
    },
    {
      "entropy": 5.051060914993286,
      "epoch": 2.2736384898095556,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004479830666867078,
      "loss": 4.8112,
      "mean_token_accuracy": 0.2182065635919571,
      "num_tokens": 50063151.0,
      "step": 27220
    },
    {
      "entropy": 4.979455041885376,
      "epoch": 2.2740561309722684,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004479640280784443,
      "loss": 4.7318,
      "mean_token_accuracy": 0.22575590908527374,
      "num_tokens": 50071207.0,
      "step": 27225
    },
    {
      "entropy": 4.928159236907959,
      "epoch": 2.2744737721349817,
      "grad_norm": 1.125,
      "learning_rate": 0.0004479449864422373,
      "loss": 4.5593,
      "mean_token_accuracy": 0.2360265851020813,
      "num_tokens": 50080122.0,
      "step": 27230
    },
    {
      "entropy": 5.045963191986084,
      "epoch": 2.2748914132976945,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004479259417784201,
      "loss": 4.7197,
      "mean_token_accuracy": 0.22965389490127563,
      "num_tokens": 50090114.0,
      "step": 27235
    },
    {
      "entropy": 4.990244770050049,
      "epoch": 2.2753090544604078,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044790689408732623,
      "loss": 4.6697,
      "mean_token_accuracy": 0.2315833792090416,
      "num_tokens": 50099328.0,
      "step": 27240
    },
    {
      "entropy": 5.046450233459472,
      "epoch": 2.2757266956231206,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044788784336928896,
      "loss": 4.8244,
      "mean_token_accuracy": 0.22067587226629257,
      "num_tokens": 50108575.0,
      "step": 27245
    },
    {
      "entropy": 5.158518600463867,
      "epoch": 2.2761443367858334,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004478687896246419,
      "loss": 4.8385,
      "mean_token_accuracy": 0.20874414294958116,
      "num_tokens": 50118786.0,
      "step": 27250
    },
    {
      "entropy": 5.069463777542114,
      "epoch": 2.2765619779485466,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044784973285371863,
      "loss": 4.7236,
      "mean_token_accuracy": 0.22275009602308274,
      "num_tokens": 50127935.0,
      "step": 27255
    },
    {
      "entropy": 5.063579797744751,
      "epoch": 2.2769796191112595,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044783067305685274,
      "loss": 4.8358,
      "mean_token_accuracy": 0.2126213625073433,
      "num_tokens": 50136940.0,
      "step": 27260
    },
    {
      "entropy": 4.994999074935913,
      "epoch": 2.2773972602739727,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004478116102343778,
      "loss": 4.5768,
      "mean_token_accuracy": 0.23551921546459198,
      "num_tokens": 50144781.0,
      "step": 27265
    },
    {
      "entropy": 4.95854811668396,
      "epoch": 2.2778149014366855,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004477925443866277,
      "loss": 4.7014,
      "mean_token_accuracy": 0.22137752026319504,
      "num_tokens": 50153545.0,
      "step": 27270
    },
    {
      "entropy": 5.045521831512451,
      "epoch": 2.278232542599399,
      "grad_norm": 1.109375,
      "learning_rate": 0.000447773475513936,
      "loss": 4.7792,
      "mean_token_accuracy": 0.21502433866262435,
      "num_tokens": 50163515.0,
      "step": 27275
    },
    {
      "entropy": 5.144568824768067,
      "epoch": 2.2786501837621116,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044775440361663677,
      "loss": 4.7091,
      "mean_token_accuracy": 0.22421592324972153,
      "num_tokens": 50171449.0,
      "step": 27280
    },
    {
      "entropy": 5.000155878067017,
      "epoch": 2.2790678249248244,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044773532869506373,
      "loss": 4.737,
      "mean_token_accuracy": 0.22168929725885392,
      "num_tokens": 50180198.0,
      "step": 27285
    },
    {
      "entropy": 5.004493141174317,
      "epoch": 2.2794854660875377,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004477162507495508,
      "loss": 4.7405,
      "mean_token_accuracy": 0.22607209980487825,
      "num_tokens": 50189838.0,
      "step": 27290
    },
    {
      "entropy": 5.0492922306060795,
      "epoch": 2.2799031072502505,
      "grad_norm": 1.140625,
      "learning_rate": 0.000447697169780432,
      "loss": 4.7375,
      "mean_token_accuracy": 0.22551899701356887,
      "num_tokens": 50200471.0,
      "step": 27295
    },
    {
      "entropy": 5.100579309463501,
      "epoch": 2.2803207484129637,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044767808578804135,
      "loss": 4.7717,
      "mean_token_accuracy": 0.22396559417247772,
      "num_tokens": 50211070.0,
      "step": 27300
    },
    {
      "entropy": 4.967407369613648,
      "epoch": 2.2807383895756765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044765899877271306,
      "loss": 4.7009,
      "mean_token_accuracy": 0.22693252563476562,
      "num_tokens": 50219884.0,
      "step": 27305
    },
    {
      "entropy": 4.9765989780426025,
      "epoch": 2.28115603073839,
      "grad_norm": 1.125,
      "learning_rate": 0.00044763990873478106,
      "loss": 4.6463,
      "mean_token_accuracy": 0.23050931394100188,
      "num_tokens": 50228598.0,
      "step": 27310
    },
    {
      "entropy": 4.867336797714233,
      "epoch": 2.2815736719011026,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044762081567457966,
      "loss": 4.5857,
      "mean_token_accuracy": 0.2305217996239662,
      "num_tokens": 50237149.0,
      "step": 27315
    },
    {
      "entropy": 5.035120296478271,
      "epoch": 2.2819913130638154,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004476017195924432,
      "loss": 4.7179,
      "mean_token_accuracy": 0.22006236463785173,
      "num_tokens": 50247009.0,
      "step": 27320
    },
    {
      "entropy": 4.9900754451751705,
      "epoch": 2.2824089542265287,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044758262048870574,
      "loss": 4.6222,
      "mean_token_accuracy": 0.2337762475013733,
      "num_tokens": 50256813.0,
      "step": 27325
    },
    {
      "entropy": 5.059978580474853,
      "epoch": 2.2828265953892415,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044756351836370184,
      "loss": 4.7107,
      "mean_token_accuracy": 0.22737374007701874,
      "num_tokens": 50265947.0,
      "step": 27330
    },
    {
      "entropy": 5.071958541870117,
      "epoch": 2.2832442365519547,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004475444132177659,
      "loss": 4.78,
      "mean_token_accuracy": 0.2187327414751053,
      "num_tokens": 50275620.0,
      "step": 27335
    },
    {
      "entropy": 5.03886866569519,
      "epoch": 2.2836618777146676,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004475253050512322,
      "loss": 4.7047,
      "mean_token_accuracy": 0.22702934145927428,
      "num_tokens": 50285229.0,
      "step": 27340
    },
    {
      "entropy": 5.017871952056884,
      "epoch": 2.2840795188773804,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044750619386443543,
      "loss": 4.6399,
      "mean_token_accuracy": 0.23293629586696624,
      "num_tokens": 50294534.0,
      "step": 27345
    },
    {
      "entropy": 5.001266622543335,
      "epoch": 2.2844971600400936,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044748707965771005,
      "loss": 4.742,
      "mean_token_accuracy": 0.22409847229719163,
      "num_tokens": 50303911.0,
      "step": 27350
    },
    {
      "entropy": 4.988085603713989,
      "epoch": 2.2849148012028064,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044746796243139077,
      "loss": 4.5928,
      "mean_token_accuracy": 0.23940402567386626,
      "num_tokens": 50312708.0,
      "step": 27355
    },
    {
      "entropy": 5.090328931808472,
      "epoch": 2.2853324423655197,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044744884218581217,
      "loss": 4.7233,
      "mean_token_accuracy": 0.22774392366409302,
      "num_tokens": 50321440.0,
      "step": 27360
    },
    {
      "entropy": 4.9491805076599125,
      "epoch": 2.2857500835282325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000447429718921309,
      "loss": 4.6611,
      "mean_token_accuracy": 0.22395415902137755,
      "num_tokens": 50330588.0,
      "step": 27365
    },
    {
      "entropy": 4.995754241943359,
      "epoch": 2.2861677246909453,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004474105926382161,
      "loss": 4.6899,
      "mean_token_accuracy": 0.21821847409009934,
      "num_tokens": 50338878.0,
      "step": 27370
    },
    {
      "entropy": 5.000873279571533,
      "epoch": 2.2865853658536586,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004473914633368683,
      "loss": 4.6924,
      "mean_token_accuracy": 0.22276170700788497,
      "num_tokens": 50346765.0,
      "step": 27375
    },
    {
      "entropy": 4.958273839950562,
      "epoch": 2.2870030070163714,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004473723310176003,
      "loss": 4.6587,
      "mean_token_accuracy": 0.23044175654649734,
      "num_tokens": 50355822.0,
      "step": 27380
    },
    {
      "entropy": 5.066641807556152,
      "epoch": 2.2874206481790846,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004473531956807471,
      "loss": 4.7109,
      "mean_token_accuracy": 0.22399798035621643,
      "num_tokens": 50364734.0,
      "step": 27385
    },
    {
      "entropy": 5.116875267028808,
      "epoch": 2.2878382893417974,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044733405732664394,
      "loss": 4.8055,
      "mean_token_accuracy": 0.224037703871727,
      "num_tokens": 50373458.0,
      "step": 27390
    },
    {
      "entropy": 5.228248023986817,
      "epoch": 2.2882559305045107,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004473149159556255,
      "loss": 4.8512,
      "mean_token_accuracy": 0.2143731012940407,
      "num_tokens": 50382148.0,
      "step": 27395
    },
    {
      "entropy": 5.075928211212158,
      "epoch": 2.2886735716672235,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044729577156802714,
      "loss": 4.6974,
      "mean_token_accuracy": 0.22687198519706725,
      "num_tokens": 50391041.0,
      "step": 27400
    },
    {
      "entropy": 4.9424323558807375,
      "epoch": 2.2890912128299363,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004472766241641839,
      "loss": 4.7,
      "mean_token_accuracy": 0.229602213203907,
      "num_tokens": 50400333.0,
      "step": 27405
    },
    {
      "entropy": 4.975290536880493,
      "epoch": 2.2895088539926496,
      "grad_norm": 1.15625,
      "learning_rate": 0.000447257473744431,
      "loss": 4.6435,
      "mean_token_accuracy": 0.23586147129535676,
      "num_tokens": 50409061.0,
      "step": 27410
    },
    {
      "entropy": 5.154702043533325,
      "epoch": 2.2899264951553624,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004472383203091036,
      "loss": 4.8312,
      "mean_token_accuracy": 0.21106395572423936,
      "num_tokens": 50418998.0,
      "step": 27415
    },
    {
      "entropy": 4.986261987686158,
      "epoch": 2.2903441363180757,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004472191638585371,
      "loss": 4.6353,
      "mean_token_accuracy": 0.23293907046318055,
      "num_tokens": 50428118.0,
      "step": 27420
    },
    {
      "entropy": 4.954024124145508,
      "epoch": 2.2907617774807885,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044720000439306694,
      "loss": 4.6346,
      "mean_token_accuracy": 0.22741747200489043,
      "num_tokens": 50436905.0,
      "step": 27425
    },
    {
      "entropy": 5.03270845413208,
      "epoch": 2.2911794186435017,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044718084191302826,
      "loss": 4.7081,
      "mean_token_accuracy": 0.22355110943317413,
      "num_tokens": 50446021.0,
      "step": 27430
    },
    {
      "entropy": 4.983155822753906,
      "epoch": 2.2915970598062145,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044716167641875685,
      "loss": 4.6829,
      "mean_token_accuracy": 0.2263592302799225,
      "num_tokens": 50454801.0,
      "step": 27435
    },
    {
      "entropy": 5.010007095336914,
      "epoch": 2.2920147009689273,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000447142507910588,
      "loss": 4.6776,
      "mean_token_accuracy": 0.2303974837064743,
      "num_tokens": 50464080.0,
      "step": 27440
    },
    {
      "entropy": 4.993154859542846,
      "epoch": 2.2924323421316406,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044712333638885726,
      "loss": 4.7056,
      "mean_token_accuracy": 0.22144514471292495,
      "num_tokens": 50473757.0,
      "step": 27445
    },
    {
      "entropy": 5.053878402709961,
      "epoch": 2.2928499832943534,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044710416185390043,
      "loss": 4.7724,
      "mean_token_accuracy": 0.2274713099002838,
      "num_tokens": 50482924.0,
      "step": 27450
    },
    {
      "entropy": 5.071264600753784,
      "epoch": 2.2932676244570667,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004470849843060531,
      "loss": 4.6363,
      "mean_token_accuracy": 0.23139723092317582,
      "num_tokens": 50491912.0,
      "step": 27455
    },
    {
      "entropy": 5.0419634819030765,
      "epoch": 2.2936852656197795,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044706580374565094,
      "loss": 4.7531,
      "mean_token_accuracy": 0.2238243266940117,
      "num_tokens": 50501787.0,
      "step": 27460
    },
    {
      "entropy": 5.005257892608642,
      "epoch": 2.2941029067824923,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004470466201730298,
      "loss": 4.6929,
      "mean_token_accuracy": 0.2274263858795166,
      "num_tokens": 50511678.0,
      "step": 27465
    },
    {
      "entropy": 4.966762399673462,
      "epoch": 2.2945205479452055,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004470274335885254,
      "loss": 4.6295,
      "mean_token_accuracy": 0.2234866663813591,
      "num_tokens": 50520123.0,
      "step": 27470
    },
    {
      "entropy": 4.992028903961182,
      "epoch": 2.2949381891079184,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004470082439924737,
      "loss": 4.6713,
      "mean_token_accuracy": 0.23445682972669601,
      "num_tokens": 50529306.0,
      "step": 27475
    },
    {
      "entropy": 5.032145023345947,
      "epoch": 2.2953558302706316,
      "grad_norm": 1.125,
      "learning_rate": 0.0004469890513852107,
      "loss": 4.7324,
      "mean_token_accuracy": 0.22076518684625626,
      "num_tokens": 50538417.0,
      "step": 27480
    },
    {
      "entropy": 5.011907911300659,
      "epoch": 2.2957734714333444,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004469698557670724,
      "loss": 4.7213,
      "mean_token_accuracy": 0.22379768937826156,
      "num_tokens": 50547462.0,
      "step": 27485
    },
    {
      "entropy": 5.060964679718017,
      "epoch": 2.2961911125960572,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004469506571383947,
      "loss": 4.7102,
      "mean_token_accuracy": 0.2228892296552658,
      "num_tokens": 50557281.0,
      "step": 27490
    },
    {
      "entropy": 5.004697513580322,
      "epoch": 2.2966087537587705,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004469314554995137,
      "loss": 4.6376,
      "mean_token_accuracy": 0.224275866150856,
      "num_tokens": 50566553.0,
      "step": 27495
    },
    {
      "entropy": 5.0497997283935545,
      "epoch": 2.2970263949214833,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044691225085076577,
      "loss": 4.7028,
      "mean_token_accuracy": 0.22539513260126115,
      "num_tokens": 50575699.0,
      "step": 27500
    },
    {
      "entropy": 5.059510135650635,
      "epoch": 2.2974440360841966,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004468930431924869,
      "loss": 4.8088,
      "mean_token_accuracy": 0.2228526458144188,
      "num_tokens": 50586210.0,
      "step": 27505
    },
    {
      "entropy": 5.039504623413086,
      "epoch": 2.2978616772469094,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044687383252501337,
      "loss": 4.6624,
      "mean_token_accuracy": 0.22934939712285995,
      "num_tokens": 50594821.0,
      "step": 27510
    },
    {
      "entropy": 4.984533596038818,
      "epoch": 2.2982793184096226,
      "grad_norm": 1.125,
      "learning_rate": 0.0004468546188486815,
      "loss": 4.7453,
      "mean_token_accuracy": 0.2223721981048584,
      "num_tokens": 50603705.0,
      "step": 27515
    },
    {
      "entropy": 5.204648685455322,
      "epoch": 2.2986969595723354,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044683540216382766,
      "loss": 4.8811,
      "mean_token_accuracy": 0.21511216908693315,
      "num_tokens": 50613136.0,
      "step": 27520
    },
    {
      "entropy": 4.999916458129883,
      "epoch": 2.2991146007350483,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004468161824707883,
      "loss": 4.5958,
      "mean_token_accuracy": 0.23518035113811492,
      "num_tokens": 50621958.0,
      "step": 27525
    },
    {
      "entropy": 4.9919655323028564,
      "epoch": 2.2995322418977615,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044679695976989985,
      "loss": 4.6715,
      "mean_token_accuracy": 0.227679942548275,
      "num_tokens": 50630950.0,
      "step": 27530
    },
    {
      "entropy": 5.001644277572632,
      "epoch": 2.2999498830604743,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044677773406149885,
      "loss": 4.6972,
      "mean_token_accuracy": 0.22512714564800262,
      "num_tokens": 50640788.0,
      "step": 27535
    },
    {
      "entropy": 4.913539123535156,
      "epoch": 2.3003675242231876,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044675850534592185,
      "loss": 4.7034,
      "mean_token_accuracy": 0.22921790182590485,
      "num_tokens": 50650566.0,
      "step": 27540
    },
    {
      "entropy": 5.024126386642456,
      "epoch": 2.3007851653859004,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044673927362350545,
      "loss": 4.6654,
      "mean_token_accuracy": 0.23060911595821382,
      "num_tokens": 50659455.0,
      "step": 27545
    },
    {
      "entropy": 5.201483297348022,
      "epoch": 2.3012028065486136,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044672003889458634,
      "loss": 4.9223,
      "mean_token_accuracy": 0.20983488708734513,
      "num_tokens": 50670012.0,
      "step": 27550
    },
    {
      "entropy": 4.994013977050781,
      "epoch": 2.3016204477113265,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044670080115950117,
      "loss": 4.706,
      "mean_token_accuracy": 0.2245055541396141,
      "num_tokens": 50678955.0,
      "step": 27555
    },
    {
      "entropy": 5.002644729614258,
      "epoch": 2.3020380888740393,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044668156041858696,
      "loss": 4.737,
      "mean_token_accuracy": 0.21851409524679183,
      "num_tokens": 50687955.0,
      "step": 27560
    },
    {
      "entropy": 5.015187788009643,
      "epoch": 2.3024557300367525,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044666231667218025,
      "loss": 4.7135,
      "mean_token_accuracy": 0.22600238174200057,
      "num_tokens": 50697031.0,
      "step": 27565
    },
    {
      "entropy": 5.052636575698853,
      "epoch": 2.3028733711994653,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044664306992061816,
      "loss": 4.791,
      "mean_token_accuracy": 0.22223135232925414,
      "num_tokens": 50706475.0,
      "step": 27570
    },
    {
      "entropy": 5.0940831184387205,
      "epoch": 2.3032910123621786,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044662382016423747,
      "loss": 4.7401,
      "mean_token_accuracy": 0.22178902626037597,
      "num_tokens": 50715848.0,
      "step": 27575
    },
    {
      "entropy": 5.049374151229858,
      "epoch": 2.3037086535248914,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004466045674033752,
      "loss": 4.7438,
      "mean_token_accuracy": 0.22754719108343124,
      "num_tokens": 50725354.0,
      "step": 27580
    },
    {
      "entropy": 5.040148544311523,
      "epoch": 2.304126294687604,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004465853116383684,
      "loss": 4.7411,
      "mean_token_accuracy": 0.22556252777576447,
      "num_tokens": 50734185.0,
      "step": 27585
    },
    {
      "entropy": 5.102357196807861,
      "epoch": 2.3045439358503175,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044656605286955424,
      "loss": 4.841,
      "mean_token_accuracy": 0.22240449488162994,
      "num_tokens": 50743030.0,
      "step": 27590
    },
    {
      "entropy": 5.018683338165284,
      "epoch": 2.3049615770130303,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004465467910972698,
      "loss": 4.6685,
      "mean_token_accuracy": 0.23249029517173767,
      "num_tokens": 50751525.0,
      "step": 27595
    },
    {
      "entropy": 4.941560983657837,
      "epoch": 2.3053792181757435,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004465275263218523,
      "loss": 4.5827,
      "mean_token_accuracy": 0.23341618925333024,
      "num_tokens": 50760877.0,
      "step": 27600
    },
    {
      "entropy": 5.074928426742554,
      "epoch": 2.3057968593384564,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044650825854363895,
      "loss": 4.8486,
      "mean_token_accuracy": 0.21145474016666413,
      "num_tokens": 50770560.0,
      "step": 27605
    },
    {
      "entropy": 5.117348098754883,
      "epoch": 2.306214500501169,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004464889877629671,
      "loss": 4.8279,
      "mean_token_accuracy": 0.21891250759363173,
      "num_tokens": 50780072.0,
      "step": 27610
    },
    {
      "entropy": 5.0503545761108395,
      "epoch": 2.3066321416638824,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004464697139801741,
      "loss": 4.6812,
      "mean_token_accuracy": 0.22688602209091185,
      "num_tokens": 50788832.0,
      "step": 27615
    },
    {
      "entropy": 5.059568262100219,
      "epoch": 2.3070497828265952,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044645043719559736,
      "loss": 4.7721,
      "mean_token_accuracy": 0.2215155303478241,
      "num_tokens": 50798969.0,
      "step": 27620
    },
    {
      "entropy": 5.029880809783935,
      "epoch": 2.3074674239893085,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044643115740957437,
      "loss": 4.6904,
      "mean_token_accuracy": 0.23164096623659133,
      "num_tokens": 50807946.0,
      "step": 27625
    },
    {
      "entropy": 4.943798017501831,
      "epoch": 2.3078850651520213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004464118746224426,
      "loss": 4.5835,
      "mean_token_accuracy": 0.2385125681757927,
      "num_tokens": 50816874.0,
      "step": 27630
    },
    {
      "entropy": 4.975468301773072,
      "epoch": 2.3083027063147346,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004463925888345396,
      "loss": 4.7487,
      "mean_token_accuracy": 0.22720092684030532,
      "num_tokens": 50826151.0,
      "step": 27635
    },
    {
      "entropy": 5.041323232650757,
      "epoch": 2.3087203474774474,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044637330004620307,
      "loss": 4.7164,
      "mean_token_accuracy": 0.21675264239311218,
      "num_tokens": 50835769.0,
      "step": 27640
    },
    {
      "entropy": 5.073754835128784,
      "epoch": 2.3091379886401606,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044635400825777065,
      "loss": 4.7544,
      "mean_token_accuracy": 0.22287668734788896,
      "num_tokens": 50844937.0,
      "step": 27645
    },
    {
      "entropy": 5.084359216690063,
      "epoch": 2.3095556298028734,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044633471346957996,
      "loss": 4.7094,
      "mean_token_accuracy": 0.23232500553131102,
      "num_tokens": 50853783.0,
      "step": 27650
    },
    {
      "entropy": 5.059469413757324,
      "epoch": 2.3099732709655862,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044631541568196896,
      "loss": 4.762,
      "mean_token_accuracy": 0.22747031301259996,
      "num_tokens": 50863566.0,
      "step": 27655
    },
    {
      "entropy": 5.039958810806274,
      "epoch": 2.3103909121282995,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044629611489527537,
      "loss": 4.7406,
      "mean_token_accuracy": 0.22736917436122894,
      "num_tokens": 50872423.0,
      "step": 27660
    },
    {
      "entropy": 5.035761404037475,
      "epoch": 2.3108085532910123,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004462768111098371,
      "loss": 4.6858,
      "mean_token_accuracy": 0.2249922424554825,
      "num_tokens": 50881403.0,
      "step": 27665
    },
    {
      "entropy": 5.0411741733551025,
      "epoch": 2.3112261944537256,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044625750432599213,
      "loss": 4.6856,
      "mean_token_accuracy": 0.22656692415475846,
      "num_tokens": 50890774.0,
      "step": 27670
    },
    {
      "entropy": 5.075244903564453,
      "epoch": 2.3116438356164384,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044623819454407844,
      "loss": 4.8182,
      "mean_token_accuracy": 0.22001171112060547,
      "num_tokens": 50901312.0,
      "step": 27675
    },
    {
      "entropy": 5.097515964508057,
      "epoch": 2.312061476779151,
      "grad_norm": 1.078125,
      "learning_rate": 0.000446218881764434,
      "loss": 4.776,
      "mean_token_accuracy": 0.22360849231481553,
      "num_tokens": 50910334.0,
      "step": 27680
    },
    {
      "entropy": 5.16759557723999,
      "epoch": 2.3124791179418644,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044619956598739686,
      "loss": 4.8176,
      "mean_token_accuracy": 0.21573885679244995,
      "num_tokens": 50919503.0,
      "step": 27685
    },
    {
      "entropy": 5.072400617599487,
      "epoch": 2.3128967591045773,
      "grad_norm": 1.25,
      "learning_rate": 0.00044618024721330523,
      "loss": 4.7335,
      "mean_token_accuracy": 0.22262586653232574,
      "num_tokens": 50928170.0,
      "step": 27690
    },
    {
      "entropy": 5.024795436859131,
      "epoch": 2.3133144002672905,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004461609254424974,
      "loss": 4.7044,
      "mean_token_accuracy": 0.2221684694290161,
      "num_tokens": 50937374.0,
      "step": 27695
    },
    {
      "entropy": 4.9773341655731205,
      "epoch": 2.3137320414300033,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044614160067531157,
      "loss": 4.6242,
      "mean_token_accuracy": 0.22634736746549605,
      "num_tokens": 50946786.0,
      "step": 27700
    },
    {
      "entropy": 5.017122888565064,
      "epoch": 2.314149682592716,
      "grad_norm": 1.09375,
      "learning_rate": 0.000446122272912086,
      "loss": 4.7533,
      "mean_token_accuracy": 0.22231768369674682,
      "num_tokens": 50956123.0,
      "step": 27705
    },
    {
      "entropy": 5.008034038543701,
      "epoch": 2.3145673237554294,
      "grad_norm": 1.1875,
      "learning_rate": 0.000446102942153159,
      "loss": 4.6589,
      "mean_token_accuracy": 0.23435151427984238,
      "num_tokens": 50964736.0,
      "step": 27710
    },
    {
      "entropy": 5.113933515548706,
      "epoch": 2.314984964918142,
      "grad_norm": 1.125,
      "learning_rate": 0.0004460836083988691,
      "loss": 4.6923,
      "mean_token_accuracy": 0.22385190427303314,
      "num_tokens": 50974529.0,
      "step": 27715
    },
    {
      "entropy": 4.969118690490722,
      "epoch": 2.3154026060808555,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044606427164955466,
      "loss": 4.6366,
      "mean_token_accuracy": 0.23396629244089126,
      "num_tokens": 50983384.0,
      "step": 27720
    },
    {
      "entropy": 5.088746118545532,
      "epoch": 2.3158202472435683,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004460449319055543,
      "loss": 4.7588,
      "mean_token_accuracy": 0.22064645290374757,
      "num_tokens": 50992239.0,
      "step": 27725
    },
    {
      "entropy": 5.0672139644622805,
      "epoch": 2.316237888406281,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004460255891672064,
      "loss": 4.7144,
      "mean_token_accuracy": 0.22754352688789367,
      "num_tokens": 51000717.0,
      "step": 27730
    },
    {
      "entropy": 5.006088733673096,
      "epoch": 2.3166555295689943,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044600624343484985,
      "loss": 4.6563,
      "mean_token_accuracy": 0.22885761111974717,
      "num_tokens": 51008777.0,
      "step": 27735
    },
    {
      "entropy": 4.999341773986816,
      "epoch": 2.317073170731707,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004459868947088231,
      "loss": 4.6099,
      "mean_token_accuracy": 0.23109411746263503,
      "num_tokens": 51017912.0,
      "step": 27740
    },
    {
      "entropy": 5.023294115066529,
      "epoch": 2.3174908118944204,
      "grad_norm": 1.125,
      "learning_rate": 0.00044596754298946494,
      "loss": 4.7464,
      "mean_token_accuracy": 0.22171511352062226,
      "num_tokens": 51026959.0,
      "step": 27745
    },
    {
      "entropy": 5.016162157058716,
      "epoch": 2.3179084530571332,
      "grad_norm": 1.0,
      "learning_rate": 0.00044594818827711416,
      "loss": 4.6944,
      "mean_token_accuracy": 0.22429829388856887,
      "num_tokens": 51036329.0,
      "step": 27750
    },
    {
      "entropy": 5.034688329696655,
      "epoch": 2.3183260942198465,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004459288305721095,
      "loss": 4.7069,
      "mean_token_accuracy": 0.22386604994535447,
      "num_tokens": 51044828.0,
      "step": 27755
    },
    {
      "entropy": 4.9763068675994875,
      "epoch": 2.3187437353825593,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044590946987479,
      "loss": 4.5798,
      "mean_token_accuracy": 0.23129891902208327,
      "num_tokens": 51052929.0,
      "step": 27760
    },
    {
      "entropy": 4.995629835128784,
      "epoch": 2.3191613765452725,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004458901061854944,
      "loss": 4.6583,
      "mean_token_accuracy": 0.23254308104515076,
      "num_tokens": 51062264.0,
      "step": 27765
    },
    {
      "entropy": 4.978682899475098,
      "epoch": 2.3195790177079854,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044587073950456193,
      "loss": 4.6612,
      "mean_token_accuracy": 0.22852080315351486,
      "num_tokens": 51071075.0,
      "step": 27770
    },
    {
      "entropy": 4.950444936752319,
      "epoch": 2.319996658870698,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004458513698323314,
      "loss": 4.6361,
      "mean_token_accuracy": 0.22704460769891738,
      "num_tokens": 51080374.0,
      "step": 27775
    },
    {
      "entropy": 4.993317317962647,
      "epoch": 2.3204143000334114,
      "grad_norm": 1.125,
      "learning_rate": 0.000445831997169142,
      "loss": 4.6528,
      "mean_token_accuracy": 0.23386246114969253,
      "num_tokens": 51088882.0,
      "step": 27780
    },
    {
      "entropy": 5.077838897705078,
      "epoch": 2.3208319411961242,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004458126215153329,
      "loss": 4.6632,
      "mean_token_accuracy": 0.22637011855840683,
      "num_tokens": 51098199.0,
      "step": 27785
    },
    {
      "entropy": 5.066616725921631,
      "epoch": 2.3212495823588375,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044579324287124324,
      "loss": 4.765,
      "mean_token_accuracy": 0.2240490809082985,
      "num_tokens": 51106466.0,
      "step": 27790
    },
    {
      "entropy": 4.9019917964935305,
      "epoch": 2.3216672235215503,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044577386123721225,
      "loss": 4.6265,
      "mean_token_accuracy": 0.23010175675153732,
      "num_tokens": 51115308.0,
      "step": 27795
    },
    {
      "entropy": 5.066980075836182,
      "epoch": 2.322084864684263,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044575447661357925,
      "loss": 4.7109,
      "mean_token_accuracy": 0.2229936823248863,
      "num_tokens": 51124748.0,
      "step": 27800
    },
    {
      "entropy": 5.114909744262695,
      "epoch": 2.3225025058469764,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044573508900068367,
      "loss": 4.7291,
      "mean_token_accuracy": 0.22254949808120728,
      "num_tokens": 51133977.0,
      "step": 27805
    },
    {
      "entropy": 5.00611367225647,
      "epoch": 2.322920147009689,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004457156983988648,
      "loss": 4.6539,
      "mean_token_accuracy": 0.22511548846960067,
      "num_tokens": 51142456.0,
      "step": 27810
    },
    {
      "entropy": 5.026314544677734,
      "epoch": 2.3233377881724024,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044569630480846214,
      "loss": 4.736,
      "mean_token_accuracy": 0.22426892966032028,
      "num_tokens": 51151600.0,
      "step": 27815
    },
    {
      "entropy": 4.985334396362305,
      "epoch": 2.3237554293351153,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004456769082298152,
      "loss": 4.6269,
      "mean_token_accuracy": 0.24027502089738845,
      "num_tokens": 51159814.0,
      "step": 27820
    },
    {
      "entropy": 4.91029725074768,
      "epoch": 2.324173070497828,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044565750866326355,
      "loss": 4.5981,
      "mean_token_accuracy": 0.2404101848602295,
      "num_tokens": 51168010.0,
      "step": 27825
    },
    {
      "entropy": 5.058001327514648,
      "epoch": 2.3245907116605413,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004456381061091467,
      "loss": 4.7416,
      "mean_token_accuracy": 0.2252634957432747,
      "num_tokens": 51177600.0,
      "step": 27830
    },
    {
      "entropy": 5.094222211837769,
      "epoch": 2.325008352823254,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044561870056780445,
      "loss": 4.8009,
      "mean_token_accuracy": 0.21356835663318635,
      "num_tokens": 51187171.0,
      "step": 27835
    },
    {
      "entropy": 5.070850276947022,
      "epoch": 2.3254259939859674,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004455992920395766,
      "loss": 4.7289,
      "mean_token_accuracy": 0.2280430555343628,
      "num_tokens": 51196477.0,
      "step": 27840
    },
    {
      "entropy": 4.9825255393981935,
      "epoch": 2.32584363514868,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044557988052480263,
      "loss": 4.6307,
      "mean_token_accuracy": 0.2285969913005829,
      "num_tokens": 51204883.0,
      "step": 27845
    },
    {
      "entropy": 5.046818447113037,
      "epoch": 2.326261276311393,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044556046602382257,
      "loss": 4.7379,
      "mean_token_accuracy": 0.23230107128620148,
      "num_tokens": 51214456.0,
      "step": 27850
    },
    {
      "entropy": 5.1356518268585205,
      "epoch": 2.3266789174741063,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004455410485369763,
      "loss": 4.843,
      "mean_token_accuracy": 0.22161711305379866,
      "num_tokens": 51224752.0,
      "step": 27855
    },
    {
      "entropy": 5.098272800445557,
      "epoch": 2.327096558636819,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004455216280646037,
      "loss": 4.7094,
      "mean_token_accuracy": 0.2313657134771347,
      "num_tokens": 51233780.0,
      "step": 27860
    },
    {
      "entropy": 5.137462997436524,
      "epoch": 2.3275141997995323,
      "grad_norm": 1.125,
      "learning_rate": 0.0004455022046070446,
      "loss": 4.8065,
      "mean_token_accuracy": 0.21847710907459258,
      "num_tokens": 51242669.0,
      "step": 27865
    },
    {
      "entropy": 5.074748516082764,
      "epoch": 2.327931840962245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044548277816463925,
      "loss": 4.7621,
      "mean_token_accuracy": 0.21940956264734268,
      "num_tokens": 51252081.0,
      "step": 27870
    },
    {
      "entropy": 5.086415767669678,
      "epoch": 2.3283494821249584,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044546334873772777,
      "loss": 4.7767,
      "mean_token_accuracy": 0.21764674335718154,
      "num_tokens": 51262103.0,
      "step": 27875
    },
    {
      "entropy": 5.07023286819458,
      "epoch": 2.328767123287671,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044544391632665004,
      "loss": 4.7836,
      "mean_token_accuracy": 0.22119968831539155,
      "num_tokens": 51270822.0,
      "step": 27880
    },
    {
      "entropy": 5.022617149353027,
      "epoch": 2.3291847644503845,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004454244809317465,
      "loss": 4.7419,
      "mean_token_accuracy": 0.21623767018318177,
      "num_tokens": 51280376.0,
      "step": 27885
    },
    {
      "entropy": 5.0734786033630375,
      "epoch": 2.3296024056130973,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044540504255335717,
      "loss": 4.7928,
      "mean_token_accuracy": 0.21544777005910873,
      "num_tokens": 51289672.0,
      "step": 27890
    },
    {
      "entropy": 5.034689998626709,
      "epoch": 2.33002004677581,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004453856011918225,
      "loss": 4.6469,
      "mean_token_accuracy": 0.223149636387825,
      "num_tokens": 51299585.0,
      "step": 27895
    },
    {
      "entropy": 4.924955701828003,
      "epoch": 2.3304376879385234,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044536615684748286,
      "loss": 4.5892,
      "mean_token_accuracy": 0.23178096264600753,
      "num_tokens": 51308700.0,
      "step": 27900
    },
    {
      "entropy": 4.992073488235474,
      "epoch": 2.330855329101236,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004453467095206785,
      "loss": 4.6819,
      "mean_token_accuracy": 0.22828032225370407,
      "num_tokens": 51316876.0,
      "step": 27905
    },
    {
      "entropy": 5.094878768920898,
      "epoch": 2.3312729702639494,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004453272592117499,
      "loss": 4.6875,
      "mean_token_accuracy": 0.23049658834934234,
      "num_tokens": 51325347.0,
      "step": 27910
    },
    {
      "entropy": 4.945746040344238,
      "epoch": 2.3316906114266622,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004453078059210377,
      "loss": 4.6203,
      "mean_token_accuracy": 0.23094865083694457,
      "num_tokens": 51333938.0,
      "step": 27915
    },
    {
      "entropy": 5.125436639785766,
      "epoch": 2.332108252589375,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044528834964888234,
      "loss": 4.9204,
      "mean_token_accuracy": 0.20959283113479615,
      "num_tokens": 51343553.0,
      "step": 27920
    },
    {
      "entropy": 5.045924329757691,
      "epoch": 2.3325258937520883,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044526889039562444,
      "loss": 4.7158,
      "mean_token_accuracy": 0.22900999784469606,
      "num_tokens": 51352272.0,
      "step": 27925
    },
    {
      "entropy": 5.064101648330689,
      "epoch": 2.332943534914801,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044524942816160463,
      "loss": 4.6907,
      "mean_token_accuracy": 0.2216926246881485,
      "num_tokens": 51361740.0,
      "step": 27930
    },
    {
      "entropy": 4.976784992218017,
      "epoch": 2.3333611760775144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044522996294716374,
      "loss": 4.5692,
      "mean_token_accuracy": 0.2365530401468277,
      "num_tokens": 51369896.0,
      "step": 27935
    },
    {
      "entropy": 5.034302234649658,
      "epoch": 2.333778817240227,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044521049475264237,
      "loss": 4.6932,
      "mean_token_accuracy": 0.22425222247838975,
      "num_tokens": 51379774.0,
      "step": 27940
    },
    {
      "entropy": 4.941820669174194,
      "epoch": 2.33419645840294,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004451910235783814,
      "loss": 4.6389,
      "mean_token_accuracy": 0.23539015501737595,
      "num_tokens": 51389626.0,
      "step": 27945
    },
    {
      "entropy": 5.021554660797119,
      "epoch": 2.3346140995656532,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044517154942472166,
      "loss": 4.7593,
      "mean_token_accuracy": 0.22947343587875366,
      "num_tokens": 51398819.0,
      "step": 27950
    },
    {
      "entropy": 5.001364231109619,
      "epoch": 2.335031740728366,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004451520722920042,
      "loss": 4.6545,
      "mean_token_accuracy": 0.22803965955972672,
      "num_tokens": 51407946.0,
      "step": 27955
    },
    {
      "entropy": 4.992988920211792,
      "epoch": 2.3354493818910793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004451325921805698,
      "loss": 4.5803,
      "mean_token_accuracy": 0.23716225177049638,
      "num_tokens": 51417306.0,
      "step": 27960
    },
    {
      "entropy": 4.95765495300293,
      "epoch": 2.335867023053792,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004451131090907597,
      "loss": 4.7336,
      "mean_token_accuracy": 0.22235341370105743,
      "num_tokens": 51427341.0,
      "step": 27965
    },
    {
      "entropy": 4.999336290359497,
      "epoch": 2.336284664216505,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004450936230229148,
      "loss": 4.6363,
      "mean_token_accuracy": 0.22434596717357635,
      "num_tokens": 51436173.0,
      "step": 27970
    },
    {
      "entropy": 5.010757637023926,
      "epoch": 2.336702305379218,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004450741339773763,
      "loss": 4.6273,
      "mean_token_accuracy": 0.2314363792538643,
      "num_tokens": 51445871.0,
      "step": 27975
    },
    {
      "entropy": 4.90652642250061,
      "epoch": 2.337119946541931,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044505464195448535,
      "loss": 4.6282,
      "mean_token_accuracy": 0.23522884845733644,
      "num_tokens": 51454380.0,
      "step": 27980
    },
    {
      "entropy": 5.068231868743896,
      "epoch": 2.3375375877046443,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044503514695458325,
      "loss": 4.7965,
      "mean_token_accuracy": 0.22003434598445892,
      "num_tokens": 51463650.0,
      "step": 27985
    },
    {
      "entropy": 5.05450587272644,
      "epoch": 2.337955228867357,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004450156489780112,
      "loss": 4.7251,
      "mean_token_accuracy": 0.22802942544221877,
      "num_tokens": 51471755.0,
      "step": 27990
    },
    {
      "entropy": 4.970090436935425,
      "epoch": 2.3383728700300703,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004449961480251106,
      "loss": 4.6534,
      "mean_token_accuracy": 0.23050907105207444,
      "num_tokens": 51481295.0,
      "step": 27995
    },
    {
      "entropy": 5.017501878738403,
      "epoch": 2.338790511192783,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004449766440962227,
      "loss": 4.71,
      "mean_token_accuracy": 0.22411434203386307,
      "num_tokens": 51490798.0,
      "step": 28000
    },
    {
      "entropy": 5.000924730300904,
      "epoch": 2.3392081523554964,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004449571371916891,
      "loss": 4.726,
      "mean_token_accuracy": 0.2213937908411026,
      "num_tokens": 51500043.0,
      "step": 28005
    },
    {
      "entropy": 5.05642409324646,
      "epoch": 2.339625793518209,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004449376273118513,
      "loss": 4.7563,
      "mean_token_accuracy": 0.22134494334459304,
      "num_tokens": 51509688.0,
      "step": 28010
    },
    {
      "entropy": 5.054797792434693,
      "epoch": 2.340043434680922,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004449181144570507,
      "loss": 4.7231,
      "mean_token_accuracy": 0.22956063449382783,
      "num_tokens": 51519735.0,
      "step": 28015
    },
    {
      "entropy": 5.062149000167847,
      "epoch": 2.3404610758436353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000444898598627629,
      "loss": 4.786,
      "mean_token_accuracy": 0.2219192326068878,
      "num_tokens": 51529153.0,
      "step": 28020
    },
    {
      "entropy": 5.047229814529419,
      "epoch": 2.340878717006348,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004448790798239278,
      "loss": 4.665,
      "mean_token_accuracy": 0.23086644858121871,
      "num_tokens": 51538707.0,
      "step": 28025
    },
    {
      "entropy": 5.0190154075622555,
      "epoch": 2.3412963581690613,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044485955804628884,
      "loss": 4.7496,
      "mean_token_accuracy": 0.22125159651041032,
      "num_tokens": 51547730.0,
      "step": 28030
    },
    {
      "entropy": 5.060393762588501,
      "epoch": 2.341713999331774,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044484003329505384,
      "loss": 4.7834,
      "mean_token_accuracy": 0.22066857069730758,
      "num_tokens": 51557714.0,
      "step": 28035
    },
    {
      "entropy": 5.08348650932312,
      "epoch": 2.342131640494487,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004448205055705647,
      "loss": 4.6781,
      "mean_token_accuracy": 0.22936010211706162,
      "num_tokens": 51567031.0,
      "step": 28040
    },
    {
      "entropy": 5.1129076957702635,
      "epoch": 2.3425492816572,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004448009748731631,
      "loss": 4.7454,
      "mean_token_accuracy": 0.22098131775856017,
      "num_tokens": 51575730.0,
      "step": 28045
    },
    {
      "entropy": 5.1291125297546385,
      "epoch": 2.342966922819913,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004447814412031911,
      "loss": 4.8119,
      "mean_token_accuracy": 0.2210909366607666,
      "num_tokens": 51585126.0,
      "step": 28050
    },
    {
      "entropy": 5.0607218742370605,
      "epoch": 2.3433845639826263,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044476190456099057,
      "loss": 4.6673,
      "mean_token_accuracy": 0.22882868647575377,
      "num_tokens": 51594153.0,
      "step": 28055
    },
    {
      "entropy": 4.94560751914978,
      "epoch": 2.343802205145339,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004447423649469036,
      "loss": 4.6452,
      "mean_token_accuracy": 0.23466755598783492,
      "num_tokens": 51603449.0,
      "step": 28060
    },
    {
      "entropy": 5.075608348846435,
      "epoch": 2.344219846308052,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004447228223612722,
      "loss": 4.7526,
      "mean_token_accuracy": 0.21664073914289475,
      "num_tokens": 51612821.0,
      "step": 28065
    },
    {
      "entropy": 5.0933901309967045,
      "epoch": 2.344637487470765,
      "grad_norm": 1.125,
      "learning_rate": 0.0004447032768044385,
      "loss": 4.8307,
      "mean_token_accuracy": 0.21760122627019882,
      "num_tokens": 51622180.0,
      "step": 28070
    },
    {
      "entropy": 5.012038278579712,
      "epoch": 2.345055128633478,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004446837282767447,
      "loss": 4.721,
      "mean_token_accuracy": 0.22166367024183273,
      "num_tokens": 51631695.0,
      "step": 28075
    },
    {
      "entropy": 5.001949644088745,
      "epoch": 2.3454727697961912,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044466417677853294,
      "loss": 4.5615,
      "mean_token_accuracy": 0.24256252944469453,
      "num_tokens": 51640328.0,
      "step": 28080
    },
    {
      "entropy": 5.032220554351807,
      "epoch": 2.345890410958904,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004446446223101456,
      "loss": 4.7717,
      "mean_token_accuracy": 0.22246777266263962,
      "num_tokens": 51650188.0,
      "step": 28085
    },
    {
      "entropy": 4.914929676055908,
      "epoch": 2.346308052121617,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044462506487192487,
      "loss": 4.5781,
      "mean_token_accuracy": 0.23452970683574675,
      "num_tokens": 51658879.0,
      "step": 28090
    },
    {
      "entropy": 5.0508630752563475,
      "epoch": 2.34672569328433,
      "grad_norm": 1.125,
      "learning_rate": 0.00044460550446421323,
      "loss": 4.7053,
      "mean_token_accuracy": 0.22444893121719361,
      "num_tokens": 51666939.0,
      "step": 28095
    },
    {
      "entropy": 5.10247688293457,
      "epoch": 2.347143334447043,
      "grad_norm": 1.09375,
      "learning_rate": 0.000444585941087353,
      "loss": 4.7461,
      "mean_token_accuracy": 0.2283298507332802,
      "num_tokens": 51676299.0,
      "step": 28100
    },
    {
      "entropy": 5.054486417770386,
      "epoch": 2.347560975609756,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004445663747416868,
      "loss": 4.665,
      "mean_token_accuracy": 0.22122225016355515,
      "num_tokens": 51685094.0,
      "step": 28105
    },
    {
      "entropy": 4.986119079589844,
      "epoch": 2.347978616772469,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044454680542755715,
      "loss": 4.6941,
      "mean_token_accuracy": 0.22054229229688643,
      "num_tokens": 51694948.0,
      "step": 28110
    },
    {
      "entropy": 4.900013780593872,
      "epoch": 2.3483962579351823,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004445272331453066,
      "loss": 4.5816,
      "mean_token_accuracy": 0.2306699886918068,
      "num_tokens": 51704186.0,
      "step": 28115
    },
    {
      "entropy": 4.946199941635132,
      "epoch": 2.348813899097895,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044450765789527777,
      "loss": 4.6153,
      "mean_token_accuracy": 0.23781894892454147,
      "num_tokens": 51712976.0,
      "step": 28120
    },
    {
      "entropy": 5.063462495803833,
      "epoch": 2.3492315402606083,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004444880796778133,
      "loss": 4.7657,
      "mean_token_accuracy": 0.22505522668361663,
      "num_tokens": 51723489.0,
      "step": 28125
    },
    {
      "entropy": 5.036405944824219,
      "epoch": 2.349649181423321,
      "grad_norm": 1.125,
      "learning_rate": 0.000444468498493256,
      "loss": 4.7588,
      "mean_token_accuracy": 0.2288749098777771,
      "num_tokens": 51732742.0,
      "step": 28130
    },
    {
      "entropy": 5.042254543304443,
      "epoch": 2.350066822586034,
      "grad_norm": 1.125,
      "learning_rate": 0.00044444891434194863,
      "loss": 4.6855,
      "mean_token_accuracy": 0.2193264737725258,
      "num_tokens": 51741865.0,
      "step": 28135
    },
    {
      "entropy": 4.976113510131836,
      "epoch": 2.350484463748747,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004444293272242341,
      "loss": 4.6103,
      "mean_token_accuracy": 0.23495450168848037,
      "num_tokens": 51750651.0,
      "step": 28140
    },
    {
      "entropy": 4.998328399658203,
      "epoch": 2.35090210491146,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044440973714045517,
      "loss": 4.6919,
      "mean_token_accuracy": 0.22647577077150344,
      "num_tokens": 51760200.0,
      "step": 28145
    },
    {
      "entropy": 4.969443130493164,
      "epoch": 2.3513197460741733,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044439014409095495,
      "loss": 4.5457,
      "mean_token_accuracy": 0.2418439045548439,
      "num_tokens": 51769106.0,
      "step": 28150
    },
    {
      "entropy": 4.9910835266113285,
      "epoch": 2.351737387236886,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004443705480760763,
      "loss": 4.6525,
      "mean_token_accuracy": 0.2290801763534546,
      "num_tokens": 51778538.0,
      "step": 28155
    },
    {
      "entropy": 5.040219736099243,
      "epoch": 2.352155028399599,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044435094909616243,
      "loss": 4.7197,
      "mean_token_accuracy": 0.21862575858831407,
      "num_tokens": 51787850.0,
      "step": 28160
    },
    {
      "entropy": 5.102751588821411,
      "epoch": 2.352572669562312,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004443313471515562,
      "loss": 4.8192,
      "mean_token_accuracy": 0.2183795005083084,
      "num_tokens": 51797487.0,
      "step": 28165
    },
    {
      "entropy": 5.097453784942627,
      "epoch": 2.352990310725025,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044431174224260106,
      "loss": 4.7865,
      "mean_token_accuracy": 0.2202304095029831,
      "num_tokens": 51807025.0,
      "step": 28170
    },
    {
      "entropy": 5.086748838424683,
      "epoch": 2.353407951887738,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044429213436964,
      "loss": 4.707,
      "mean_token_accuracy": 0.22051712423563002,
      "num_tokens": 51816038.0,
      "step": 28175
    },
    {
      "entropy": 5.066150140762329,
      "epoch": 2.353825593050451,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044427252353301637,
      "loss": 4.7557,
      "mean_token_accuracy": 0.21880001276731492,
      "num_tokens": 51825115.0,
      "step": 28180
    },
    {
      "entropy": 5.05403733253479,
      "epoch": 2.354243234213164,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044425290973307345,
      "loss": 4.7502,
      "mean_token_accuracy": 0.21787556260824203,
      "num_tokens": 51834584.0,
      "step": 28185
    },
    {
      "entropy": 5.043879508972168,
      "epoch": 2.354660875375877,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044423329297015463,
      "loss": 4.7225,
      "mean_token_accuracy": 0.2279866486787796,
      "num_tokens": 51843759.0,
      "step": 28190
    },
    {
      "entropy": 4.9934063911437985,
      "epoch": 2.35507851653859,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044421367324460335,
      "loss": 4.7014,
      "mean_token_accuracy": 0.2248843491077423,
      "num_tokens": 51853594.0,
      "step": 28195
    },
    {
      "entropy": 5.087533903121948,
      "epoch": 2.355496157701303,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004441940505567629,
      "loss": 4.8487,
      "mean_token_accuracy": 0.22123677134513856,
      "num_tokens": 51863139.0,
      "step": 28200
    },
    {
      "entropy": 4.978259801864624,
      "epoch": 2.355913798864016,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044417442490697704,
      "loss": 4.6888,
      "mean_token_accuracy": 0.2288337007164955,
      "num_tokens": 51872117.0,
      "step": 28205
    },
    {
      "entropy": 4.94784951210022,
      "epoch": 2.356331440026729,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044415479629558924,
      "loss": 4.5822,
      "mean_token_accuracy": 0.23291940987110138,
      "num_tokens": 51881394.0,
      "step": 28210
    },
    {
      "entropy": 4.959408855438232,
      "epoch": 2.356749081189442,
      "grad_norm": 1.125,
      "learning_rate": 0.00044413516472294315,
      "loss": 4.5927,
      "mean_token_accuracy": 0.2261655181646347,
      "num_tokens": 51890443.0,
      "step": 28215
    },
    {
      "entropy": 5.091761207580566,
      "epoch": 2.357166722352155,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004441155301893823,
      "loss": 4.7897,
      "mean_token_accuracy": 0.2233905240893364,
      "num_tokens": 51900012.0,
      "step": 28220
    },
    {
      "entropy": 5.028991174697876,
      "epoch": 2.357584363514868,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004440958926952506,
      "loss": 4.7753,
      "mean_token_accuracy": 0.22565001547336577,
      "num_tokens": 51909166.0,
      "step": 28225
    },
    {
      "entropy": 5.034745550155639,
      "epoch": 2.358002004677581,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004440762522408917,
      "loss": 4.7543,
      "mean_token_accuracy": 0.21884857416152953,
      "num_tokens": 51918631.0,
      "step": 28230
    },
    {
      "entropy": 5.1232781410217285,
      "epoch": 2.358419645840294,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004440566088266496,
      "loss": 4.8149,
      "mean_token_accuracy": 0.22194276452064515,
      "num_tokens": 51928382.0,
      "step": 28235
    },
    {
      "entropy": 5.093289136886597,
      "epoch": 2.358837287003007,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044403696245286803,
      "loss": 4.662,
      "mean_token_accuracy": 0.22408814132213592,
      "num_tokens": 51937162.0,
      "step": 28240
    },
    {
      "entropy": 4.975168037414551,
      "epoch": 2.3592549281657202,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044401731311989086,
      "loss": 4.7267,
      "mean_token_accuracy": 0.22527031898498534,
      "num_tokens": 51946002.0,
      "step": 28245
    },
    {
      "entropy": 5.016817903518676,
      "epoch": 2.359672569328433,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004439976608280623,
      "loss": 4.669,
      "mean_token_accuracy": 0.22602367848157884,
      "num_tokens": 51955716.0,
      "step": 28250
    },
    {
      "entropy": 4.959379386901856,
      "epoch": 2.360090210491146,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044397800557772614,
      "loss": 4.5857,
      "mean_token_accuracy": 0.23715246617794036,
      "num_tokens": 51964521.0,
      "step": 28255
    },
    {
      "entropy": 5.08110408782959,
      "epoch": 2.360507851653859,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044395834736922675,
      "loss": 4.8126,
      "mean_token_accuracy": 0.22127519100904464,
      "num_tokens": 51975751.0,
      "step": 28260
    },
    {
      "entropy": 5.085246133804321,
      "epoch": 2.360925492816572,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004439386862029081,
      "loss": 4.772,
      "mean_token_accuracy": 0.22363122105598449,
      "num_tokens": 51984126.0,
      "step": 28265
    },
    {
      "entropy": 5.088105916976929,
      "epoch": 2.361343133979285,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004439190220791143,
      "loss": 4.7591,
      "mean_token_accuracy": 0.2174350842833519,
      "num_tokens": 51993396.0,
      "step": 28270
    },
    {
      "entropy": 5.052322483062744,
      "epoch": 2.361760775141998,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044389935499818974,
      "loss": 4.7575,
      "mean_token_accuracy": 0.2261730968952179,
      "num_tokens": 52002986.0,
      "step": 28275
    },
    {
      "entropy": 5.106848812103271,
      "epoch": 2.362178416304711,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004438796849604786,
      "loss": 4.7648,
      "mean_token_accuracy": 0.21834796965122222,
      "num_tokens": 52012506.0,
      "step": 28280
    },
    {
      "entropy": 5.027579498291016,
      "epoch": 2.362596057467424,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044386001196632536,
      "loss": 4.7308,
      "mean_token_accuracy": 0.22303004264831544,
      "num_tokens": 52021560.0,
      "step": 28285
    },
    {
      "entropy": 5.091634368896484,
      "epoch": 2.363013698630137,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004438403360160744,
      "loss": 4.7623,
      "mean_token_accuracy": 0.22454452514648438,
      "num_tokens": 52031388.0,
      "step": 28290
    },
    {
      "entropy": 5.0023720264434814,
      "epoch": 2.36343133979285,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044382065711007,
      "loss": 4.5872,
      "mean_token_accuracy": 0.23187541365623474,
      "num_tokens": 52040038.0,
      "step": 28295
    },
    {
      "entropy": 4.979033613204956,
      "epoch": 2.363848980955563,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044380097524865684,
      "loss": 4.6912,
      "mean_token_accuracy": 0.22259222865104675,
      "num_tokens": 52049459.0,
      "step": 28300
    },
    {
      "entropy": 5.085794305801391,
      "epoch": 2.3642666221182758,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044378129043217947,
      "loss": 4.8513,
      "mean_token_accuracy": 0.2162954643368721,
      "num_tokens": 52058603.0,
      "step": 28305
    },
    {
      "entropy": 5.106886339187622,
      "epoch": 2.364684263280989,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004437616026609824,
      "loss": 4.7569,
      "mean_token_accuracy": 0.21498668789863587,
      "num_tokens": 52068265.0,
      "step": 28310
    },
    {
      "entropy": 5.0479659080505375,
      "epoch": 2.365101904443702,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044374191193541037,
      "loss": 4.6388,
      "mean_token_accuracy": 0.2253141313791275,
      "num_tokens": 52077038.0,
      "step": 28315
    },
    {
      "entropy": 4.973205327987671,
      "epoch": 2.365519545606415,
      "grad_norm": 1.125,
      "learning_rate": 0.00044372221825580803,
      "loss": 4.694,
      "mean_token_accuracy": 0.22454287111759186,
      "num_tokens": 52086790.0,
      "step": 28320
    },
    {
      "entropy": 5.064486837387085,
      "epoch": 2.365937186769128,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044370252162252016,
      "loss": 4.8019,
      "mean_token_accuracy": 0.2132396936416626,
      "num_tokens": 52097213.0,
      "step": 28325
    },
    {
      "entropy": 5.103704452514648,
      "epoch": 2.366354827931841,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004436828220358916,
      "loss": 4.7587,
      "mean_token_accuracy": 0.2174467474222183,
      "num_tokens": 52106119.0,
      "step": 28330
    },
    {
      "entropy": 5.051385164260864,
      "epoch": 2.366772469094554,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004436631194962672,
      "loss": 4.818,
      "mean_token_accuracy": 0.22011719942092894,
      "num_tokens": 52115633.0,
      "step": 28335
    },
    {
      "entropy": 4.982537269592285,
      "epoch": 2.3671901102572668,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044364341400399186,
      "loss": 4.7103,
      "mean_token_accuracy": 0.2287757948040962,
      "num_tokens": 52124592.0,
      "step": 28340
    },
    {
      "entropy": 5.096143627166748,
      "epoch": 2.36760775141998,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044362370555941057,
      "loss": 4.7956,
      "mean_token_accuracy": 0.2177107185125351,
      "num_tokens": 52134113.0,
      "step": 28345
    },
    {
      "entropy": 5.119878482818604,
      "epoch": 2.368025392582693,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004436039941628684,
      "loss": 4.8452,
      "mean_token_accuracy": 0.21320240944623947,
      "num_tokens": 52143515.0,
      "step": 28350
    },
    {
      "entropy": 5.032259750366211,
      "epoch": 2.368443033745406,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004435842798147103,
      "loss": 4.741,
      "mean_token_accuracy": 0.21346283555030823,
      "num_tokens": 52152626.0,
      "step": 28355
    },
    {
      "entropy": 5.00815601348877,
      "epoch": 2.368860674908119,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044356456251528145,
      "loss": 4.7707,
      "mean_token_accuracy": 0.22028156369924545,
      "num_tokens": 52162454.0,
      "step": 28360
    },
    {
      "entropy": 5.066653919219971,
      "epoch": 2.369278316070832,
      "grad_norm": 1.125,
      "learning_rate": 0.0004435448422649271,
      "loss": 4.6908,
      "mean_token_accuracy": 0.228229384124279,
      "num_tokens": 52171624.0,
      "step": 28365
    },
    {
      "entropy": 4.984856176376343,
      "epoch": 2.369695957233545,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004435251190639923,
      "loss": 4.6636,
      "mean_token_accuracy": 0.23069707453250884,
      "num_tokens": 52181792.0,
      "step": 28370
    },
    {
      "entropy": 4.92995171546936,
      "epoch": 2.370113598396258,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044350539291282247,
      "loss": 4.6686,
      "mean_token_accuracy": 0.23400080054998398,
      "num_tokens": 52189814.0,
      "step": 28375
    },
    {
      "entropy": 5.115924930572509,
      "epoch": 2.370531239558971,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004434856638117629,
      "loss": 4.8573,
      "mean_token_accuracy": 0.21008004546165465,
      "num_tokens": 52198561.0,
      "step": 28380
    },
    {
      "entropy": 4.9809760570526125,
      "epoch": 2.370948880721684,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044346593176115904,
      "loss": 4.6772,
      "mean_token_accuracy": 0.22735024392604827,
      "num_tokens": 52207014.0,
      "step": 28385
    },
    {
      "entropy": 5.080011034011841,
      "epoch": 2.371366521884397,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044344619676135624,
      "loss": 4.7329,
      "mean_token_accuracy": 0.23034323900938034,
      "num_tokens": 52216884.0,
      "step": 28390
    },
    {
      "entropy": 5.116549253463745,
      "epoch": 2.37178416304711,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004434264588127,
      "loss": 4.8018,
      "mean_token_accuracy": 0.21990921199321747,
      "num_tokens": 52226850.0,
      "step": 28395
    },
    {
      "entropy": 5.000787162780762,
      "epoch": 2.3722018042098227,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044340671791553586,
      "loss": 4.7035,
      "mean_token_accuracy": 0.2264939084649086,
      "num_tokens": 52235998.0,
      "step": 28400
    },
    {
      "entropy": 5.095290327072144,
      "epoch": 2.372619445372536,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004433869740702094,
      "loss": 4.7498,
      "mean_token_accuracy": 0.21964357495307923,
      "num_tokens": 52245570.0,
      "step": 28405
    },
    {
      "entropy": 5.129317378997802,
      "epoch": 2.373037086535249,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004433672272770663,
      "loss": 4.8481,
      "mean_token_accuracy": 0.2119593322277069,
      "num_tokens": 52254613.0,
      "step": 28410
    },
    {
      "entropy": 5.060100078582764,
      "epoch": 2.373454727697962,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044334747753645224,
      "loss": 4.7533,
      "mean_token_accuracy": 0.2209625020623207,
      "num_tokens": 52264427.0,
      "step": 28415
    },
    {
      "entropy": 5.047259283065796,
      "epoch": 2.373872368860675,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004433277248487129,
      "loss": 4.6817,
      "mean_token_accuracy": 0.22742525786161422,
      "num_tokens": 52274673.0,
      "step": 28420
    },
    {
      "entropy": 5.129214334487915,
      "epoch": 2.3742900100233877,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004433079692141941,
      "loss": 4.888,
      "mean_token_accuracy": 0.2149788498878479,
      "num_tokens": 52283967.0,
      "step": 28425
    },
    {
      "entropy": 5.1341639995574955,
      "epoch": 2.374707651186101,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044328821063324177,
      "loss": 4.8363,
      "mean_token_accuracy": 0.22374726384878157,
      "num_tokens": 52293013.0,
      "step": 28430
    },
    {
      "entropy": 5.024908351898193,
      "epoch": 2.3751252923488138,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004432684491062017,
      "loss": 4.6559,
      "mean_token_accuracy": 0.23270078897476196,
      "num_tokens": 52301777.0,
      "step": 28435
    },
    {
      "entropy": 5.068070316314698,
      "epoch": 2.375542933511527,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044324868463341987,
      "loss": 4.7828,
      "mean_token_accuracy": 0.21980923116207124,
      "num_tokens": 52312366.0,
      "step": 28440
    },
    {
      "entropy": 5.063067245483398,
      "epoch": 2.37596057467424,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044322891721524233,
      "loss": 4.7544,
      "mean_token_accuracy": 0.22238309532403946,
      "num_tokens": 52321093.0,
      "step": 28445
    },
    {
      "entropy": 5.034194040298462,
      "epoch": 2.376378215836953,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044320914685201516,
      "loss": 4.6597,
      "mean_token_accuracy": 0.22571324110031127,
      "num_tokens": 52330166.0,
      "step": 28450
    },
    {
      "entropy": 5.046360778808594,
      "epoch": 2.376795856999666,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004431893735440843,
      "loss": 4.6947,
      "mean_token_accuracy": 0.22889635115861892,
      "num_tokens": 52339377.0,
      "step": 28455
    },
    {
      "entropy": 4.992163753509521,
      "epoch": 2.3772134981623787,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044316959729179603,
      "loss": 4.6955,
      "mean_token_accuracy": 0.22101196348667146,
      "num_tokens": 52349721.0,
      "step": 28460
    },
    {
      "entropy": 5.002919006347656,
      "epoch": 2.377631139325092,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044314981809549656,
      "loss": 4.7268,
      "mean_token_accuracy": 0.22921405285596846,
      "num_tokens": 52358332.0,
      "step": 28465
    },
    {
      "entropy": 5.048385238647461,
      "epoch": 2.3780487804878048,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004431300359555321,
      "loss": 4.7232,
      "mean_token_accuracy": 0.22688011676073075,
      "num_tokens": 52368000.0,
      "step": 28470
    },
    {
      "entropy": 5.086547565460205,
      "epoch": 2.378466421650518,
      "grad_norm": 1.09375,
      "learning_rate": 0.000443110250872249,
      "loss": 4.7746,
      "mean_token_accuracy": 0.22350021749734877,
      "num_tokens": 52376790.0,
      "step": 28475
    },
    {
      "entropy": 4.993002653121948,
      "epoch": 2.378884062813231,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004430904628459935,
      "loss": 4.648,
      "mean_token_accuracy": 0.2268161803483963,
      "num_tokens": 52385876.0,
      "step": 28480
    },
    {
      "entropy": 5.038763427734375,
      "epoch": 2.379301703975944,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044307067187711227,
      "loss": 4.7426,
      "mean_token_accuracy": 0.22848640233278275,
      "num_tokens": 52395350.0,
      "step": 28485
    },
    {
      "entropy": 4.886916875839233,
      "epoch": 2.379719345138657,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004430508779659516,
      "loss": 4.6478,
      "mean_token_accuracy": 0.23394013196229935,
      "num_tokens": 52404626.0,
      "step": 28490
    },
    {
      "entropy": 5.120145225524903,
      "epoch": 2.3801369863013697,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044303108111285795,
      "loss": 4.8619,
      "mean_token_accuracy": 0.21597510129213332,
      "num_tokens": 52414053.0,
      "step": 28495
    },
    {
      "entropy": 5.0525775909423825,
      "epoch": 2.380554627464083,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044301128131817805,
      "loss": 4.6601,
      "mean_token_accuracy": 0.23576609045267105,
      "num_tokens": 52422754.0,
      "step": 28500
    },
    {
      "entropy": 5.006632280349732,
      "epoch": 2.380972268626796,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044299147858225835,
      "loss": 4.735,
      "mean_token_accuracy": 0.21562550067901612,
      "num_tokens": 52431727.0,
      "step": 28505
    },
    {
      "entropy": 5.078313970565796,
      "epoch": 2.381389909789509,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004429716729054456,
      "loss": 4.681,
      "mean_token_accuracy": 0.2301594629883766,
      "num_tokens": 52440515.0,
      "step": 28510
    },
    {
      "entropy": 5.066996431350708,
      "epoch": 2.381807550952222,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004429518642880866,
      "loss": 4.7579,
      "mean_token_accuracy": 0.2218200072646141,
      "num_tokens": 52449091.0,
      "step": 28515
    },
    {
      "entropy": 5.048136949539185,
      "epoch": 2.3822251921149347,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044293205273052805,
      "loss": 4.725,
      "mean_token_accuracy": 0.2232707306742668,
      "num_tokens": 52457990.0,
      "step": 28520
    },
    {
      "entropy": 5.096951103210449,
      "epoch": 2.382642833277648,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004429122382331168,
      "loss": 4.7399,
      "mean_token_accuracy": 0.21771990656852722,
      "num_tokens": 52466954.0,
      "step": 28525
    },
    {
      "entropy": 5.0457910060882565,
      "epoch": 2.3830604744403607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044289242079619955,
      "loss": 4.7423,
      "mean_token_accuracy": 0.2278387188911438,
      "num_tokens": 52476757.0,
      "step": 28530
    },
    {
      "entropy": 4.97066740989685,
      "epoch": 2.383478115603074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044287260042012353,
      "loss": 4.6491,
      "mean_token_accuracy": 0.23299776315689086,
      "num_tokens": 52485776.0,
      "step": 28535
    },
    {
      "entropy": 4.9492607593536375,
      "epoch": 2.383895756765787,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004428527771052355,
      "loss": 4.6203,
      "mean_token_accuracy": 0.2334069326519966,
      "num_tokens": 52494267.0,
      "step": 28540
    },
    {
      "entropy": 5.133308982849121,
      "epoch": 2.3843133979284996,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044283295085188263,
      "loss": 4.8451,
      "mean_token_accuracy": 0.21299617886543273,
      "num_tokens": 52504163.0,
      "step": 28545
    },
    {
      "entropy": 5.04723219871521,
      "epoch": 2.384731039091213,
      "grad_norm": 1.125,
      "learning_rate": 0.00044281312166041196,
      "loss": 4.6344,
      "mean_token_accuracy": 0.23361170738935472,
      "num_tokens": 52513333.0,
      "step": 28550
    },
    {
      "entropy": 5.003926467895508,
      "epoch": 2.3851486802539257,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004427932895311705,
      "loss": 4.7006,
      "mean_token_accuracy": 0.22494760751724244,
      "num_tokens": 52522846.0,
      "step": 28555
    },
    {
      "entropy": 5.018396949768066,
      "epoch": 2.385566321416639,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004427734544645055,
      "loss": 4.7805,
      "mean_token_accuracy": 0.22147061228752135,
      "num_tokens": 52532130.0,
      "step": 28560
    },
    {
      "entropy": 5.079034900665283,
      "epoch": 2.3859839625793517,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044275361646076427,
      "loss": 4.7565,
      "mean_token_accuracy": 0.2170762002468109,
      "num_tokens": 52540952.0,
      "step": 28565
    },
    {
      "entropy": 5.036730623245239,
      "epoch": 2.386401603742065,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044273377552029415,
      "loss": 4.7323,
      "mean_token_accuracy": 0.21953799873590468,
      "num_tokens": 52551173.0,
      "step": 28570
    },
    {
      "entropy": 4.93903694152832,
      "epoch": 2.386819244904778,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004427139316434423,
      "loss": 4.6286,
      "mean_token_accuracy": 0.23544431924819947,
      "num_tokens": 52560535.0,
      "step": 28575
    },
    {
      "entropy": 4.959468412399292,
      "epoch": 2.3872368860674906,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004426940848305563,
      "loss": 4.5632,
      "mean_token_accuracy": 0.23203002512454987,
      "num_tokens": 52569643.0,
      "step": 28580
    },
    {
      "entropy": 4.993225765228272,
      "epoch": 2.387654527230204,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004426742350819834,
      "loss": 4.6758,
      "mean_token_accuracy": 0.2293240249156952,
      "num_tokens": 52578547.0,
      "step": 28585
    },
    {
      "entropy": 4.97934308052063,
      "epoch": 2.3880721683929167,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044265438239807117,
      "loss": 4.6957,
      "mean_token_accuracy": 0.2248530000448227,
      "num_tokens": 52588111.0,
      "step": 28590
    },
    {
      "entropy": 5.012687158584595,
      "epoch": 2.38848980955563,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004426345267791672,
      "loss": 4.7349,
      "mean_token_accuracy": 0.2185040682554245,
      "num_tokens": 52596755.0,
      "step": 28595
    },
    {
      "entropy": 5.138187599182129,
      "epoch": 2.3889074507183428,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044261466822561907,
      "loss": 4.8018,
      "mean_token_accuracy": 0.21560253649950029,
      "num_tokens": 52607159.0,
      "step": 28600
    },
    {
      "entropy": 5.049730014801026,
      "epoch": 2.389325091881056,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004425948067377744,
      "loss": 4.7698,
      "mean_token_accuracy": 0.2205239787697792,
      "num_tokens": 52617797.0,
      "step": 28605
    },
    {
      "entropy": 5.001480627059936,
      "epoch": 2.389742733043769,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044257494231598094,
      "loss": 4.6747,
      "mean_token_accuracy": 0.23655619323253632,
      "num_tokens": 52626910.0,
      "step": 28610
    },
    {
      "entropy": 4.9932637214660645,
      "epoch": 2.3901603742064816,
      "grad_norm": 1.125,
      "learning_rate": 0.0004425550749605863,
      "loss": 4.6938,
      "mean_token_accuracy": 0.22449645400047302,
      "num_tokens": 52635730.0,
      "step": 28615
    },
    {
      "entropy": 4.999620962142944,
      "epoch": 2.390578015369195,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044253520467193855,
      "loss": 4.654,
      "mean_token_accuracy": 0.22251056432723998,
      "num_tokens": 52644723.0,
      "step": 28620
    },
    {
      "entropy": 5.036747932434082,
      "epoch": 2.3909956565319077,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044251533145038525,
      "loss": 4.7779,
      "mean_token_accuracy": 0.21858801394701005,
      "num_tokens": 52654407.0,
      "step": 28625
    },
    {
      "entropy": 5.02938060760498,
      "epoch": 2.391413297694621,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004424954552962744,
      "loss": 4.7322,
      "mean_token_accuracy": 0.22383615970611573,
      "num_tokens": 52663347.0,
      "step": 28630
    },
    {
      "entropy": 5.052218341827393,
      "epoch": 2.3918309388573338,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044247557620995416,
      "loss": 4.623,
      "mean_token_accuracy": 0.23397864997386933,
      "num_tokens": 52672449.0,
      "step": 28635
    },
    {
      "entropy": 5.039639139175415,
      "epoch": 2.3922485800200466,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004424556941917723,
      "loss": 4.692,
      "mean_token_accuracy": 0.22515685707330704,
      "num_tokens": 52681077.0,
      "step": 28640
    },
    {
      "entropy": 4.9245383739471436,
      "epoch": 2.39266622118276,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044243580924207686,
      "loss": 4.6837,
      "mean_token_accuracy": 0.22816684246063232,
      "num_tokens": 52690499.0,
      "step": 28645
    },
    {
      "entropy": 5.025797700881958,
      "epoch": 2.3930838623454727,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004424159213612161,
      "loss": 4.6705,
      "mean_token_accuracy": 0.215700101852417,
      "num_tokens": 52700330.0,
      "step": 28650
    },
    {
      "entropy": 4.982071304321289,
      "epoch": 2.393501503508186,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004423960305495381,
      "loss": 4.6581,
      "mean_token_accuracy": 0.2328794151544571,
      "num_tokens": 52709738.0,
      "step": 28655
    },
    {
      "entropy": 5.100012731552124,
      "epoch": 2.3939191446708987,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004423761368073911,
      "loss": 4.7839,
      "mean_token_accuracy": 0.2231995642185211,
      "num_tokens": 52719002.0,
      "step": 28660
    },
    {
      "entropy": 5.045068550109863,
      "epoch": 2.3943367858336115,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004423562401351234,
      "loss": 4.6599,
      "mean_token_accuracy": 0.22973365038633348,
      "num_tokens": 52727160.0,
      "step": 28665
    },
    {
      "entropy": 5.012835502624512,
      "epoch": 2.394754426996325,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004423363405330833,
      "loss": 4.6695,
      "mean_token_accuracy": 0.2344977855682373,
      "num_tokens": 52735839.0,
      "step": 28670
    },
    {
      "entropy": 4.952998542785645,
      "epoch": 2.3951720681590376,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044231643800161897,
      "loss": 4.5748,
      "mean_token_accuracy": 0.2313125893473625,
      "num_tokens": 52744388.0,
      "step": 28675
    },
    {
      "entropy": 4.976932668685913,
      "epoch": 2.395589709321751,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004422965325410791,
      "loss": 4.6953,
      "mean_token_accuracy": 0.23279836922883987,
      "num_tokens": 52754560.0,
      "step": 28680
    },
    {
      "entropy": 5.07656888961792,
      "epoch": 2.3960073504844637,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044227662415181207,
      "loss": 4.7864,
      "mean_token_accuracy": 0.21723434180021287,
      "num_tokens": 52764164.0,
      "step": 28685
    },
    {
      "entropy": 4.976916885375976,
      "epoch": 2.396424991647177,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044225671283416634,
      "loss": 4.6055,
      "mean_token_accuracy": 0.2398191660642624,
      "num_tokens": 52773352.0,
      "step": 28690
    },
    {
      "entropy": 5.111955547332764,
      "epoch": 2.3968426328098897,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044223679858849057,
      "loss": 4.8016,
      "mean_token_accuracy": 0.216179496049881,
      "num_tokens": 52782399.0,
      "step": 28695
    },
    {
      "entropy": 5.0433755874633786,
      "epoch": 2.3972602739726026,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004422168814151333,
      "loss": 4.7211,
      "mean_token_accuracy": 0.22101287692785263,
      "num_tokens": 52791029.0,
      "step": 28700
    },
    {
      "entropy": 5.0393617153167725,
      "epoch": 2.397677915135316,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044219696131444326,
      "loss": 4.7413,
      "mean_token_accuracy": 0.22128960490226746,
      "num_tokens": 52800368.0,
      "step": 28705
    },
    {
      "entropy": 5.045047807693481,
      "epoch": 2.3980955562980286,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044217703828676913,
      "loss": 4.7335,
      "mean_token_accuracy": 0.22817093580961229,
      "num_tokens": 52809820.0,
      "step": 28710
    },
    {
      "entropy": 5.029381895065308,
      "epoch": 2.398513197460742,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044215711233245973,
      "loss": 4.6621,
      "mean_token_accuracy": 0.22403862625360488,
      "num_tokens": 52819442.0,
      "step": 28715
    },
    {
      "entropy": 5.008009195327759,
      "epoch": 2.3989308386234547,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044213718345186394,
      "loss": 4.7842,
      "mean_token_accuracy": 0.21725236475467682,
      "num_tokens": 52828852.0,
      "step": 28720
    },
    {
      "entropy": 5.017750072479248,
      "epoch": 2.399348479786168,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044211725164533045,
      "loss": 4.6876,
      "mean_token_accuracy": 0.22928241044282913,
      "num_tokens": 52837295.0,
      "step": 28725
    },
    {
      "entropy": 5.032516145706177,
      "epoch": 2.3997661209488808,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044209731691320837,
      "loss": 4.6411,
      "mean_token_accuracy": 0.22482388466596603,
      "num_tokens": 52846347.0,
      "step": 28730
    },
    {
      "entropy": 5.165672397613525,
      "epoch": 2.4001837621115936,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004420773792558467,
      "loss": 4.8239,
      "mean_token_accuracy": 0.22054870426654816,
      "num_tokens": 52855191.0,
      "step": 28735
    },
    {
      "entropy": 5.066412496566772,
      "epoch": 2.400601403274307,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004420574386735943,
      "loss": 4.7102,
      "mean_token_accuracy": 0.22818684726953506,
      "num_tokens": 52865231.0,
      "step": 28740
    },
    {
      "entropy": 4.97991418838501,
      "epoch": 2.4010190444370196,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044203749516680035,
      "loss": 4.683,
      "mean_token_accuracy": 0.2227127209305763,
      "num_tokens": 52874065.0,
      "step": 28745
    },
    {
      "entropy": 4.987019157409668,
      "epoch": 2.401436685599733,
      "grad_norm": 1.125,
      "learning_rate": 0.00044201754873581396,
      "loss": 4.5971,
      "mean_token_accuracy": 0.23226651251316072,
      "num_tokens": 52882740.0,
      "step": 28750
    },
    {
      "entropy": 4.928406667709351,
      "epoch": 2.4018543267624457,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044199759938098437,
      "loss": 4.6108,
      "mean_token_accuracy": 0.22815473824739457,
      "num_tokens": 52891011.0,
      "step": 28755
    },
    {
      "entropy": 5.165027761459351,
      "epoch": 2.4022719679251585,
      "grad_norm": 1.125,
      "learning_rate": 0.00044197764710266087,
      "loss": 4.8542,
      "mean_token_accuracy": 0.22195156365633012,
      "num_tokens": 52900364.0,
      "step": 28760
    },
    {
      "entropy": 5.125201654434204,
      "epoch": 2.4026896090878718,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004419576919011926,
      "loss": 4.8055,
      "mean_token_accuracy": 0.21753200739622117,
      "num_tokens": 52910155.0,
      "step": 28765
    },
    {
      "entropy": 4.9985230445861815,
      "epoch": 2.4031072502505846,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044193773377692886,
      "loss": 4.7145,
      "mean_token_accuracy": 0.2219867929816246,
      "num_tokens": 52919134.0,
      "step": 28770
    },
    {
      "entropy": 5.068641519546508,
      "epoch": 2.403524891413298,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044191777273021933,
      "loss": 4.8077,
      "mean_token_accuracy": 0.22532174736261368,
      "num_tokens": 52927293.0,
      "step": 28775
    },
    {
      "entropy": 5.038974952697754,
      "epoch": 2.4039425325760106,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004418978087614131,
      "loss": 4.734,
      "mean_token_accuracy": 0.22160226553678514,
      "num_tokens": 52936144.0,
      "step": 28780
    },
    {
      "entropy": 5.1277244091033936,
      "epoch": 2.4043601737387235,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044187784187085994,
      "loss": 4.7802,
      "mean_token_accuracy": 0.218555049598217,
      "num_tokens": 52945666.0,
      "step": 28785
    },
    {
      "entropy": 4.96799259185791,
      "epoch": 2.4047778149014367,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004418578720589093,
      "loss": 4.6457,
      "mean_token_accuracy": 0.2324896812438965,
      "num_tokens": 52954301.0,
      "step": 28790
    },
    {
      "entropy": 4.935872888565063,
      "epoch": 2.4051954560641495,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004418378993259107,
      "loss": 4.6134,
      "mean_token_accuracy": 0.23519655764102937,
      "num_tokens": 52962794.0,
      "step": 28795
    },
    {
      "entropy": 4.956109046936035,
      "epoch": 2.405613097226863,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004418179236722139,
      "loss": 4.5631,
      "mean_token_accuracy": 0.23968268930912018,
      "num_tokens": 52970927.0,
      "step": 28800
    },
    {
      "entropy": 5.055432271957398,
      "epoch": 2.4060307383895756,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004417979450981685,
      "loss": 4.6919,
      "mean_token_accuracy": 0.2295960783958435,
      "num_tokens": 52979638.0,
      "step": 28805
    },
    {
      "entropy": 5.001510810852051,
      "epoch": 2.406448379552289,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044177796360412436,
      "loss": 4.6801,
      "mean_token_accuracy": 0.2227442443370819,
      "num_tokens": 52989268.0,
      "step": 28810
    },
    {
      "entropy": 5.026500511169433,
      "epoch": 2.4068660207150017,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004417579791904312,
      "loss": 4.7085,
      "mean_token_accuracy": 0.22356806844472885,
      "num_tokens": 52997858.0,
      "step": 28815
    },
    {
      "entropy": 5.068287086486817,
      "epoch": 2.4072836618777145,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004417379918574388,
      "loss": 4.678,
      "mean_token_accuracy": 0.22840783894062042,
      "num_tokens": 53006830.0,
      "step": 28820
    },
    {
      "entropy": 5.03721513748169,
      "epoch": 2.4077013030404277,
      "grad_norm": 1.125,
      "learning_rate": 0.0004417180016054972,
      "loss": 4.7452,
      "mean_token_accuracy": 0.22189656645059586,
      "num_tokens": 53016202.0,
      "step": 28825
    },
    {
      "entropy": 5.045368289947509,
      "epoch": 2.4081189442031405,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044169800843495637,
      "loss": 4.7866,
      "mean_token_accuracy": 0.21660268306732178,
      "num_tokens": 53025441.0,
      "step": 28830
    },
    {
      "entropy": 5.093586063385009,
      "epoch": 2.408536585365854,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004416780123461662,
      "loss": 4.8609,
      "mean_token_accuracy": 0.21008424907922746,
      "num_tokens": 53036072.0,
      "step": 28835
    },
    {
      "entropy": 5.021963691711425,
      "epoch": 2.4089542265285666,
      "grad_norm": 1.125,
      "learning_rate": 0.00044165801333947676,
      "loss": 4.7491,
      "mean_token_accuracy": 0.23254174292087554,
      "num_tokens": 53046272.0,
      "step": 28840
    },
    {
      "entropy": 5.057261943817139,
      "epoch": 2.40937186769128,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004416380114152382,
      "loss": 4.6602,
      "mean_token_accuracy": 0.22078338861465455,
      "num_tokens": 53054944.0,
      "step": 28845
    },
    {
      "entropy": 4.892693424224854,
      "epoch": 2.4097895088539927,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044161800657380076,
      "loss": 4.6113,
      "mean_token_accuracy": 0.23253184407949448,
      "num_tokens": 53063770.0,
      "step": 28850
    },
    {
      "entropy": 5.022906160354614,
      "epoch": 2.4102071500167055,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004415979988155145,
      "loss": 4.6728,
      "mean_token_accuracy": 0.23243079334497452,
      "num_tokens": 53073413.0,
      "step": 28855
    },
    {
      "entropy": 5.045170259475708,
      "epoch": 2.4106247911794187,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004415779881407297,
      "loss": 4.7395,
      "mean_token_accuracy": 0.22051485031843185,
      "num_tokens": 53082908.0,
      "step": 28860
    },
    {
      "entropy": 5.050991106033325,
      "epoch": 2.4110424323421316,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004415579745497967,
      "loss": 4.7516,
      "mean_token_accuracy": 0.22463317811489106,
      "num_tokens": 53092549.0,
      "step": 28865
    },
    {
      "entropy": 5.000395488739014,
      "epoch": 2.411460073504845,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004415379580430659,
      "loss": 4.6937,
      "mean_token_accuracy": 0.22574810385704042,
      "num_tokens": 53102334.0,
      "step": 28870
    },
    {
      "entropy": 5.027732419967651,
      "epoch": 2.4118777146675576,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044151793862088766,
      "loss": 4.7789,
      "mean_token_accuracy": 0.22060059607028962,
      "num_tokens": 53112603.0,
      "step": 28875
    },
    {
      "entropy": 5.1395463943481445,
      "epoch": 2.4122953558302704,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004414979162836125,
      "loss": 4.8059,
      "mean_token_accuracy": 0.22079577147960663,
      "num_tokens": 53122020.0,
      "step": 28880
    },
    {
      "entropy": 5.051037931442261,
      "epoch": 2.4127129969929837,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004414778910315909,
      "loss": 4.6719,
      "mean_token_accuracy": 0.2244867354631424,
      "num_tokens": 53130776.0,
      "step": 28885
    },
    {
      "entropy": 4.946846199035645,
      "epoch": 2.4131306381556965,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044145786286517343,
      "loss": 4.7052,
      "mean_token_accuracy": 0.22868459969758986,
      "num_tokens": 53140007.0,
      "step": 28890
    },
    {
      "entropy": 5.1029175281524655,
      "epoch": 2.4135482793184098,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004414378317847107,
      "loss": 4.8031,
      "mean_token_accuracy": 0.22119836807250975,
      "num_tokens": 53149218.0,
      "step": 28895
    },
    {
      "entropy": 5.037054586410522,
      "epoch": 2.4139659204811226,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044141779779055334,
      "loss": 4.6663,
      "mean_token_accuracy": 0.22930338978767395,
      "num_tokens": 53158259.0,
      "step": 28900
    },
    {
      "entropy": 4.962104988098145,
      "epoch": 2.4143835616438354,
      "grad_norm": 1.125,
      "learning_rate": 0.00044139776088305215,
      "loss": 4.7043,
      "mean_token_accuracy": 0.22880886048078536,
      "num_tokens": 53167131.0,
      "step": 28905
    },
    {
      "entropy": 5.015885686874389,
      "epoch": 2.4148012028065486,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044137772106255793,
      "loss": 4.6552,
      "mean_token_accuracy": 0.22340963631868363,
      "num_tokens": 53176171.0,
      "step": 28910
    },
    {
      "entropy": 5.034735631942749,
      "epoch": 2.4152188439692615,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044135767832942136,
      "loss": 4.7244,
      "mean_token_accuracy": 0.22452976405620576,
      "num_tokens": 53185770.0,
      "step": 28915
    },
    {
      "entropy": 4.979569768905639,
      "epoch": 2.4156364851319747,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004413376326839934,
      "loss": 4.515,
      "mean_token_accuracy": 0.2362914577126503,
      "num_tokens": 53194085.0,
      "step": 28920
    },
    {
      "entropy": 4.988331604003906,
      "epoch": 2.4160541262946875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044131758412662493,
      "loss": 4.7368,
      "mean_token_accuracy": 0.2218383863568306,
      "num_tokens": 53202978.0,
      "step": 28925
    },
    {
      "entropy": 4.960284090042114,
      "epoch": 2.4164717674574008,
      "grad_norm": 1.21875,
      "learning_rate": 0.000441297532657667,
      "loss": 4.5967,
      "mean_token_accuracy": 0.22979626953601837,
      "num_tokens": 53212202.0,
      "step": 28930
    },
    {
      "entropy": 5.028318881988525,
      "epoch": 2.4168894086201136,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004412774782774705,
      "loss": 4.6772,
      "mean_token_accuracy": 0.23099337965250016,
      "num_tokens": 53220893.0,
      "step": 28935
    },
    {
      "entropy": 5.027711391448975,
      "epoch": 2.417307049782827,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004412574209863867,
      "loss": 4.6481,
      "mean_token_accuracy": 0.2279279425740242,
      "num_tokens": 53229253.0,
      "step": 28940
    },
    {
      "entropy": 4.940762567520141,
      "epoch": 2.4177246909455397,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004412373607847666,
      "loss": 4.6334,
      "mean_token_accuracy": 0.23169131726026534,
      "num_tokens": 53238507.0,
      "step": 28945
    },
    {
      "entropy": 5.03739366531372,
      "epoch": 2.4181423321082525,
      "grad_norm": 1.125,
      "learning_rate": 0.00044121729767296135,
      "loss": 4.74,
      "mean_token_accuracy": 0.22726851105690002,
      "num_tokens": 53247109.0,
      "step": 28950
    },
    {
      "entropy": 5.096235132217407,
      "epoch": 2.4185599732709657,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044119723165132225,
      "loss": 4.7137,
      "mean_token_accuracy": 0.21960866898298265,
      "num_tokens": 53255933.0,
      "step": 28955
    },
    {
      "entropy": 5.0256805419921875,
      "epoch": 2.4189776144336785,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044117716272020055,
      "loss": 4.6493,
      "mean_token_accuracy": 0.2289960965514183,
      "num_tokens": 53264378.0,
      "step": 28960
    },
    {
      "entropy": 4.981594753265381,
      "epoch": 2.419395255596392,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004411570908799476,
      "loss": 4.7003,
      "mean_token_accuracy": 0.22637448459863663,
      "num_tokens": 53273805.0,
      "step": 28965
    },
    {
      "entropy": 4.964417457580566,
      "epoch": 2.4198128967591046,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044113701613091483,
      "loss": 4.6466,
      "mean_token_accuracy": 0.2214195489883423,
      "num_tokens": 53282381.0,
      "step": 28970
    },
    {
      "entropy": 5.00276746749878,
      "epoch": 2.4202305379218174,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044111693847345354,
      "loss": 4.7754,
      "mean_token_accuracy": 0.2173010915517807,
      "num_tokens": 53291400.0,
      "step": 28975
    },
    {
      "entropy": 5.113553524017334,
      "epoch": 2.4206481790845307,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004410968579079153,
      "loss": 4.8164,
      "mean_token_accuracy": 0.21713705807924272,
      "num_tokens": 53300825.0,
      "step": 28980
    },
    {
      "entropy": 5.010876846313477,
      "epoch": 2.4210658202472435,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044107677443465165,
      "loss": 4.6441,
      "mean_token_accuracy": 0.22852799892425538,
      "num_tokens": 53309932.0,
      "step": 28985
    },
    {
      "entropy": 5.125916528701782,
      "epoch": 2.4214834614099567,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044105668805401415,
      "loss": 4.852,
      "mean_token_accuracy": 0.21337681710720063,
      "num_tokens": 53318941.0,
      "step": 28990
    },
    {
      "entropy": 5.122243165969849,
      "epoch": 2.4219011025726696,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044103659876635443,
      "loss": 4.8088,
      "mean_token_accuracy": 0.22333623021841048,
      "num_tokens": 53328006.0,
      "step": 28995
    },
    {
      "entropy": 4.995333194732666,
      "epoch": 2.4223187437353824,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004410165065720242,
      "loss": 4.6158,
      "mean_token_accuracy": 0.23472701460123063,
      "num_tokens": 53337024.0,
      "step": 29000
    },
    {
      "entropy": 5.014283895492554,
      "epoch": 2.4227363848980956,
      "grad_norm": 1.125,
      "learning_rate": 0.0004409964114713752,
      "loss": 4.6857,
      "mean_token_accuracy": 0.23195817470550537,
      "num_tokens": 53346295.0,
      "step": 29005
    },
    {
      "entropy": 5.046065998077393,
      "epoch": 2.4231540260608084,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004409763134647592,
      "loss": 4.7423,
      "mean_token_accuracy": 0.22628345042467118,
      "num_tokens": 53356184.0,
      "step": 29010
    },
    {
      "entropy": 4.964841508865357,
      "epoch": 2.4235716672235217,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000440956212552528,
      "loss": 4.5232,
      "mean_token_accuracy": 0.24238962084054946,
      "num_tokens": 53365001.0,
      "step": 29015
    },
    {
      "entropy": 5.057470846176147,
      "epoch": 2.4239893083862345,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004409361087350336,
      "loss": 4.7774,
      "mean_token_accuracy": 0.21825411468744277,
      "num_tokens": 53374643.0,
      "step": 29020
    },
    {
      "entropy": 4.948200702667236,
      "epoch": 2.4244069495489473,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004409160020126279,
      "loss": 4.6696,
      "mean_token_accuracy": 0.2285086065530777,
      "num_tokens": 53382878.0,
      "step": 29025
    },
    {
      "entropy": 5.114639472961426,
      "epoch": 2.4248245907116606,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004408958923856628,
      "loss": 4.8534,
      "mean_token_accuracy": 0.21669959872961045,
      "num_tokens": 53393290.0,
      "step": 29030
    },
    {
      "entropy": 5.176379919052124,
      "epoch": 2.4252422318743734,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044087577985449047,
      "loss": 4.8486,
      "mean_token_accuracy": 0.2141724035143852,
      "num_tokens": 53403453.0,
      "step": 29035
    },
    {
      "entropy": 5.036980533599854,
      "epoch": 2.4256598730370866,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004408556644194629,
      "loss": 4.7717,
      "mean_token_accuracy": 0.219194994866848,
      "num_tokens": 53413519.0,
      "step": 29040
    },
    {
      "entropy": 5.040273904800415,
      "epoch": 2.4260775141997994,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004408355460809323,
      "loss": 4.723,
      "mean_token_accuracy": 0.22030321955680848,
      "num_tokens": 53423123.0,
      "step": 29045
    },
    {
      "entropy": 4.981621217727661,
      "epoch": 2.4264951553625127,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044081542483925086,
      "loss": 4.7619,
      "mean_token_accuracy": 0.21811582148075104,
      "num_tokens": 53432243.0,
      "step": 29050
    },
    {
      "entropy": 5.053811168670654,
      "epoch": 2.4269127965252255,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004407953006947708,
      "loss": 4.6969,
      "mean_token_accuracy": 0.2272563859820366,
      "num_tokens": 53441679.0,
      "step": 29055
    },
    {
      "entropy": 5.010194635391235,
      "epoch": 2.4273304376879388,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044077517364784455,
      "loss": 4.7041,
      "mean_token_accuracy": 0.23072677701711655,
      "num_tokens": 53450674.0,
      "step": 29060
    },
    {
      "entropy": 5.003877735137939,
      "epoch": 2.4277480788506516,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044075504369882416,
      "loss": 4.6698,
      "mean_token_accuracy": 0.2292722135782242,
      "num_tokens": 53459075.0,
      "step": 29065
    },
    {
      "entropy": 5.055096006393432,
      "epoch": 2.4281657200133644,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004407349108480623,
      "loss": 4.7137,
      "mean_token_accuracy": 0.22611218988895415,
      "num_tokens": 53468135.0,
      "step": 29070
    },
    {
      "entropy": 4.999924516677856,
      "epoch": 2.4285833611760776,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044071477509591126,
      "loss": 4.7518,
      "mean_token_accuracy": 0.22420028299093248,
      "num_tokens": 53477777.0,
      "step": 29075
    },
    {
      "entropy": 5.054559707641602,
      "epoch": 2.4290010023387905,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004406946364427237,
      "loss": 4.6833,
      "mean_token_accuracy": 0.22374757677316665,
      "num_tokens": 53487126.0,
      "step": 29080
    },
    {
      "entropy": 4.949953365325928,
      "epoch": 2.4294186435015037,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044067449488885207,
      "loss": 4.6654,
      "mean_token_accuracy": 0.2300649642944336,
      "num_tokens": 53497304.0,
      "step": 29085
    },
    {
      "entropy": 4.9375440120697025,
      "epoch": 2.4298362846642165,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044065435043464903,
      "loss": 4.7131,
      "mean_token_accuracy": 0.22662238478660585,
      "num_tokens": 53505842.0,
      "step": 29090
    },
    {
      "entropy": 4.966595935821533,
      "epoch": 2.4302539258269293,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004406342030804671,
      "loss": 4.6428,
      "mean_token_accuracy": 0.2261927008628845,
      "num_tokens": 53514504.0,
      "step": 29095
    },
    {
      "entropy": 4.9929869174957275,
      "epoch": 2.4306715669896426,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004406140528266591,
      "loss": 4.6564,
      "mean_token_accuracy": 0.22671499997377395,
      "num_tokens": 53523691.0,
      "step": 29100
    },
    {
      "entropy": 5.012039756774902,
      "epoch": 2.4310892081523554,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044059389967357774,
      "loss": 4.7227,
      "mean_token_accuracy": 0.22077531963586808,
      "num_tokens": 53532402.0,
      "step": 29105
    },
    {
      "entropy": 4.946609687805176,
      "epoch": 2.4315068493150687,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044057374362157587,
      "loss": 4.6365,
      "mean_token_accuracy": 0.23440106809139252,
      "num_tokens": 53541085.0,
      "step": 29110
    },
    {
      "entropy": 4.945417976379394,
      "epoch": 2.4319244904777815,
      "grad_norm": 1.125,
      "learning_rate": 0.00044055358467100627,
      "loss": 4.5879,
      "mean_token_accuracy": 0.23087803274393082,
      "num_tokens": 53550422.0,
      "step": 29115
    },
    {
      "entropy": 4.9597900867462155,
      "epoch": 2.4323421316404943,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044053342282222196,
      "loss": 4.6656,
      "mean_token_accuracy": 0.23419995307922364,
      "num_tokens": 53560238.0,
      "step": 29120
    },
    {
      "entropy": 5.129540920257568,
      "epoch": 2.4327597728032075,
      "grad_norm": 1.0,
      "learning_rate": 0.0004405132580755758,
      "loss": 4.8181,
      "mean_token_accuracy": 0.214924593269825,
      "num_tokens": 53571467.0,
      "step": 29125
    },
    {
      "entropy": 5.1008693218231205,
      "epoch": 2.4331774139659204,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044049309043142084,
      "loss": 4.6445,
      "mean_token_accuracy": 0.2157386288046837,
      "num_tokens": 53580443.0,
      "step": 29130
    },
    {
      "entropy": 5.042001724243164,
      "epoch": 2.4335950551286336,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004404729198901101,
      "loss": 4.8104,
      "mean_token_accuracy": 0.21891495436429978,
      "num_tokens": 53589957.0,
      "step": 29135
    },
    {
      "entropy": 5.060759496688843,
      "epoch": 2.4340126962913464,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044045274645199665,
      "loss": 4.8332,
      "mean_token_accuracy": 0.21082940697669983,
      "num_tokens": 53599616.0,
      "step": 29140
    },
    {
      "entropy": 4.95823040008545,
      "epoch": 2.4344303374540592,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004404325701174338,
      "loss": 4.6326,
      "mean_token_accuracy": 0.22292779982089997,
      "num_tokens": 53609872.0,
      "step": 29145
    },
    {
      "entropy": 4.966677856445313,
      "epoch": 2.4348479786167725,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004404123908867747,
      "loss": 4.6768,
      "mean_token_accuracy": 0.23112272024154662,
      "num_tokens": 53618713.0,
      "step": 29150
    },
    {
      "entropy": 5.038806581497193,
      "epoch": 2.4352656197794853,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044039220876037255,
      "loss": 4.6427,
      "mean_token_accuracy": 0.23449686020612717,
      "num_tokens": 53627782.0,
      "step": 29155
    },
    {
      "entropy": 5.021997737884521,
      "epoch": 2.4356832609421986,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004403720237385808,
      "loss": 4.6629,
      "mean_token_accuracy": 0.23171934485435486,
      "num_tokens": 53637012.0,
      "step": 29160
    },
    {
      "entropy": 4.964478158950806,
      "epoch": 2.4361009021049114,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044035183582175264,
      "loss": 4.724,
      "mean_token_accuracy": 0.22383038997650145,
      "num_tokens": 53646145.0,
      "step": 29165
    },
    {
      "entropy": 5.081682443618774,
      "epoch": 2.4365185432676246,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044033164501024153,
      "loss": 4.8068,
      "mean_token_accuracy": 0.21758115738630296,
      "num_tokens": 53655283.0,
      "step": 29170
    },
    {
      "entropy": 5.055565309524536,
      "epoch": 2.4369361844303374,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000440311451304401,
      "loss": 4.6873,
      "mean_token_accuracy": 0.23130788654088974,
      "num_tokens": 53664786.0,
      "step": 29175
    },
    {
      "entropy": 4.9739748477935795,
      "epoch": 2.4373538255930507,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044029125470458445,
      "loss": 4.5933,
      "mean_token_accuracy": 0.23299514800310134,
      "num_tokens": 53673566.0,
      "step": 29180
    },
    {
      "entropy": 5.0599370956420895,
      "epoch": 2.4377714667557635,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004402710552111456,
      "loss": 4.7812,
      "mean_token_accuracy": 0.2210545212030411,
      "num_tokens": 53683772.0,
      "step": 29185
    },
    {
      "entropy": 5.018598937988282,
      "epoch": 2.4381891079184763,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044025085282443807,
      "loss": 4.6946,
      "mean_token_accuracy": 0.22380512654781343,
      "num_tokens": 53692202.0,
      "step": 29190
    },
    {
      "entropy": 5.101478433609008,
      "epoch": 2.4386067490811896,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044023064754481535,
      "loss": 4.7195,
      "mean_token_accuracy": 0.22506927251815795,
      "num_tokens": 53702098.0,
      "step": 29195
    },
    {
      "entropy": 4.950622129440307,
      "epoch": 2.4390243902439024,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004402104393726313,
      "loss": 4.6459,
      "mean_token_accuracy": 0.22444025427103043,
      "num_tokens": 53711069.0,
      "step": 29200
    },
    {
      "entropy": 5.093327617645263,
      "epoch": 2.4394420314066156,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004401902283082396,
      "loss": 4.7658,
      "mean_token_accuracy": 0.22485154569149018,
      "num_tokens": 53720335.0,
      "step": 29205
    },
    {
      "entropy": 5.0368678092956545,
      "epoch": 2.4398596725693285,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044017001435199413,
      "loss": 4.7233,
      "mean_token_accuracy": 0.22020622789859773,
      "num_tokens": 53729401.0,
      "step": 29210
    },
    {
      "entropy": 5.087573003768921,
      "epoch": 2.4402773137320413,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004401497975042487,
      "loss": 4.755,
      "mean_token_accuracy": 0.22576136887073517,
      "num_tokens": 53738642.0,
      "step": 29215
    },
    {
      "entropy": 4.957791948318482,
      "epoch": 2.4406949548947545,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044012957776535735,
      "loss": 4.6483,
      "mean_token_accuracy": 0.22726439982652663,
      "num_tokens": 53747316.0,
      "step": 29220
    },
    {
      "entropy": 5.019256162643432,
      "epoch": 2.4411125960574673,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004401093551356739,
      "loss": 4.6396,
      "mean_token_accuracy": 0.22941708266735078,
      "num_tokens": 53756914.0,
      "step": 29225
    },
    {
      "entropy": 4.837207174301147,
      "epoch": 2.4415302372201806,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044008912961555246,
      "loss": 4.5126,
      "mean_token_accuracy": 0.24281605780124665,
      "num_tokens": 53765740.0,
      "step": 29230
    },
    {
      "entropy": 4.980676746368408,
      "epoch": 2.4419478783828934,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004400689012053472,
      "loss": 4.7336,
      "mean_token_accuracy": 0.22211433798074723,
      "num_tokens": 53775616.0,
      "step": 29235
    },
    {
      "entropy": 5.0652648448944095,
      "epoch": 2.442365519545606,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000440048669905412,
      "loss": 4.7666,
      "mean_token_accuracy": 0.22099242359399796,
      "num_tokens": 53785063.0,
      "step": 29240
    },
    {
      "entropy": 4.950164079666138,
      "epoch": 2.4427831607083195,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044002843571610124,
      "loss": 4.5773,
      "mean_token_accuracy": 0.23543634861707688,
      "num_tokens": 53794190.0,
      "step": 29245
    },
    {
      "entropy": 4.982102918624878,
      "epoch": 2.4432008018710323,
      "grad_norm": 1.09375,
      "learning_rate": 0.000440008198637769,
      "loss": 4.665,
      "mean_token_accuracy": 0.2310412898659706,
      "num_tokens": 53803121.0,
      "step": 29250
    },
    {
      "entropy": 4.911992454528809,
      "epoch": 2.4436184430337455,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004399879586707697,
      "loss": 4.5702,
      "mean_token_accuracy": 0.23437263071537018,
      "num_tokens": 53812538.0,
      "step": 29255
    },
    {
      "entropy": 5.076948833465576,
      "epoch": 2.4440360841964583,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004399677158154576,
      "loss": 4.7281,
      "mean_token_accuracy": 0.22211321592330932,
      "num_tokens": 53821574.0,
      "step": 29260
    },
    {
      "entropy": 5.040156984329224,
      "epoch": 2.444453725359171,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000439947470072187,
      "loss": 4.7005,
      "mean_token_accuracy": 0.2214355304837227,
      "num_tokens": 53829996.0,
      "step": 29265
    },
    {
      "entropy": 5.072239685058594,
      "epoch": 2.4448713665218844,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004399272214413124,
      "loss": 4.7644,
      "mean_token_accuracy": 0.2284940242767334,
      "num_tokens": 53838708.0,
      "step": 29270
    },
    {
      "entropy": 5.045899868011475,
      "epoch": 2.4452890076845972,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004399069699231883,
      "loss": 4.8145,
      "mean_token_accuracy": 0.2128199815750122,
      "num_tokens": 53849354.0,
      "step": 29275
    },
    {
      "entropy": 5.042806577682495,
      "epoch": 2.4457066488473105,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043988671551816914,
      "loss": 4.7066,
      "mean_token_accuracy": 0.22346633821725845,
      "num_tokens": 53858286.0,
      "step": 29280
    },
    {
      "entropy": 4.989195919036865,
      "epoch": 2.4461242900100233,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043986645822660956,
      "loss": 4.6661,
      "mean_token_accuracy": 0.22817434668540953,
      "num_tokens": 53867823.0,
      "step": 29285
    },
    {
      "entropy": 5.0794964790344235,
      "epoch": 2.4465419311727366,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043984619804886423,
      "loss": 4.7837,
      "mean_token_accuracy": 0.2259153753519058,
      "num_tokens": 53877007.0,
      "step": 29290
    },
    {
      "entropy": 4.930950355529785,
      "epoch": 2.4469595723354494,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004398259349852877,
      "loss": 4.6217,
      "mean_token_accuracy": 0.23037814795970918,
      "num_tokens": 53885518.0,
      "step": 29295
    },
    {
      "entropy": 5.08004355430603,
      "epoch": 2.4473772134981626,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043980566903623487,
      "loss": 4.7698,
      "mean_token_accuracy": 0.22262530028820038,
      "num_tokens": 53894919.0,
      "step": 29300
    },
    {
      "entropy": 4.993763446807861,
      "epoch": 2.4477948546608754,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004397854002020604,
      "loss": 4.6592,
      "mean_token_accuracy": 0.23083720952272416,
      "num_tokens": 53904393.0,
      "step": 29305
    },
    {
      "entropy": 4.994530200958252,
      "epoch": 2.4482124958235882,
      "grad_norm": 1.125,
      "learning_rate": 0.00043976512848311906,
      "loss": 4.6397,
      "mean_token_accuracy": 0.22421533316373826,
      "num_tokens": 53913586.0,
      "step": 29310
    },
    {
      "entropy": 4.997750663757325,
      "epoch": 2.4486301369863015,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004397448538797659,
      "loss": 4.6811,
      "mean_token_accuracy": 0.22950809299945832,
      "num_tokens": 53922458.0,
      "step": 29315
    },
    {
      "entropy": 4.975944328308105,
      "epoch": 2.4490477781490143,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004397245763923557,
      "loss": 4.6609,
      "mean_token_accuracy": 0.22766398638486862,
      "num_tokens": 53930953.0,
      "step": 29320
    },
    {
      "entropy": 4.998822259902954,
      "epoch": 2.4494654193117276,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004397042960212436,
      "loss": 4.6578,
      "mean_token_accuracy": 0.2211569920182228,
      "num_tokens": 53940519.0,
      "step": 29325
    },
    {
      "entropy": 5.030440139770508,
      "epoch": 2.4498830604744404,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004396840127667845,
      "loss": 4.6589,
      "mean_token_accuracy": 0.23288680464029313,
      "num_tokens": 53949938.0,
      "step": 29330
    },
    {
      "entropy": 5.0593496799469,
      "epoch": 2.450300701637153,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004396637266293335,
      "loss": 4.8666,
      "mean_token_accuracy": 0.2099646121263504,
      "num_tokens": 53958714.0,
      "step": 29335
    },
    {
      "entropy": 5.030391359329224,
      "epoch": 2.4507183427998664,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004396434376092458,
      "loss": 4.66,
      "mean_token_accuracy": 0.23345884680747986,
      "num_tokens": 53968160.0,
      "step": 29340
    },
    {
      "entropy": 5.037456607818603,
      "epoch": 2.4511359839625793,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004396231457068765,
      "loss": 4.6594,
      "mean_token_accuracy": 0.23329162299633027,
      "num_tokens": 53977198.0,
      "step": 29345
    },
    {
      "entropy": 5.077903366088867,
      "epoch": 2.4515536251252925,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043960285092258097,
      "loss": 4.7609,
      "mean_token_accuracy": 0.21869688034057616,
      "num_tokens": 53986278.0,
      "step": 29350
    },
    {
      "entropy": 5.001395750045776,
      "epoch": 2.4519712662880053,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004395825532567143,
      "loss": 4.7084,
      "mean_token_accuracy": 0.2263369917869568,
      "num_tokens": 53995197.0,
      "step": 29355
    },
    {
      "entropy": 4.979094934463501,
      "epoch": 2.452388907450718,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000439562252709632,
      "loss": 4.6743,
      "mean_token_accuracy": 0.2270834267139435,
      "num_tokens": 54004636.0,
      "step": 29360
    },
    {
      "entropy": 4.929143381118775,
      "epoch": 2.4528065486134314,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004395419492816894,
      "loss": 4.5598,
      "mean_token_accuracy": 0.24430026710033417,
      "num_tokens": 54012757.0,
      "step": 29365
    },
    {
      "entropy": 5.077042102813721,
      "epoch": 2.453224189776144,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004395216429732418,
      "loss": 4.8191,
      "mean_token_accuracy": 0.219437937438488,
      "num_tokens": 54022416.0,
      "step": 29370
    },
    {
      "entropy": 4.992264080047607,
      "epoch": 2.4536418309388575,
      "grad_norm": 1.125,
      "learning_rate": 0.00043950133378464485,
      "loss": 4.6416,
      "mean_token_accuracy": 0.22922578752040862,
      "num_tokens": 54031340.0,
      "step": 29375
    },
    {
      "entropy": 5.010110235214233,
      "epoch": 2.4540594721015703,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004394810217162541,
      "loss": 4.6733,
      "mean_token_accuracy": 0.2190912514925003,
      "num_tokens": 54041511.0,
      "step": 29380
    },
    {
      "entropy": 5.014117908477783,
      "epoch": 2.454477113264283,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004394607067684251,
      "loss": 4.6749,
      "mean_token_accuracy": 0.2226187378168106,
      "num_tokens": 54050075.0,
      "step": 29385
    },
    {
      "entropy": 5.010887193679809,
      "epoch": 2.4548947544269963,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043944038894151336,
      "loss": 4.7115,
      "mean_token_accuracy": 0.22663243860006332,
      "num_tokens": 54059907.0,
      "step": 29390
    },
    {
      "entropy": 4.946980619430542,
      "epoch": 2.455312395589709,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004394200682358748,
      "loss": 4.6672,
      "mean_token_accuracy": 0.2202286183834076,
      "num_tokens": 54070680.0,
      "step": 29395
    },
    {
      "entropy": 4.981841564178467,
      "epoch": 2.4557300367524224,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004393997446518649,
      "loss": 4.6454,
      "mean_token_accuracy": 0.22937535643577575,
      "num_tokens": 54079866.0,
      "step": 29400
    },
    {
      "entropy": 5.052939796447754,
      "epoch": 2.456147677915135,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004393794181898397,
      "loss": 4.7227,
      "mean_token_accuracy": 0.23014288097620011,
      "num_tokens": 54088001.0,
      "step": 29405
    },
    {
      "entropy": 5.003994798660278,
      "epoch": 2.4565653190778485,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004393590888501548,
      "loss": 4.6776,
      "mean_token_accuracy": 0.2280384212732315,
      "num_tokens": 54097350.0,
      "step": 29410
    },
    {
      "entropy": 4.965285205841065,
      "epoch": 2.4569829602405613,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004393387566331663,
      "loss": 4.66,
      "mean_token_accuracy": 0.23060558438301088,
      "num_tokens": 54106706.0,
      "step": 29415
    },
    {
      "entropy": 4.996157884597778,
      "epoch": 2.4574006014032745,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043931842153923,
      "loss": 4.7363,
      "mean_token_accuracy": 0.23010325729846953,
      "num_tokens": 54116089.0,
      "step": 29420
    },
    {
      "entropy": 4.962317943572998,
      "epoch": 2.4578182425659874,
      "grad_norm": 1.125,
      "learning_rate": 0.000439298083568702,
      "loss": 4.6742,
      "mean_token_accuracy": 0.22325651198625565,
      "num_tokens": 54125852.0,
      "step": 29425
    },
    {
      "entropy": 4.967683267593384,
      "epoch": 2.4582358837287,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004392777427219382,
      "loss": 4.5802,
      "mean_token_accuracy": 0.23767400085926055,
      "num_tokens": 54134532.0,
      "step": 29430
    },
    {
      "entropy": 5.021471691131592,
      "epoch": 2.4586535248914134,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043925739899929484,
      "loss": 4.7753,
      "mean_token_accuracy": 0.22245322167873383,
      "num_tokens": 54144121.0,
      "step": 29435
    },
    {
      "entropy": 4.967955303192139,
      "epoch": 2.4590711660541262,
      "grad_norm": 1.0625,
      "learning_rate": 0.000439237052401128,
      "loss": 4.5821,
      "mean_token_accuracy": 0.23700257986783982,
      "num_tokens": 54153018.0,
      "step": 29440
    },
    {
      "entropy": 5.045514345169067,
      "epoch": 2.4594888072168395,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043921670292779386,
      "loss": 4.7078,
      "mean_token_accuracy": 0.22063106298446655,
      "num_tokens": 54162391.0,
      "step": 29445
    },
    {
      "entropy": 5.029855489730835,
      "epoch": 2.4599064483795523,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004391963505796487,
      "loss": 4.6547,
      "mean_token_accuracy": 0.21974731981754303,
      "num_tokens": 54171575.0,
      "step": 29450
    },
    {
      "entropy": 5.049818229675293,
      "epoch": 2.460324089542265,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004391759953570487,
      "loss": 4.788,
      "mean_token_accuracy": 0.21177101582288743,
      "num_tokens": 54181379.0,
      "step": 29455
    },
    {
      "entropy": 5.075631999969483,
      "epoch": 2.4607417307049784,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004391556372603503,
      "loss": 4.6885,
      "mean_token_accuracy": 0.22479359805583954,
      "num_tokens": 54190582.0,
      "step": 29460
    },
    {
      "entropy": 5.101151657104492,
      "epoch": 2.461159371867691,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043913527628990984,
      "loss": 4.736,
      "mean_token_accuracy": 0.21776345372200012,
      "num_tokens": 54199884.0,
      "step": 29465
    },
    {
      "entropy": 5.019132995605469,
      "epoch": 2.4615770130304044,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043911491244608386,
      "loss": 4.728,
      "mean_token_accuracy": 0.22151595503091812,
      "num_tokens": 54209562.0,
      "step": 29470
    },
    {
      "entropy": 5.080878829956054,
      "epoch": 2.4619946541931172,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004390945457292288,
      "loss": 4.791,
      "mean_token_accuracy": 0.21958024501800538,
      "num_tokens": 54218652.0,
      "step": 29475
    },
    {
      "entropy": 5.0230389595031735,
      "epoch": 2.46241229535583,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004390741761397012,
      "loss": 4.6672,
      "mean_token_accuracy": 0.2247052252292633,
      "num_tokens": 54227231.0,
      "step": 29480
    },
    {
      "entropy": 5.008385181427002,
      "epoch": 2.4628299365185433,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043905380367785757,
      "loss": 4.6629,
      "mean_token_accuracy": 0.22831055372953415,
      "num_tokens": 54236357.0,
      "step": 29485
    },
    {
      "entropy": 4.979161024093628,
      "epoch": 2.463247577681256,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004390334283440547,
      "loss": 4.7014,
      "mean_token_accuracy": 0.21235381364822387,
      "num_tokens": 54245315.0,
      "step": 29490
    },
    {
      "entropy": 5.013259506225586,
      "epoch": 2.4636652188439694,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043901305013864923,
      "loss": 4.6727,
      "mean_token_accuracy": 0.2268316552042961,
      "num_tokens": 54253567.0,
      "step": 29495
    },
    {
      "entropy": 5.01284646987915,
      "epoch": 2.464082860006682,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00043899266906199784,
      "loss": 4.7292,
      "mean_token_accuracy": 0.21863813698291779,
      "num_tokens": 54263366.0,
      "step": 29500
    },
    {
      "entropy": 5.062762594223022,
      "epoch": 2.464500501169395,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004389722851144574,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2269522711634636,
      "num_tokens": 54272938.0,
      "step": 29505
    },
    {
      "entropy": 5.0034332275390625,
      "epoch": 2.4649181423321083,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043895189829638485,
      "loss": 4.6631,
      "mean_token_accuracy": 0.22686034440994263,
      "num_tokens": 54281547.0,
      "step": 29510
    },
    {
      "entropy": 4.9559979915618895,
      "epoch": 2.465335783494821,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004389315086081368,
      "loss": 4.666,
      "mean_token_accuracy": 0.23260541260242462,
      "num_tokens": 54289633.0,
      "step": 29515
    },
    {
      "entropy": 5.057028913497925,
      "epoch": 2.4657534246575343,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043891111605007043,
      "loss": 4.6647,
      "mean_token_accuracy": 0.22501132190227507,
      "num_tokens": 54298698.0,
      "step": 29520
    },
    {
      "entropy": 4.917834329605102,
      "epoch": 2.466171065820247,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004388907206225427,
      "loss": 4.6435,
      "mean_token_accuracy": 0.2273445323109627,
      "num_tokens": 54307049.0,
      "step": 29525
    },
    {
      "entropy": 5.000425481796265,
      "epoch": 2.4665887069829604,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004388703223259106,
      "loss": 4.6624,
      "mean_token_accuracy": 0.23753853142261505,
      "num_tokens": 54315658.0,
      "step": 29530
    },
    {
      "entropy": 4.99507999420166,
      "epoch": 2.467006348145673,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043884992116053137,
      "loss": 4.6809,
      "mean_token_accuracy": 0.22681910991668702,
      "num_tokens": 54324627.0,
      "step": 29535
    },
    {
      "entropy": 4.990749883651733,
      "epoch": 2.4674239893083865,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043882951712676183,
      "loss": 4.7315,
      "mean_token_accuracy": 0.22103019952774047,
      "num_tokens": 54333829.0,
      "step": 29540
    },
    {
      "entropy": 4.989173841476441,
      "epoch": 2.4678416304710993,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043880911022495955,
      "loss": 4.7215,
      "mean_token_accuracy": 0.2229451894760132,
      "num_tokens": 54342395.0,
      "step": 29545
    },
    {
      "entropy": 5.010736989974975,
      "epoch": 2.468259271633812,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004387887004554816,
      "loss": 4.6438,
      "mean_token_accuracy": 0.22656549364328385,
      "num_tokens": 54351411.0,
      "step": 29550
    },
    {
      "entropy": 4.986594200134277,
      "epoch": 2.4686769127965253,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004387682878186852,
      "loss": 4.6515,
      "mean_token_accuracy": 0.22884147465229035,
      "num_tokens": 54361082.0,
      "step": 29555
    },
    {
      "entropy": 5.018958663940429,
      "epoch": 2.469094553959238,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004387478723149279,
      "loss": 4.7109,
      "mean_token_accuracy": 0.22604213804006576,
      "num_tokens": 54369669.0,
      "step": 29560
    },
    {
      "entropy": 5.110363483428955,
      "epoch": 2.4695121951219514,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043872745394456697,
      "loss": 4.7474,
      "mean_token_accuracy": 0.2237729698419571,
      "num_tokens": 54379947.0,
      "step": 29565
    },
    {
      "entropy": 5.124659156799316,
      "epoch": 2.4699298362846642,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043870703270795986,
      "loss": 4.8032,
      "mean_token_accuracy": 0.2153582528233528,
      "num_tokens": 54390187.0,
      "step": 29570
    },
    {
      "entropy": 5.021060514450073,
      "epoch": 2.470347477447377,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004386866086054641,
      "loss": 4.7194,
      "mean_token_accuracy": 0.2250710681080818,
      "num_tokens": 54399496.0,
      "step": 29575
    },
    {
      "entropy": 5.0611090660095215,
      "epoch": 2.4707651186100903,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043866618163743723,
      "loss": 4.7704,
      "mean_token_accuracy": 0.2189513698220253,
      "num_tokens": 54408697.0,
      "step": 29580
    },
    {
      "entropy": 5.029923915863037,
      "epoch": 2.471182759772803,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043864575180423685,
      "loss": 4.7172,
      "mean_token_accuracy": 0.22888316363096237,
      "num_tokens": 54417760.0,
      "step": 29585
    },
    {
      "entropy": 5.145903253555298,
      "epoch": 2.4716004009355164,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043862531910622056,
      "loss": 4.8648,
      "mean_token_accuracy": 0.21597041338682174,
      "num_tokens": 54427308.0,
      "step": 29590
    },
    {
      "entropy": 4.988644170761108,
      "epoch": 2.472018042098229,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043860488354374613,
      "loss": 4.7393,
      "mean_token_accuracy": 0.22234202325344085,
      "num_tokens": 54437602.0,
      "step": 29595
    },
    {
      "entropy": 5.06865086555481,
      "epoch": 2.472435683260942,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043858444511717126,
      "loss": 4.7256,
      "mean_token_accuracy": 0.22931522578001023,
      "num_tokens": 54446533.0,
      "step": 29600
    },
    {
      "entropy": 4.928246688842774,
      "epoch": 2.4728533244236552,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043856400382685383,
      "loss": 4.5429,
      "mean_token_accuracy": 0.2393283262848854,
      "num_tokens": 54455670.0,
      "step": 29605
    },
    {
      "entropy": 5.040971517562866,
      "epoch": 2.473270965586368,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004385435596731516,
      "loss": 4.6947,
      "mean_token_accuracy": 0.21425127983093262,
      "num_tokens": 54466202.0,
      "step": 29610
    },
    {
      "entropy": 5.051441478729248,
      "epoch": 2.4736886067490813,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043852311265642246,
      "loss": 4.6984,
      "mean_token_accuracy": 0.2294282555580139,
      "num_tokens": 54474662.0,
      "step": 29615
    },
    {
      "entropy": 4.930345249176026,
      "epoch": 2.474106247911794,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043850266277702446,
      "loss": 4.594,
      "mean_token_accuracy": 0.23083053678274154,
      "num_tokens": 54483748.0,
      "step": 29620
    },
    {
      "entropy": 5.0185229778289795,
      "epoch": 2.4745238890745074,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004384822100353155,
      "loss": 4.8226,
      "mean_token_accuracy": 0.2138156458735466,
      "num_tokens": 54494346.0,
      "step": 29625
    },
    {
      "entropy": 5.028608942031861,
      "epoch": 2.47494153023722,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004384617544316537,
      "loss": 4.6691,
      "mean_token_accuracy": 0.22414208203554153,
      "num_tokens": 54503511.0,
      "step": 29630
    },
    {
      "entropy": 5.047468566894532,
      "epoch": 2.475359171399933,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043844129596639714,
      "loss": 4.6769,
      "mean_token_accuracy": 0.22716744840145112,
      "num_tokens": 54512168.0,
      "step": 29635
    },
    {
      "entropy": 4.9121985912323,
      "epoch": 2.4757768125626463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000438420834639904,
      "loss": 4.6392,
      "mean_token_accuracy": 0.22803011685609817,
      "num_tokens": 54520906.0,
      "step": 29640
    },
    {
      "entropy": 5.027965784072876,
      "epoch": 2.476194453725359,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043840037045253234,
      "loss": 4.6968,
      "mean_token_accuracy": 0.2278583750128746,
      "num_tokens": 54530211.0,
      "step": 29645
    },
    {
      "entropy": 5.05372052192688,
      "epoch": 2.4766120948880723,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004383799034046405,
      "loss": 4.7449,
      "mean_token_accuracy": 0.22151085883378982,
      "num_tokens": 54538885.0,
      "step": 29650
    },
    {
      "entropy": 5.014594793319702,
      "epoch": 2.477029736050785,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004383594334965869,
      "loss": 4.6653,
      "mean_token_accuracy": 0.22440161257982255,
      "num_tokens": 54548034.0,
      "step": 29655
    },
    {
      "entropy": 5.1157989501953125,
      "epoch": 2.4774473772134984,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004383389607287297,
      "loss": 4.7167,
      "mean_token_accuracy": 0.2291305959224701,
      "num_tokens": 54556328.0,
      "step": 29660
    },
    {
      "entropy": 4.957929515838623,
      "epoch": 2.477865018376211,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004383184851014274,
      "loss": 4.6912,
      "mean_token_accuracy": 0.23211260586977006,
      "num_tokens": 54565891.0,
      "step": 29665
    },
    {
      "entropy": 4.990735054016113,
      "epoch": 2.478282659538924,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043829800661503843,
      "loss": 4.6995,
      "mean_token_accuracy": 0.2293139234185219,
      "num_tokens": 54576440.0,
      "step": 29670
    },
    {
      "entropy": 5.122193479537964,
      "epoch": 2.4787003007016373,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043827752526992125,
      "loss": 4.748,
      "mean_token_accuracy": 0.22709666937589645,
      "num_tokens": 54585805.0,
      "step": 29675
    },
    {
      "entropy": 5.001657485961914,
      "epoch": 2.47911794186435,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043825704106643454,
      "loss": 4.6406,
      "mean_token_accuracy": 0.2350724905729294,
      "num_tokens": 54595327.0,
      "step": 29680
    },
    {
      "entropy": 4.968381547927857,
      "epoch": 2.4795355830270633,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043823655400493676,
      "loss": 4.7453,
      "mean_token_accuracy": 0.22359900176525116,
      "num_tokens": 54604252.0,
      "step": 29685
    },
    {
      "entropy": 5.022830009460449,
      "epoch": 2.479953224189776,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043821606408578665,
      "loss": 4.6551,
      "mean_token_accuracy": 0.22828968167304992,
      "num_tokens": 54613049.0,
      "step": 29690
    },
    {
      "entropy": 5.002430009841919,
      "epoch": 2.480370865352489,
      "grad_norm": 1.125,
      "learning_rate": 0.0004381955713093428,
      "loss": 4.6483,
      "mean_token_accuracy": 0.22954309731721878,
      "num_tokens": 54622797.0,
      "step": 29695
    },
    {
      "entropy": 4.97469801902771,
      "epoch": 2.480788506515202,
      "grad_norm": 1.125,
      "learning_rate": 0.00043817507567596415,
      "loss": 4.6836,
      "mean_token_accuracy": 0.22740351110696794,
      "num_tokens": 54632039.0,
      "step": 29700
    },
    {
      "entropy": 5.0725044250488285,
      "epoch": 2.481206147677915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043815457718600933,
      "loss": 4.7274,
      "mean_token_accuracy": 0.22577583342790603,
      "num_tokens": 54641989.0,
      "step": 29705
    },
    {
      "entropy": 4.966227006912232,
      "epoch": 2.4816237888406283,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004381340758398372,
      "loss": 4.7578,
      "mean_token_accuracy": 0.22276390790939332,
      "num_tokens": 54650567.0,
      "step": 29710
    },
    {
      "entropy": 5.039543151855469,
      "epoch": 2.482041430003341,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043811357163780676,
      "loss": 4.7236,
      "mean_token_accuracy": 0.22724349647760392,
      "num_tokens": 54660028.0,
      "step": 29715
    },
    {
      "entropy": 4.989207363128662,
      "epoch": 2.482459071166054,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004380930645802769,
      "loss": 4.6082,
      "mean_token_accuracy": 0.23221104443073273,
      "num_tokens": 54669490.0,
      "step": 29720
    },
    {
      "entropy": 5.019079923629761,
      "epoch": 2.482876712328767,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004380725546676067,
      "loss": 4.7456,
      "mean_token_accuracy": 0.22412191927433014,
      "num_tokens": 54678284.0,
      "step": 29725
    },
    {
      "entropy": 5.068228769302368,
      "epoch": 2.48329435349148,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043805204190015503,
      "loss": 4.8354,
      "mean_token_accuracy": 0.21039935648441316,
      "num_tokens": 54688903.0,
      "step": 29730
    },
    {
      "entropy": 5.105949401855469,
      "epoch": 2.4837119946541932,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004380315262782811,
      "loss": 4.7636,
      "mean_token_accuracy": 0.21989082247018815,
      "num_tokens": 54698011.0,
      "step": 29735
    },
    {
      "entropy": 5.059952306747436,
      "epoch": 2.484129635816906,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004380110078023441,
      "loss": 4.638,
      "mean_token_accuracy": 0.22925180643796922,
      "num_tokens": 54707226.0,
      "step": 29740
    },
    {
      "entropy": 5.029758167266846,
      "epoch": 2.4845472769796193,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004379904864727032,
      "loss": 4.7886,
      "mean_token_accuracy": 0.22091748863458632,
      "num_tokens": 54715889.0,
      "step": 29745
    },
    {
      "entropy": 5.0752037525177,
      "epoch": 2.484964918142332,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043796996228971766,
      "loss": 4.7969,
      "mean_token_accuracy": 0.22378707379102708,
      "num_tokens": 54724938.0,
      "step": 29750
    },
    {
      "entropy": 5.000101947784424,
      "epoch": 2.485382559305045,
      "grad_norm": 1.125,
      "learning_rate": 0.0004379494352537467,
      "loss": 4.6175,
      "mean_token_accuracy": 0.23370094001293182,
      "num_tokens": 54734266.0,
      "step": 29755
    },
    {
      "entropy": 4.910245513916015,
      "epoch": 2.485800200467758,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004379289053651498,
      "loss": 4.6119,
      "mean_token_accuracy": 0.22525040358304976,
      "num_tokens": 54744105.0,
      "step": 29760
    },
    {
      "entropy": 5.042551803588867,
      "epoch": 2.486217841630471,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004379083726242862,
      "loss": 4.7774,
      "mean_token_accuracy": 0.22202294021844865,
      "num_tokens": 54753303.0,
      "step": 29765
    },
    {
      "entropy": 5.10030837059021,
      "epoch": 2.4866354827931842,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004378878370315156,
      "loss": 4.7304,
      "mean_token_accuracy": 0.2228455901145935,
      "num_tokens": 54762670.0,
      "step": 29770
    },
    {
      "entropy": 5.029805994033813,
      "epoch": 2.487053123955897,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004378672985871972,
      "loss": 4.6914,
      "mean_token_accuracy": 0.22502330988645552,
      "num_tokens": 54772748.0,
      "step": 29775
    },
    {
      "entropy": 5.006698083877564,
      "epoch": 2.4874707651186103,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043784675729169076,
      "loss": 4.7597,
      "mean_token_accuracy": 0.21820629090070726,
      "num_tokens": 54782339.0,
      "step": 29780
    },
    {
      "entropy": 5.053822994232178,
      "epoch": 2.487888406281323,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004378262131453558,
      "loss": 4.7151,
      "mean_token_accuracy": 0.22525831907987595,
      "num_tokens": 54791698.0,
      "step": 29785
    },
    {
      "entropy": 4.990624952316284,
      "epoch": 2.488306047444036,
      "grad_norm": 1.140625,
      "learning_rate": 0.000437805666148552,
      "loss": 4.6846,
      "mean_token_accuracy": 0.22586403489112855,
      "num_tokens": 54800662.0,
      "step": 29790
    },
    {
      "entropy": 5.07534499168396,
      "epoch": 2.488723688606749,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043778511630163905,
      "loss": 4.6909,
      "mean_token_accuracy": 0.22489008605480193,
      "num_tokens": 54810052.0,
      "step": 29795
    },
    {
      "entropy": 5.063573837280273,
      "epoch": 2.489141329769462,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043776456360497665,
      "loss": 4.772,
      "mean_token_accuracy": 0.22488994598388673,
      "num_tokens": 54819519.0,
      "step": 29800
    },
    {
      "entropy": 4.894841146469116,
      "epoch": 2.4895589709321753,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004377440080589247,
      "loss": 4.6251,
      "mean_token_accuracy": 0.23256967812776566,
      "num_tokens": 54828190.0,
      "step": 29805
    },
    {
      "entropy": 5.034190273284912,
      "epoch": 2.489976612094888,
      "grad_norm": 1.140625,
      "learning_rate": 0.000437723449663843,
      "loss": 4.7231,
      "mean_token_accuracy": 0.22943077087402344,
      "num_tokens": 54837375.0,
      "step": 29810
    },
    {
      "entropy": 4.992992305755616,
      "epoch": 2.490394253257601,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004377028884200914,
      "loss": 4.6638,
      "mean_token_accuracy": 0.23223376423120498,
      "num_tokens": 54847090.0,
      "step": 29815
    },
    {
      "entropy": 4.997753858566284,
      "epoch": 2.490811894420314,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004376823243280298,
      "loss": 4.7449,
      "mean_token_accuracy": 0.21910440623760224,
      "num_tokens": 54856753.0,
      "step": 29820
    },
    {
      "entropy": 5.061763954162598,
      "epoch": 2.491229535583027,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004376617573880185,
      "loss": 4.7887,
      "mean_token_accuracy": 0.21684341728687287,
      "num_tokens": 54865899.0,
      "step": 29825
    },
    {
      "entropy": 5.032311153411865,
      "epoch": 2.49164717674574,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004376411876004172,
      "loss": 4.6423,
      "mean_token_accuracy": 0.22859692424535752,
      "num_tokens": 54874524.0,
      "step": 29830
    },
    {
      "entropy": 4.978237581253052,
      "epoch": 2.492064817908453,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004376206149655862,
      "loss": 4.6524,
      "mean_token_accuracy": 0.2295512616634369,
      "num_tokens": 54883885.0,
      "step": 29835
    },
    {
      "entropy": 4.953192138671875,
      "epoch": 2.492482459071166,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004376000394838856,
      "loss": 4.6476,
      "mean_token_accuracy": 0.22946895956993102,
      "num_tokens": 54892792.0,
      "step": 29840
    },
    {
      "entropy": 5.0086150646209715,
      "epoch": 2.492900100233879,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004375794611556755,
      "loss": 4.636,
      "mean_token_accuracy": 0.22966762781143188,
      "num_tokens": 54901591.0,
      "step": 29845
    },
    {
      "entropy": 4.95533299446106,
      "epoch": 2.493317741396592,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004375588799813163,
      "loss": 4.6479,
      "mean_token_accuracy": 0.22734307795763015,
      "num_tokens": 54911146.0,
      "step": 29850
    },
    {
      "entropy": 5.021447229385376,
      "epoch": 2.493735382559305,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004375382959611682,
      "loss": 4.7036,
      "mean_token_accuracy": 0.22545846402645112,
      "num_tokens": 54920193.0,
      "step": 29855
    },
    {
      "entropy": 5.101010274887085,
      "epoch": 2.494153023722018,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043751770909559155,
      "loss": 4.6929,
      "mean_token_accuracy": 0.22173014879226685,
      "num_tokens": 54931374.0,
      "step": 29860
    },
    {
      "entropy": 5.02647967338562,
      "epoch": 2.4945706648847312,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043749711938494693,
      "loss": 4.7161,
      "mean_token_accuracy": 0.22025876492261887,
      "num_tokens": 54940758.0,
      "step": 29865
    },
    {
      "entropy": 4.974259567260742,
      "epoch": 2.494988306047444,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043747652682959455,
      "loss": 4.6674,
      "mean_token_accuracy": 0.22440795749425888,
      "num_tokens": 54950195.0,
      "step": 29870
    },
    {
      "entropy": 5.034496641159057,
      "epoch": 2.495405947210157,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004374559314298949,
      "loss": 4.804,
      "mean_token_accuracy": 0.22212871611118318,
      "num_tokens": 54958829.0,
      "step": 29875
    },
    {
      "entropy": 4.980949592590332,
      "epoch": 2.49582358837287,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043743533318620873,
      "loss": 4.6472,
      "mean_token_accuracy": 0.23415371924638748,
      "num_tokens": 54967946.0,
      "step": 29880
    },
    {
      "entropy": 5.103853464126587,
      "epoch": 2.496241229535583,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043741473209889654,
      "loss": 4.7742,
      "mean_token_accuracy": 0.2252495542168617,
      "num_tokens": 54977178.0,
      "step": 29885
    },
    {
      "entropy": 5.030395746231079,
      "epoch": 2.496658870698296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004373941281683189,
      "loss": 4.6606,
      "mean_token_accuracy": 0.23229423314332961,
      "num_tokens": 54985753.0,
      "step": 29890
    },
    {
      "entropy": 5.014884567260742,
      "epoch": 2.497076511861009,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004373735213948366,
      "loss": 4.7337,
      "mean_token_accuracy": 0.2196944385766983,
      "num_tokens": 54994728.0,
      "step": 29895
    },
    {
      "entropy": 4.925800514221192,
      "epoch": 2.4974941530237222,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043735291177881045,
      "loss": 4.593,
      "mean_token_accuracy": 0.23957917094230652,
      "num_tokens": 55003685.0,
      "step": 29900
    },
    {
      "entropy": 5.00201096534729,
      "epoch": 2.497911794186435,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000437332299320601,
      "loss": 4.6678,
      "mean_token_accuracy": 0.22679000943899155,
      "num_tokens": 55012571.0,
      "step": 29905
    },
    {
      "entropy": 5.124615859985352,
      "epoch": 2.498329435349148,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004373116840205693,
      "loss": 4.838,
      "mean_token_accuracy": 0.21440956443548204,
      "num_tokens": 55022908.0,
      "step": 29910
    },
    {
      "entropy": 4.998402786254883,
      "epoch": 2.498747076511861,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043729106587907625,
      "loss": 4.5839,
      "mean_token_accuracy": 0.23555291593074798,
      "num_tokens": 55030896.0,
      "step": 29915
    },
    {
      "entropy": 5.0559851169586185,
      "epoch": 2.499164717674574,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043727044489648263,
      "loss": 4.8375,
      "mean_token_accuracy": 0.2259227976202965,
      "num_tokens": 55039487.0,
      "step": 29920
    },
    {
      "entropy": 5.027026224136352,
      "epoch": 2.499582358837287,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004372498210731497,
      "loss": 4.7006,
      "mean_token_accuracy": 0.2242559775710106,
      "num_tokens": 55048205.0,
      "step": 29925
    },
    {
      "entropy": 4.973114252090454,
      "epoch": 2.5,
      "grad_norm": 1.125,
      "learning_rate": 0.00043722919440943833,
      "loss": 4.6193,
      "mean_token_accuracy": 0.23215944916009904,
      "num_tokens": 55057060.0,
      "step": 29930
    },
    {
      "entropy": 4.947098255157471,
      "epoch": 2.500417641162713,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004372085649057095,
      "loss": 4.603,
      "mean_token_accuracy": 0.2293909505009651,
      "num_tokens": 55065905.0,
      "step": 29935
    },
    {
      "entropy": 5.07386736869812,
      "epoch": 2.500835282325426,
      "grad_norm": 1.125,
      "learning_rate": 0.00043718793256232463,
      "loss": 4.7133,
      "mean_token_accuracy": 0.2316938027739525,
      "num_tokens": 55074411.0,
      "step": 29940
    },
    {
      "entropy": 5.046506118774414,
      "epoch": 2.501252923488139,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043716729737964475,
      "loss": 4.7782,
      "mean_token_accuracy": 0.2234795168042183,
      "num_tokens": 55082987.0,
      "step": 29945
    },
    {
      "entropy": 4.942578125,
      "epoch": 2.501670564650852,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043714665935803103,
      "loss": 4.5865,
      "mean_token_accuracy": 0.23223154097795487,
      "num_tokens": 55092067.0,
      "step": 29950
    },
    {
      "entropy": 5.02805986404419,
      "epoch": 2.502088205813565,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043712601849784494,
      "loss": 4.7078,
      "mean_token_accuracy": 0.2193588897585869,
      "num_tokens": 55101087.0,
      "step": 29955
    },
    {
      "entropy": 5.062891674041748,
      "epoch": 2.5025058469762778,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043710537479944774,
      "loss": 4.8526,
      "mean_token_accuracy": 0.21493545770645142,
      "num_tokens": 55110782.0,
      "step": 29960
    },
    {
      "entropy": 4.949912405014038,
      "epoch": 2.502923488138991,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043708472826320085,
      "loss": 4.5559,
      "mean_token_accuracy": 0.24527808874845505,
      "num_tokens": 55118573.0,
      "step": 29965
    },
    {
      "entropy": 5.029082202911377,
      "epoch": 2.503341129301704,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004370640788894656,
      "loss": 4.7382,
      "mean_token_accuracy": 0.23148789405822753,
      "num_tokens": 55128084.0,
      "step": 29970
    },
    {
      "entropy": 4.9858564853668215,
      "epoch": 2.503758770464417,
      "grad_norm": 1.125,
      "learning_rate": 0.0004370434266786037,
      "loss": 4.6309,
      "mean_token_accuracy": 0.23487247228622438,
      "num_tokens": 55137256.0,
      "step": 29975
    },
    {
      "entropy": 4.950522232055664,
      "epoch": 2.50417641162713,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004370227716309765,
      "loss": 4.6365,
      "mean_token_accuracy": 0.22509715259075164,
      "num_tokens": 55147252.0,
      "step": 29980
    },
    {
      "entropy": 5.054384660720825,
      "epoch": 2.5045940527898427,
      "grad_norm": 1.125,
      "learning_rate": 0.00043700211374694565,
      "loss": 4.7922,
      "mean_token_accuracy": 0.22055012583732606,
      "num_tokens": 55155591.0,
      "step": 29985
    },
    {
      "entropy": 5.070456266403198,
      "epoch": 2.505011693952556,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004369814530268728,
      "loss": 4.6634,
      "mean_token_accuracy": 0.2276929050683975,
      "num_tokens": 55164668.0,
      "step": 29990
    },
    {
      "entropy": 5.128434133529663,
      "epoch": 2.505429335115269,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004369607894711196,
      "loss": 4.7197,
      "mean_token_accuracy": 0.22334203273057937,
      "num_tokens": 55173483.0,
      "step": 29995
    },
    {
      "entropy": 4.942606592178345,
      "epoch": 2.505846976277982,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004369401230800479,
      "loss": 4.611,
      "mean_token_accuracy": 0.23367864340543748,
      "num_tokens": 55181696.0,
      "step": 30000
    },
    {
      "epoch": 2.505846976277982,
      "eval_entropy": 4.710070815739857,
      "eval_loss": 4.744089126586914,
      "eval_mean_token_accuracy": 0.2351193139248671,
      "eval_num_tokens": 55181696.0,
      "eval_runtime": 27.7211,
      "eval_samples_per_second": 1362.246,
      "eval_steps_per_second": 170.303,
      "step": 30000
    },
    {
      "entropy": 4.836050128936767,
      "epoch": 2.506264617440695,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004369194538540194,
      "loss": 4.5136,
      "mean_token_accuracy": 0.24048410058021547,
      "num_tokens": 55191107.0,
      "step": 30005
    },
    {
      "entropy": 4.990440893173218,
      "epoch": 2.506682258603408,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043689878179339593,
      "loss": 4.6585,
      "mean_token_accuracy": 0.2270812451839447,
      "num_tokens": 55199562.0,
      "step": 30010
    },
    {
      "entropy": 5.037631273269653,
      "epoch": 2.507099899766121,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004368781068985394,
      "loss": 4.7201,
      "mean_token_accuracy": 0.22691605240106583,
      "num_tokens": 55208880.0,
      "step": 30015
    },
    {
      "entropy": 4.943087768554688,
      "epoch": 2.507517540928834,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004368574291698118,
      "loss": 4.6088,
      "mean_token_accuracy": 0.23055688589811324,
      "num_tokens": 55218402.0,
      "step": 30020
    },
    {
      "entropy": 5.0013233661651615,
      "epoch": 2.507935182091547,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000436836748607575,
      "loss": 4.7048,
      "mean_token_accuracy": 0.23579045981168748,
      "num_tokens": 55227310.0,
      "step": 30025
    },
    {
      "entropy": 4.907833814620972,
      "epoch": 2.50835282325426,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004368160652121912,
      "loss": 4.5935,
      "mean_token_accuracy": 0.2325818046927452,
      "num_tokens": 55236769.0,
      "step": 30030
    },
    {
      "entropy": 5.034113359451294,
      "epoch": 2.508770464416973,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004367953789840224,
      "loss": 4.8274,
      "mean_token_accuracy": 0.21731277853250502,
      "num_tokens": 55246632.0,
      "step": 30035
    },
    {
      "entropy": 5.0467854022979735,
      "epoch": 2.509188105579686,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043677468992343074,
      "loss": 4.658,
      "mean_token_accuracy": 0.22605662792921066,
      "num_tokens": 55255600.0,
      "step": 30040
    },
    {
      "entropy": 5.107480716705322,
      "epoch": 2.509605746742399,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043675399803077837,
      "loss": 4.7723,
      "mean_token_accuracy": 0.21978584229946135,
      "num_tokens": 55265051.0,
      "step": 30045
    },
    {
      "entropy": 4.992791080474854,
      "epoch": 2.510023387905112,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004367333033064275,
      "loss": 4.6549,
      "mean_token_accuracy": 0.23272036612033845,
      "num_tokens": 55274148.0,
      "step": 30050
    },
    {
      "entropy": 5.0183876037597654,
      "epoch": 2.5104410290678247,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004367126057507405,
      "loss": 4.7277,
      "mean_token_accuracy": 0.21677012145519256,
      "num_tokens": 55282841.0,
      "step": 30055
    },
    {
      "entropy": 5.146202468872071,
      "epoch": 2.510858670230538,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004366919053640797,
      "loss": 4.8509,
      "mean_token_accuracy": 0.21793281137943268,
      "num_tokens": 55293136.0,
      "step": 30060
    },
    {
      "entropy": 5.110984754562378,
      "epoch": 2.511276311393251,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004366712021468076,
      "loss": 4.8188,
      "mean_token_accuracy": 0.22099236100912095,
      "num_tokens": 55303491.0,
      "step": 30065
    },
    {
      "entropy": 5.082115221023559,
      "epoch": 2.511693952555964,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043665049609928634,
      "loss": 4.7197,
      "mean_token_accuracy": 0.22070466876029968,
      "num_tokens": 55312746.0,
      "step": 30070
    },
    {
      "entropy": 4.992985773086548,
      "epoch": 2.512111593718677,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004366297872218786,
      "loss": 4.6668,
      "mean_token_accuracy": 0.22631352990865708,
      "num_tokens": 55322166.0,
      "step": 30075
    },
    {
      "entropy": 5.010161685943603,
      "epoch": 2.5125292348813897,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043660907551494696,
      "loss": 4.6646,
      "mean_token_accuracy": 0.23235988169908522,
      "num_tokens": 55331733.0,
      "step": 30080
    },
    {
      "entropy": 5.032168579101563,
      "epoch": 2.512946876044103,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043658836097885387,
      "loss": 4.7007,
      "mean_token_accuracy": 0.22712698131799697,
      "num_tokens": 55340941.0,
      "step": 30085
    },
    {
      "entropy": 4.99653902053833,
      "epoch": 2.5133645172068158,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043656764361396194,
      "loss": 4.5708,
      "mean_token_accuracy": 0.2329285502433777,
      "num_tokens": 55349570.0,
      "step": 30090
    },
    {
      "entropy": 4.979613256454468,
      "epoch": 2.513782158369529,
      "grad_norm": 1.03125,
      "learning_rate": 0.000436546923420634,
      "loss": 4.6946,
      "mean_token_accuracy": 0.22867047190666198,
      "num_tokens": 55359043.0,
      "step": 30095
    },
    {
      "entropy": 5.010455513000489,
      "epoch": 2.514199799532242,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043652620039923276,
      "loss": 4.7029,
      "mean_token_accuracy": 0.2300919309258461,
      "num_tokens": 55369034.0,
      "step": 30100
    },
    {
      "entropy": 5.010351276397705,
      "epoch": 2.5146174406949546,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004365054745501209,
      "loss": 4.662,
      "mean_token_accuracy": 0.22311634123325347,
      "num_tokens": 55378002.0,
      "step": 30105
    },
    {
      "entropy": 5.0607561588287355,
      "epoch": 2.515035081857668,
      "grad_norm": 1.125,
      "learning_rate": 0.00043648474587366134,
      "loss": 4.6886,
      "mean_token_accuracy": 0.21940386146306992,
      "num_tokens": 55387129.0,
      "step": 30110
    },
    {
      "entropy": 5.035100889205933,
      "epoch": 2.515452723020381,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004364640143702169,
      "loss": 4.7849,
      "mean_token_accuracy": 0.22787240147590637,
      "num_tokens": 55396452.0,
      "step": 30115
    },
    {
      "entropy": 4.99557580947876,
      "epoch": 2.515870364183094,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004364432800401505,
      "loss": 4.6451,
      "mean_token_accuracy": 0.2359452575445175,
      "num_tokens": 55405202.0,
      "step": 30120
    },
    {
      "entropy": 5.0065696239471436,
      "epoch": 2.5162880053458068,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043642254288382514,
      "loss": 4.6445,
      "mean_token_accuracy": 0.2323583722114563,
      "num_tokens": 55414075.0,
      "step": 30125
    },
    {
      "entropy": 5.001539325714111,
      "epoch": 2.51670564650852,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004364018029016039,
      "loss": 4.7349,
      "mean_token_accuracy": 0.22373124659061433,
      "num_tokens": 55423964.0,
      "step": 30130
    },
    {
      "entropy": 4.972276067733764,
      "epoch": 2.517123287671233,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004363810600938498,
      "loss": 4.6547,
      "mean_token_accuracy": 0.22575754970312117,
      "num_tokens": 55433594.0,
      "step": 30135
    },
    {
      "entropy": 5.079157829284668,
      "epoch": 2.517540928833946,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043636031446092605,
      "loss": 4.7372,
      "mean_token_accuracy": 0.22336320728063583,
      "num_tokens": 55443246.0,
      "step": 30140
    },
    {
      "entropy": 5.000527811050415,
      "epoch": 2.517958569996659,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043633956600319575,
      "loss": 4.6375,
      "mean_token_accuracy": 0.2285942628979683,
      "num_tokens": 55451973.0,
      "step": 30145
    },
    {
      "entropy": 5.085242128372192,
      "epoch": 2.5183762111593717,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043631881472102214,
      "loss": 4.771,
      "mean_token_accuracy": 0.22071921974420547,
      "num_tokens": 55461023.0,
      "step": 30150
    },
    {
      "entropy": 4.965851497650147,
      "epoch": 2.518793852322085,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043629806061476847,
      "loss": 4.6218,
      "mean_token_accuracy": 0.23555876314640045,
      "num_tokens": 55469433.0,
      "step": 30155
    },
    {
      "entropy": 5.014393091201782,
      "epoch": 2.519211493484798,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004362773036847981,
      "loss": 4.6836,
      "mean_token_accuracy": 0.22435836642980575,
      "num_tokens": 55479086.0,
      "step": 30160
    },
    {
      "entropy": 5.079253149032593,
      "epoch": 2.519629134647511,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043625654393147445,
      "loss": 4.7752,
      "mean_token_accuracy": 0.23180919289588928,
      "num_tokens": 55487787.0,
      "step": 30165
    },
    {
      "entropy": 5.044316816329956,
      "epoch": 2.520046775810224,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004362357813551608,
      "loss": 4.6907,
      "mean_token_accuracy": 0.2327631413936615,
      "num_tokens": 55496574.0,
      "step": 30170
    },
    {
      "entropy": 5.040924596786499,
      "epoch": 2.5204644169729367,
      "grad_norm": 1.125,
      "learning_rate": 0.00043621501595622077,
      "loss": 4.7334,
      "mean_token_accuracy": 0.2209316000342369,
      "num_tokens": 55505851.0,
      "step": 30175
    },
    {
      "entropy": 4.971868467330933,
      "epoch": 2.52088205813565,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004361942477350178,
      "loss": 4.6823,
      "mean_token_accuracy": 0.22353572398424149,
      "num_tokens": 55514597.0,
      "step": 30180
    },
    {
      "entropy": 5.083057451248169,
      "epoch": 2.5212996992983627,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004361734766919156,
      "loss": 4.7159,
      "mean_token_accuracy": 0.22003111541271209,
      "num_tokens": 55523467.0,
      "step": 30185
    },
    {
      "entropy": 5.061150407791137,
      "epoch": 2.521717340461076,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004361527028272776,
      "loss": 4.7457,
      "mean_token_accuracy": 0.221624718606472,
      "num_tokens": 55532786.0,
      "step": 30190
    },
    {
      "entropy": 5.100564813613891,
      "epoch": 2.522134981623789,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043613192614146753,
      "loss": 4.8512,
      "mean_token_accuracy": 0.21093617379665375,
      "num_tokens": 55541454.0,
      "step": 30195
    },
    {
      "entropy": 5.07801628112793,
      "epoch": 2.5225526227865016,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004361111466348492,
      "loss": 4.7283,
      "mean_token_accuracy": 0.2252323657274246,
      "num_tokens": 55550600.0,
      "step": 30200
    },
    {
      "entropy": 5.050239133834839,
      "epoch": 2.522970263949215,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043609036430778623,
      "loss": 4.7143,
      "mean_token_accuracy": 0.22340261489152907,
      "num_tokens": 55559413.0,
      "step": 30205
    },
    {
      "entropy": 4.979958438873291,
      "epoch": 2.5233879051119277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004360695791606426,
      "loss": 4.6434,
      "mean_token_accuracy": 0.22458267211914062,
      "num_tokens": 55568446.0,
      "step": 30210
    },
    {
      "entropy": 5.059675455093384,
      "epoch": 2.523805546274641,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004360487911937821,
      "loss": 4.7641,
      "mean_token_accuracy": 0.23188960999250413,
      "num_tokens": 55577343.0,
      "step": 30215
    },
    {
      "entropy": 4.957174587249756,
      "epoch": 2.5242231874373537,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004360280004075687,
      "loss": 4.6181,
      "mean_token_accuracy": 0.23442779779434203,
      "num_tokens": 55586202.0,
      "step": 30220
    },
    {
      "entropy": 5.018587398529053,
      "epoch": 2.5246408286000666,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004360072068023662,
      "loss": 4.718,
      "mean_token_accuracy": 0.22076638638973237,
      "num_tokens": 55595385.0,
      "step": 30225
    },
    {
      "entropy": 5.068341302871704,
      "epoch": 2.52505846976278,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043598641037853885,
      "loss": 4.6909,
      "mean_token_accuracy": 0.2290903478860855,
      "num_tokens": 55604118.0,
      "step": 30230
    },
    {
      "entropy": 5.038752794265747,
      "epoch": 2.525476110925493,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043596561113645055,
      "loss": 4.506,
      "mean_token_accuracy": 0.2436787396669388,
      "num_tokens": 55613315.0,
      "step": 30235
    },
    {
      "entropy": 4.966917562484741,
      "epoch": 2.525893752088206,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004359448090764655,
      "loss": 4.6853,
      "mean_token_accuracy": 0.22591643631458283,
      "num_tokens": 55623509.0,
      "step": 30240
    },
    {
      "entropy": 4.954662609100342,
      "epoch": 2.5263113932509187,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043592400419894785,
      "loss": 4.6056,
      "mean_token_accuracy": 0.23749227374792098,
      "num_tokens": 55632299.0,
      "step": 30245
    },
    {
      "entropy": 5.051137828826905,
      "epoch": 2.526729034413632,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004359031965042617,
      "loss": 4.7341,
      "mean_token_accuracy": 0.22960255593061446,
      "num_tokens": 55641001.0,
      "step": 30250
    },
    {
      "entropy": 4.984944105148315,
      "epoch": 2.5271466755763448,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043588238599277156,
      "loss": 4.712,
      "mean_token_accuracy": 0.22302065789699554,
      "num_tokens": 55649366.0,
      "step": 30255
    },
    {
      "entropy": 4.988801193237305,
      "epoch": 2.527564316739058,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043586157266484155,
      "loss": 4.6388,
      "mean_token_accuracy": 0.23381123542785645,
      "num_tokens": 55658543.0,
      "step": 30260
    },
    {
      "entropy": 5.063041496276855,
      "epoch": 2.527981957901771,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004358407565208361,
      "loss": 4.7789,
      "mean_token_accuracy": 0.22224838882684708,
      "num_tokens": 55668039.0,
      "step": 30265
    },
    {
      "entropy": 5.013068914413452,
      "epoch": 2.5283995990644836,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043581993756111956,
      "loss": 4.6269,
      "mean_token_accuracy": 0.22790263295173646,
      "num_tokens": 55676941.0,
      "step": 30270
    },
    {
      "entropy": 5.1644023895263675,
      "epoch": 2.528817240227197,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004357991157860565,
      "loss": 4.8107,
      "mean_token_accuracy": 0.2224975660443306,
      "num_tokens": 55686703.0,
      "step": 30275
    },
    {
      "entropy": 5.081228971481323,
      "epoch": 2.5292348813899097,
      "grad_norm": 1.125,
      "learning_rate": 0.0004357782911960113,
      "loss": 4.7191,
      "mean_token_accuracy": 0.22059297561645508,
      "num_tokens": 55696349.0,
      "step": 30280
    },
    {
      "entropy": 5.027966547012329,
      "epoch": 2.529652522552623,
      "grad_norm": 1.125,
      "learning_rate": 0.0004357574637913486,
      "loss": 4.7695,
      "mean_token_accuracy": 0.22240751534700393,
      "num_tokens": 55704644.0,
      "step": 30285
    },
    {
      "entropy": 5.082309818267822,
      "epoch": 2.5300701637153358,
      "grad_norm": 1.1875,
      "learning_rate": 0.000435736633572433,
      "loss": 4.6609,
      "mean_token_accuracy": 0.2391667291522026,
      "num_tokens": 55713783.0,
      "step": 30290
    },
    {
      "entropy": 4.977870082855224,
      "epoch": 2.5304878048780486,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004357158005396291,
      "loss": 4.6185,
      "mean_token_accuracy": 0.23387140035629272,
      "num_tokens": 55723210.0,
      "step": 30295
    },
    {
      "entropy": 4.877847957611084,
      "epoch": 2.530905446040762,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043569496469330173,
      "loss": 4.5175,
      "mean_token_accuracy": 0.2409648209810257,
      "num_tokens": 55731152.0,
      "step": 30300
    },
    {
      "entropy": 5.076578664779663,
      "epoch": 2.5313230872034747,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004356741260338156,
      "loss": 4.7779,
      "mean_token_accuracy": 0.21354439109563828,
      "num_tokens": 55740321.0,
      "step": 30305
    },
    {
      "entropy": 4.930563354492188,
      "epoch": 2.531740728366188,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043565328456153534,
      "loss": 4.5683,
      "mean_token_accuracy": 0.23789604306221007,
      "num_tokens": 55748398.0,
      "step": 30310
    },
    {
      "entropy": 4.996127033233643,
      "epoch": 2.5321583695289007,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004356324402768261,
      "loss": 4.6302,
      "mean_token_accuracy": 0.227285797894001,
      "num_tokens": 55757357.0,
      "step": 30315
    },
    {
      "entropy": 5.075280284881591,
      "epoch": 2.5325760106916135,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004356115931800526,
      "loss": 4.786,
      "mean_token_accuracy": 0.22199821323156357,
      "num_tokens": 55765874.0,
      "step": 30320
    },
    {
      "entropy": 5.085766744613648,
      "epoch": 2.532993651854327,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004355907432715798,
      "loss": 4.8141,
      "mean_token_accuracy": 0.21925319284200667,
      "num_tokens": 55774569.0,
      "step": 30325
    },
    {
      "entropy": 5.016178274154663,
      "epoch": 2.5334112930170396,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043556989055177275,
      "loss": 4.6473,
      "mean_token_accuracy": 0.22813538014888762,
      "num_tokens": 55784829.0,
      "step": 30330
    },
    {
      "entropy": 5.014798307418824,
      "epoch": 2.533828934179753,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043554903502099644,
      "loss": 4.6736,
      "mean_token_accuracy": 0.22651979625225066,
      "num_tokens": 55793393.0,
      "step": 30335
    },
    {
      "entropy": 5.003944969177246,
      "epoch": 2.5342465753424657,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043552817667961616,
      "loss": 4.6982,
      "mean_token_accuracy": 0.23076115250587464,
      "num_tokens": 55802988.0,
      "step": 30340
    },
    {
      "entropy": 5.114545392990112,
      "epoch": 2.5346642165051785,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043550731552799676,
      "loss": 4.7798,
      "mean_token_accuracy": 0.22238215059041977,
      "num_tokens": 55812698.0,
      "step": 30345
    },
    {
      "entropy": 5.032493209838867,
      "epoch": 2.5350818576678917,
      "grad_norm": 1.125,
      "learning_rate": 0.00043548645156650376,
      "loss": 4.7242,
      "mean_token_accuracy": 0.22101350128650665,
      "num_tokens": 55821752.0,
      "step": 30350
    },
    {
      "entropy": 5.102636480331421,
      "epoch": 2.535499498830605,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004354655847955021,
      "loss": 4.7805,
      "mean_token_accuracy": 0.222718046605587,
      "num_tokens": 55830707.0,
      "step": 30355
    },
    {
      "entropy": 5.08576135635376,
      "epoch": 2.535917139993318,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043544471521535726,
      "loss": 4.8289,
      "mean_token_accuracy": 0.21732427030801774,
      "num_tokens": 55840816.0,
      "step": 30360
    },
    {
      "entropy": 5.072657060623169,
      "epoch": 2.5363347811560306,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043542384282643467,
      "loss": 4.7311,
      "mean_token_accuracy": 0.2264278307557106,
      "num_tokens": 55849462.0,
      "step": 30365
    },
    {
      "entropy": 4.98287296295166,
      "epoch": 2.536752422318744,
      "grad_norm": 1.125,
      "learning_rate": 0.0004354029676290995,
      "loss": 4.6661,
      "mean_token_accuracy": 0.2337902382016182,
      "num_tokens": 55858420.0,
      "step": 30370
    },
    {
      "entropy": 5.114178133010864,
      "epoch": 2.5371700634814567,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004353820896237173,
      "loss": 4.7848,
      "mean_token_accuracy": 0.2195681795477867,
      "num_tokens": 55868576.0,
      "step": 30375
    },
    {
      "entropy": 4.964525318145752,
      "epoch": 2.53758770464417,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004353612088106536,
      "loss": 4.6349,
      "mean_token_accuracy": 0.23192695379257203,
      "num_tokens": 55877348.0,
      "step": 30380
    },
    {
      "entropy": 5.033477640151977,
      "epoch": 2.5380053458068828,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043534032519027395,
      "loss": 4.6984,
      "mean_token_accuracy": 0.22675966024398803,
      "num_tokens": 55887016.0,
      "step": 30385
    },
    {
      "entropy": 5.0648046970367435,
      "epoch": 2.5384229869695956,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004353194387629438,
      "loss": 4.7078,
      "mean_token_accuracy": 0.2317677065730095,
      "num_tokens": 55896445.0,
      "step": 30390
    },
    {
      "entropy": 5.07045431137085,
      "epoch": 2.538840628132309,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000435298549529029,
      "loss": 4.7011,
      "mean_token_accuracy": 0.22456980496644974,
      "num_tokens": 55905122.0,
      "step": 30395
    },
    {
      "entropy": 4.982398891448975,
      "epoch": 2.5392582692950216,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043527765748889506,
      "loss": 4.7404,
      "mean_token_accuracy": 0.22193360775709153,
      "num_tokens": 55914759.0,
      "step": 30400
    },
    {
      "entropy": 5.066453695297241,
      "epoch": 2.539675910457735,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004352567626429079,
      "loss": 4.7273,
      "mean_token_accuracy": 0.2165827989578247,
      "num_tokens": 55925884.0,
      "step": 30405
    },
    {
      "entropy": 5.06363434791565,
      "epoch": 2.5400935516204477,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004352358649914331,
      "loss": 4.6982,
      "mean_token_accuracy": 0.22881697714328766,
      "num_tokens": 55935199.0,
      "step": 30410
    },
    {
      "entropy": 4.985807180404663,
      "epoch": 2.5405111927831605,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043521496453483667,
      "loss": 4.6606,
      "mean_token_accuracy": 0.23790544420480728,
      "num_tokens": 55943351.0,
      "step": 30415
    },
    {
      "entropy": 5.023849725723267,
      "epoch": 2.5409288339458738,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043519406127348435,
      "loss": 4.7025,
      "mean_token_accuracy": 0.22961650639772416,
      "num_tokens": 55951715.0,
      "step": 30420
    },
    {
      "entropy": 5.026906538009643,
      "epoch": 2.5413464751085866,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004351731552077423,
      "loss": 4.6577,
      "mean_token_accuracy": 0.2278609275817871,
      "num_tokens": 55960395.0,
      "step": 30425
    },
    {
      "entropy": 4.8656140804290775,
      "epoch": 2.5417641162713,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004351522463379763,
      "loss": 4.5417,
      "mean_token_accuracy": 0.23874365389347077,
      "num_tokens": 55969386.0,
      "step": 30430
    },
    {
      "entropy": 4.981404399871826,
      "epoch": 2.5421817574340126,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004351313346645525,
      "loss": 4.6605,
      "mean_token_accuracy": 0.22866163551807403,
      "num_tokens": 55978727.0,
      "step": 30435
    },
    {
      "entropy": 5.057233476638794,
      "epoch": 2.5425993985967255,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004351104201878368,
      "loss": 4.6963,
      "mean_token_accuracy": 0.2239602416753769,
      "num_tokens": 55987273.0,
      "step": 30440
    },
    {
      "entropy": 5.017791604995727,
      "epoch": 2.5430170397594387,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004350895029081956,
      "loss": 4.7481,
      "mean_token_accuracy": 0.22705432921648025,
      "num_tokens": 55996601.0,
      "step": 30445
    },
    {
      "entropy": 5.072534704208374,
      "epoch": 2.5434346809221515,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043506858282599487,
      "loss": 4.7368,
      "mean_token_accuracy": 0.21672551929950715,
      "num_tokens": 56005952.0,
      "step": 30450
    },
    {
      "entropy": 5.028097915649414,
      "epoch": 2.543852322084865,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000435047659941601,
      "loss": 4.7629,
      "mean_token_accuracy": 0.21996112763881684,
      "num_tokens": 56015197.0,
      "step": 30455
    },
    {
      "entropy": 5.0263587474823,
      "epoch": 2.5442699632475776,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043502673425538016,
      "loss": 4.6379,
      "mean_token_accuracy": 0.22777758836746215,
      "num_tokens": 56023662.0,
      "step": 30460
    },
    {
      "entropy": 5.086950016021729,
      "epoch": 2.5446876044102904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043500580576769874,
      "loss": 4.7987,
      "mean_token_accuracy": 0.2216857209801674,
      "num_tokens": 56033414.0,
      "step": 30465
    },
    {
      "entropy": 4.975443553924561,
      "epoch": 2.5451052455730037,
      "grad_norm": 1.140625,
      "learning_rate": 0.000434984874478923,
      "loss": 4.5664,
      "mean_token_accuracy": 0.235675410926342,
      "num_tokens": 56042425.0,
      "step": 30470
    },
    {
      "entropy": 4.992159080505371,
      "epoch": 2.545522886735717,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043496394038941956,
      "loss": 4.6663,
      "mean_token_accuracy": 0.22382782846689225,
      "num_tokens": 56051858.0,
      "step": 30475
    },
    {
      "entropy": 4.931417036056518,
      "epoch": 2.5459405278984297,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043494300349955473,
      "loss": 4.5805,
      "mean_token_accuracy": 0.23814429193735123,
      "num_tokens": 56060903.0,
      "step": 30480
    },
    {
      "entropy": 4.993144702911377,
      "epoch": 2.5463581690611425,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004349220638096952,
      "loss": 4.6437,
      "mean_token_accuracy": 0.232941110432148,
      "num_tokens": 56069365.0,
      "step": 30485
    },
    {
      "entropy": 4.998958253860474,
      "epoch": 2.546775810223856,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004349011213202073,
      "loss": 4.6128,
      "mean_token_accuracy": 0.23178038001060486,
      "num_tokens": 56078387.0,
      "step": 30490
    },
    {
      "entropy": 4.905283355712891,
      "epoch": 2.5471934513865686,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043488017603145787,
      "loss": 4.6337,
      "mean_token_accuracy": 0.22724654376506806,
      "num_tokens": 56087696.0,
      "step": 30495
    },
    {
      "entropy": 5.011586141586304,
      "epoch": 2.547611092549282,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043485922794381357,
      "loss": 4.6708,
      "mean_token_accuracy": 0.22404577881097792,
      "num_tokens": 56097064.0,
      "step": 30500
    },
    {
      "entropy": 4.9659095287323,
      "epoch": 2.5480287337119947,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043483827705764094,
      "loss": 4.638,
      "mean_token_accuracy": 0.22843272089958191,
      "num_tokens": 56106395.0,
      "step": 30505
    },
    {
      "entropy": 5.013437414169312,
      "epoch": 2.5484463748747075,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043481732337330697,
      "loss": 4.7085,
      "mean_token_accuracy": 0.22437625676393508,
      "num_tokens": 56115642.0,
      "step": 30510
    },
    {
      "entropy": 5.011914587020874,
      "epoch": 2.5488640160374207,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043479636689117834,
      "loss": 4.6799,
      "mean_token_accuracy": 0.23024199157953262,
      "num_tokens": 56125369.0,
      "step": 30515
    },
    {
      "entropy": 5.044667768478393,
      "epoch": 2.5492816572001336,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043477540761162196,
      "loss": 4.7245,
      "mean_token_accuracy": 0.21802729219198227,
      "num_tokens": 56134408.0,
      "step": 30520
    },
    {
      "entropy": 4.888294506072998,
      "epoch": 2.549699298362847,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004347544455350047,
      "loss": 4.5377,
      "mean_token_accuracy": 0.23377467393875123,
      "num_tokens": 56143454.0,
      "step": 30525
    },
    {
      "entropy": 5.032129335403442,
      "epoch": 2.5501169395255596,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043473348066169363,
      "loss": 4.6758,
      "mean_token_accuracy": 0.22602162808179854,
      "num_tokens": 56152239.0,
      "step": 30530
    },
    {
      "entropy": 4.9801530838012695,
      "epoch": 2.5505345806882724,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004347125129920557,
      "loss": 4.6597,
      "mean_token_accuracy": 0.22525801807641982,
      "num_tokens": 56161299.0,
      "step": 30535
    },
    {
      "entropy": 5.011531162261963,
      "epoch": 2.5509522218509857,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004346915425264581,
      "loss": 4.6692,
      "mean_token_accuracy": 0.22632636725902558,
      "num_tokens": 56169961.0,
      "step": 30540
    },
    {
      "entropy": 4.959675598144531,
      "epoch": 2.5513698630136985,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004346705692652677,
      "loss": 4.6109,
      "mean_token_accuracy": 0.23716597110033036,
      "num_tokens": 56179189.0,
      "step": 30545
    },
    {
      "entropy": 4.966413497924805,
      "epoch": 2.5517875041764118,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004346495932088518,
      "loss": 4.6591,
      "mean_token_accuracy": 0.22635245621204375,
      "num_tokens": 56188643.0,
      "step": 30550
    },
    {
      "entropy": 5.0899358749389645,
      "epoch": 2.5522051453391246,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004346286143575776,
      "loss": 4.7227,
      "mean_token_accuracy": 0.2307007372379303,
      "num_tokens": 56198084.0,
      "step": 30555
    },
    {
      "entropy": 4.984431171417237,
      "epoch": 2.5526227865018374,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043460763271181235,
      "loss": 4.6351,
      "mean_token_accuracy": 0.23701891899108887,
      "num_tokens": 56206773.0,
      "step": 30560
    },
    {
      "entropy": 4.954773092269898,
      "epoch": 2.5530404276645506,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043458664827192333,
      "loss": 4.7285,
      "mean_token_accuracy": 0.22035176157951356,
      "num_tokens": 56216381.0,
      "step": 30565
    },
    {
      "entropy": 5.0194073677062985,
      "epoch": 2.5534580688272634,
      "grad_norm": 1.125,
      "learning_rate": 0.0004345656610382781,
      "loss": 4.6677,
      "mean_token_accuracy": 0.2323532909154892,
      "num_tokens": 56224963.0,
      "step": 30570
    },
    {
      "entropy": 5.065360736846924,
      "epoch": 2.5538757099899767,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004345446710112438,
      "loss": 4.7573,
      "mean_token_accuracy": 0.22134315073490143,
      "num_tokens": 56234239.0,
      "step": 30575
    },
    {
      "entropy": 5.036963844299317,
      "epoch": 2.5542933511526895,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043452367819118797,
      "loss": 4.74,
      "mean_token_accuracy": 0.21748898178339005,
      "num_tokens": 56243689.0,
      "step": 30580
    },
    {
      "entropy": 4.982497978210449,
      "epoch": 2.5547109923154023,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043450268257847814,
      "loss": 4.6017,
      "mean_token_accuracy": 0.24098596572875977,
      "num_tokens": 56252262.0,
      "step": 30585
    },
    {
      "entropy": 5.059906816482544,
      "epoch": 2.5551286334781156,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004344816841734819,
      "loss": 4.6884,
      "mean_token_accuracy": 0.2302256092429161,
      "num_tokens": 56261012.0,
      "step": 30590
    },
    {
      "entropy": 5.035729694366455,
      "epoch": 2.555546274640829,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004344606829765668,
      "loss": 4.7143,
      "mean_token_accuracy": 0.22032633572816848,
      "num_tokens": 56271324.0,
      "step": 30595
    },
    {
      "entropy": 5.00416316986084,
      "epoch": 2.5559639158035417,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004344396789881005,
      "loss": 4.659,
      "mean_token_accuracy": 0.2230440631508827,
      "num_tokens": 56280007.0,
      "step": 30600
    },
    {
      "entropy": 4.9707365989685055,
      "epoch": 2.5563815569662545,
      "grad_norm": 1.125,
      "learning_rate": 0.0004344186722084507,
      "loss": 4.5962,
      "mean_token_accuracy": 0.2338648945093155,
      "num_tokens": 56288939.0,
      "step": 30605
    },
    {
      "entropy": 4.886863374710083,
      "epoch": 2.5567991981289677,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043439766263798506,
      "loss": 4.572,
      "mean_token_accuracy": 0.2451679915189743,
      "num_tokens": 56297628.0,
      "step": 30610
    },
    {
      "entropy": 4.928222036361694,
      "epoch": 2.5572168392916805,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043437665027707155,
      "loss": 4.5831,
      "mean_token_accuracy": 0.2328586608171463,
      "num_tokens": 56306009.0,
      "step": 30615
    },
    {
      "entropy": 5.021965789794922,
      "epoch": 2.557634480454394,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043435563512607796,
      "loss": 4.7168,
      "mean_token_accuracy": 0.21298278272151946,
      "num_tokens": 56314587.0,
      "step": 30620
    },
    {
      "entropy": 5.105061674118042,
      "epoch": 2.5580521216171066,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004343346171853722,
      "loss": 4.7935,
      "mean_token_accuracy": 0.2174234002828598,
      "num_tokens": 56323754.0,
      "step": 30625
    },
    {
      "entropy": 5.064725017547607,
      "epoch": 2.5584697627798194,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004343135964553221,
      "loss": 4.7151,
      "mean_token_accuracy": 0.22864601612091065,
      "num_tokens": 56332406.0,
      "step": 30630
    },
    {
      "entropy": 5.0246223449707035,
      "epoch": 2.5588874039425327,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004342925729362957,
      "loss": 4.6809,
      "mean_token_accuracy": 0.22325865924358368,
      "num_tokens": 56341288.0,
      "step": 30635
    },
    {
      "entropy": 4.95399317741394,
      "epoch": 2.5593050451052455,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004342715466286611,
      "loss": 4.5901,
      "mean_token_accuracy": 0.23619039505720138,
      "num_tokens": 56349054.0,
      "step": 30640
    },
    {
      "entropy": 4.997623252868652,
      "epoch": 2.5597226862679587,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043425051753278633,
      "loss": 4.7575,
      "mean_token_accuracy": 0.22802478522062303,
      "num_tokens": 56357967.0,
      "step": 30645
    },
    {
      "entropy": 5.016992044448853,
      "epoch": 2.5601403274306715,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004342294856490397,
      "loss": 4.6082,
      "mean_token_accuracy": 0.23291121274232865,
      "num_tokens": 56366338.0,
      "step": 30650
    },
    {
      "entropy": 4.999181461334229,
      "epoch": 2.5605579685933844,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043420845097778913,
      "loss": 4.6788,
      "mean_token_accuracy": 0.226304529607296,
      "num_tokens": 56375779.0,
      "step": 30655
    },
    {
      "entropy": 4.985859632492065,
      "epoch": 2.5609756097560976,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000434187413519403,
      "loss": 4.6935,
      "mean_token_accuracy": 0.2319295436143875,
      "num_tokens": 56385454.0,
      "step": 30660
    },
    {
      "entropy": 4.981514883041382,
      "epoch": 2.5613932509188104,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043416637327424955,
      "loss": 4.6442,
      "mean_token_accuracy": 0.23058406561613082,
      "num_tokens": 56393834.0,
      "step": 30665
    },
    {
      "entropy": 4.990794324874878,
      "epoch": 2.5618108920815237,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004341453302426972,
      "loss": 4.6864,
      "mean_token_accuracy": 0.2277348279953003,
      "num_tokens": 56402959.0,
      "step": 30670
    },
    {
      "entropy": 5.000248146057129,
      "epoch": 2.5622285332442365,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004341242844251143,
      "loss": 4.6329,
      "mean_token_accuracy": 0.22256070077419282,
      "num_tokens": 56412468.0,
      "step": 30675
    },
    {
      "entropy": 4.987673711776734,
      "epoch": 2.5626461744069493,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004341032358218692,
      "loss": 4.7372,
      "mean_token_accuracy": 0.22306836843490602,
      "num_tokens": 56421312.0,
      "step": 30680
    },
    {
      "entropy": 5.015114641189575,
      "epoch": 2.5630638155696626,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043408218443333056,
      "loss": 4.6634,
      "mean_token_accuracy": 0.23166847974061966,
      "num_tokens": 56430248.0,
      "step": 30685
    },
    {
      "entropy": 4.960431146621704,
      "epoch": 2.563481456732376,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043406113025986666,
      "loss": 4.5632,
      "mean_token_accuracy": 0.23593445867300034,
      "num_tokens": 56438895.0,
      "step": 30690
    },
    {
      "entropy": 4.955180549621582,
      "epoch": 2.5638990978950886,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004340400733018463,
      "loss": 4.6016,
      "mean_token_accuracy": 0.23630058020353317,
      "num_tokens": 56448229.0,
      "step": 30695
    },
    {
      "entropy": 4.978836441040039,
      "epoch": 2.5643167390578014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000434019013559638,
      "loss": 4.6284,
      "mean_token_accuracy": 0.22960177958011627,
      "num_tokens": 56457017.0,
      "step": 30700
    },
    {
      "entropy": 4.987419700622558,
      "epoch": 2.5647343802205147,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043399795103361046,
      "loss": 4.6735,
      "mean_token_accuracy": 0.2264266863465309,
      "num_tokens": 56466214.0,
      "step": 30705
    },
    {
      "entropy": 5.010708236694336,
      "epoch": 2.5651520213832275,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043397688572413236,
      "loss": 4.7569,
      "mean_token_accuracy": 0.2243591859936714,
      "num_tokens": 56475613.0,
      "step": 30710
    },
    {
      "entropy": 4.981833124160767,
      "epoch": 2.5655696625459408,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043395581763157256,
      "loss": 4.6122,
      "mean_token_accuracy": 0.2317723661661148,
      "num_tokens": 56484137.0,
      "step": 30715
    },
    {
      "entropy": 4.982231092453003,
      "epoch": 2.5659873037086536,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043393474675629983,
      "loss": 4.6316,
      "mean_token_accuracy": 0.2357809752225876,
      "num_tokens": 56492903.0,
      "step": 30720
    },
    {
      "entropy": 5.023267078399658,
      "epoch": 2.5664049448713664,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043391367309868304,
      "loss": 4.6798,
      "mean_token_accuracy": 0.22369183450937272,
      "num_tokens": 56503025.0,
      "step": 30725
    },
    {
      "entropy": 4.986454725265503,
      "epoch": 2.5668225860340796,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043389259665909107,
      "loss": 4.6591,
      "mean_token_accuracy": 0.23153338879346846,
      "num_tokens": 56512288.0,
      "step": 30730
    },
    {
      "entropy": 5.063553810119629,
      "epoch": 2.5672402271967925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000433871517437893,
      "loss": 4.7266,
      "mean_token_accuracy": 0.2298281818628311,
      "num_tokens": 56521636.0,
      "step": 30735
    },
    {
      "entropy": 4.952017021179199,
      "epoch": 2.5676578683595057,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043385043543545777,
      "loss": 4.6098,
      "mean_token_accuracy": 0.23795831054449082,
      "num_tokens": 56530364.0,
      "step": 30740
    },
    {
      "entropy": 4.895663070678711,
      "epoch": 2.5680755095222185,
      "grad_norm": 1.0,
      "learning_rate": 0.00043382935065215444,
      "loss": 4.518,
      "mean_token_accuracy": 0.23650759607553482,
      "num_tokens": 56539672.0,
      "step": 30745
    },
    {
      "entropy": 5.0224462985992435,
      "epoch": 2.5684931506849313,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043380826308835226,
      "loss": 4.7504,
      "mean_token_accuracy": 0.21867417991161348,
      "num_tokens": 56549620.0,
      "step": 30750
    },
    {
      "entropy": 5.047282934188843,
      "epoch": 2.5689107918476446,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043378717274442015,
      "loss": 4.7573,
      "mean_token_accuracy": 0.22731243818998337,
      "num_tokens": 56558615.0,
      "step": 30755
    },
    {
      "entropy": 4.944527053833008,
      "epoch": 2.5693284330103574,
      "grad_norm": 1.125,
      "learning_rate": 0.00043376607962072747,
      "loss": 4.6035,
      "mean_token_accuracy": 0.24000087827444078,
      "num_tokens": 56568011.0,
      "step": 30760
    },
    {
      "entropy": 5.024690055847168,
      "epoch": 2.5697460741730707,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043374498371764344,
      "loss": 4.7072,
      "mean_token_accuracy": 0.22278545498847963,
      "num_tokens": 56577100.0,
      "step": 30765
    },
    {
      "entropy": 5.065305376052857,
      "epoch": 2.5701637153357835,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043372388503553743,
      "loss": 4.732,
      "mean_token_accuracy": 0.22573486119508743,
      "num_tokens": 56586707.0,
      "step": 30770
    },
    {
      "entropy": 5.0716972827911375,
      "epoch": 2.5705813564984963,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043370278357477874,
      "loss": 4.8264,
      "mean_token_accuracy": 0.2169950544834137,
      "num_tokens": 56595685.0,
      "step": 30775
    },
    {
      "entropy": 5.037875652313232,
      "epoch": 2.5709989976612095,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043368167933573675,
      "loss": 4.5789,
      "mean_token_accuracy": 0.2391623690724373,
      "num_tokens": 56604474.0,
      "step": 30780
    },
    {
      "entropy": 4.997287178039551,
      "epoch": 2.5714166388239224,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000433660572318781,
      "loss": 4.6375,
      "mean_token_accuracy": 0.2320246919989586,
      "num_tokens": 56612699.0,
      "step": 30785
    },
    {
      "entropy": 5.025967264175415,
      "epoch": 2.5718342799866356,
      "grad_norm": 1.078125,
      "learning_rate": 0.000433639462524281,
      "loss": 4.7861,
      "mean_token_accuracy": 0.21962717771530152,
      "num_tokens": 56622145.0,
      "step": 30790
    },
    {
      "entropy": 4.981334829330445,
      "epoch": 2.5722519211493484,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004336183499526062,
      "loss": 4.6462,
      "mean_token_accuracy": 0.23755952417850495,
      "num_tokens": 56630716.0,
      "step": 30795
    },
    {
      "entropy": 4.978836965560913,
      "epoch": 2.5726695623120612,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043359723460412625,
      "loss": 4.6397,
      "mean_token_accuracy": 0.2412180259823799,
      "num_tokens": 56640006.0,
      "step": 30800
    },
    {
      "entropy": 5.1074906349182125,
      "epoch": 2.5730872034747745,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004335761164792108,
      "loss": 4.7281,
      "mean_token_accuracy": 0.22385169565677643,
      "num_tokens": 56649286.0,
      "step": 30805
    },
    {
      "entropy": 4.9472638130187985,
      "epoch": 2.5735048446374877,
      "grad_norm": 1.125,
      "learning_rate": 0.0004335549955782296,
      "loss": 4.6063,
      "mean_token_accuracy": 0.23364781141281127,
      "num_tokens": 56658444.0,
      "step": 30810
    },
    {
      "entropy": 4.9689497470855715,
      "epoch": 2.5739224858002006,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004335338719015523,
      "loss": 4.6947,
      "mean_token_accuracy": 0.22488323152065276,
      "num_tokens": 56668359.0,
      "step": 30815
    },
    {
      "entropy": 4.999911308288574,
      "epoch": 2.5743401269629134,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004335127454495488,
      "loss": 4.7329,
      "mean_token_accuracy": 0.22523049265146255,
      "num_tokens": 56677504.0,
      "step": 30820
    },
    {
      "entropy": 4.990752410888672,
      "epoch": 2.5747577681256266,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004334916162225889,
      "loss": 4.6701,
      "mean_token_accuracy": 0.23661897480487823,
      "num_tokens": 56686684.0,
      "step": 30825
    },
    {
      "entropy": 5.023148059844971,
      "epoch": 2.5751754092883394,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004334704842210424,
      "loss": 4.7255,
      "mean_token_accuracy": 0.22134305089712142,
      "num_tokens": 56695394.0,
      "step": 30830
    },
    {
      "entropy": 5.081557178497315,
      "epoch": 2.5755930504510527,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004334493494452794,
      "loss": 4.6989,
      "mean_token_accuracy": 0.2252110868692398,
      "num_tokens": 56705275.0,
      "step": 30835
    },
    {
      "entropy": 5.049327325820923,
      "epoch": 2.5760106916137655,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004334282118956697,
      "loss": 4.7429,
      "mean_token_accuracy": 0.22478724271059036,
      "num_tokens": 56715301.0,
      "step": 30840
    },
    {
      "entropy": 5.06113109588623,
      "epoch": 2.5764283327764783,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004334070715725836,
      "loss": 4.7108,
      "mean_token_accuracy": 0.22436042726039887,
      "num_tokens": 56725319.0,
      "step": 30845
    },
    {
      "entropy": 5.092880630493164,
      "epoch": 2.5768459739391916,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043338592847639093,
      "loss": 4.7611,
      "mean_token_accuracy": 0.22770674675703048,
      "num_tokens": 56734599.0,
      "step": 30850
    },
    {
      "entropy": 4.948880195617676,
      "epoch": 2.5772636151019044,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004333647826074619,
      "loss": 4.6181,
      "mean_token_accuracy": 0.23484479635953903,
      "num_tokens": 56743704.0,
      "step": 30855
    },
    {
      "entropy": 5.09867901802063,
      "epoch": 2.5776812562646176,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004333436339661668,
      "loss": 4.7424,
      "mean_token_accuracy": 0.2311744436621666,
      "num_tokens": 56753260.0,
      "step": 30860
    },
    {
      "entropy": 5.033992195129395,
      "epoch": 2.5780988974273304,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043332248255287584,
      "loss": 4.6969,
      "mean_token_accuracy": 0.226128651201725,
      "num_tokens": 56763139.0,
      "step": 30865
    },
    {
      "entropy": 4.9606434345245365,
      "epoch": 2.5785165385900433,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043330132836795916,
      "loss": 4.6779,
      "mean_token_accuracy": 0.23448651880025864,
      "num_tokens": 56772595.0,
      "step": 30870
    },
    {
      "entropy": 5.015968465805054,
      "epoch": 2.5789341797527565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043328017141178724,
      "loss": 4.7171,
      "mean_token_accuracy": 0.2287769392132759,
      "num_tokens": 56782297.0,
      "step": 30875
    },
    {
      "entropy": 5.03973879814148,
      "epoch": 2.5793518209154693,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043325901168473035,
      "loss": 4.7129,
      "mean_token_accuracy": 0.22444235533475876,
      "num_tokens": 56792080.0,
      "step": 30880
    },
    {
      "entropy": 5.064612960815429,
      "epoch": 2.5797694620781826,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043323784918715896,
      "loss": 4.7602,
      "mean_token_accuracy": 0.22084080278873444,
      "num_tokens": 56802141.0,
      "step": 30885
    },
    {
      "entropy": 5.101237058639526,
      "epoch": 2.5801871032408954,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043321668391944364,
      "loss": 4.7249,
      "mean_token_accuracy": 0.22383217364549637,
      "num_tokens": 56811525.0,
      "step": 30890
    },
    {
      "entropy": 5.074109697341919,
      "epoch": 2.580604744403608,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004331955158819547,
      "loss": 4.7404,
      "mean_token_accuracy": 0.21919707357883453,
      "num_tokens": 56821052.0,
      "step": 30895
    },
    {
      "entropy": 4.958997392654419,
      "epoch": 2.5810223855663215,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043317434507506285,
      "loss": 4.558,
      "mean_token_accuracy": 0.2387296050786972,
      "num_tokens": 56830166.0,
      "step": 30900
    },
    {
      "entropy": 4.902561140060425,
      "epoch": 2.5814400267290343,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00043315317149913876,
      "loss": 4.591,
      "mean_token_accuracy": 0.23696701973676682,
      "num_tokens": 56839999.0,
      "step": 30905
    },
    {
      "entropy": 5.095736742019653,
      "epoch": 2.5818576678917475,
      "grad_norm": 1.0625,
      "learning_rate": 0.000433131995154553,
      "loss": 4.8134,
      "mean_token_accuracy": 0.21361175924539566,
      "num_tokens": 56850078.0,
      "step": 30910
    },
    {
      "entropy": 4.917168188095093,
      "epoch": 2.5822753090544603,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004331108160416764,
      "loss": 4.456,
      "mean_token_accuracy": 0.2441622197628021,
      "num_tokens": 56858148.0,
      "step": 30915
    },
    {
      "entropy": 5.011640739440918,
      "epoch": 2.582692950217173,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004330896341608795,
      "loss": 4.7463,
      "mean_token_accuracy": 0.2263013854622841,
      "num_tokens": 56867452.0,
      "step": 30920
    },
    {
      "entropy": 4.971128273010254,
      "epoch": 2.5831105913798864,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043306844951253336,
      "loss": 4.6285,
      "mean_token_accuracy": 0.23402538448572158,
      "num_tokens": 56876594.0,
      "step": 30925
    },
    {
      "entropy": 5.038482618331909,
      "epoch": 2.5835282325425997,
      "grad_norm": 1.125,
      "learning_rate": 0.00043304726209700874,
      "loss": 4.7041,
      "mean_token_accuracy": 0.22408445179462433,
      "num_tokens": 56885271.0,
      "step": 30930
    },
    {
      "entropy": 4.977520751953125,
      "epoch": 2.5839458737053125,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004330260719146765,
      "loss": 4.6701,
      "mean_token_accuracy": 0.2312503159046173,
      "num_tokens": 56894442.0,
      "step": 30935
    },
    {
      "entropy": 5.105996751785279,
      "epoch": 2.5843635148680253,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004330048789659077,
      "loss": 4.8142,
      "mean_token_accuracy": 0.21451411694288253,
      "num_tokens": 56903969.0,
      "step": 30940
    },
    {
      "entropy": 5.133035135269165,
      "epoch": 2.5847811560307385,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043298368325107335,
      "loss": 4.727,
      "mean_token_accuracy": 0.22011161148548125,
      "num_tokens": 56914168.0,
      "step": 30945
    },
    {
      "entropy": 4.95918459892273,
      "epoch": 2.5851987971934514,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004329624847705444,
      "loss": 4.6956,
      "mean_token_accuracy": 0.23241257965564727,
      "num_tokens": 56923040.0,
      "step": 30950
    },
    {
      "entropy": 4.971990013122559,
      "epoch": 2.5856164383561646,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043294128352469195,
      "loss": 4.6576,
      "mean_token_accuracy": 0.22773414403200148,
      "num_tokens": 56933352.0,
      "step": 30955
    },
    {
      "entropy": 5.064724922180176,
      "epoch": 2.5860340795188774,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043292007951388733,
      "loss": 4.7252,
      "mean_token_accuracy": 0.22045668065547944,
      "num_tokens": 56943678.0,
      "step": 30960
    },
    {
      "entropy": 5.027716541290284,
      "epoch": 2.5864517206815902,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043289887273850167,
      "loss": 4.7027,
      "mean_token_accuracy": 0.22318318784236907,
      "num_tokens": 56953469.0,
      "step": 30965
    },
    {
      "entropy": 4.994379758834839,
      "epoch": 2.5868693618443035,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000432877663198906,
      "loss": 4.755,
      "mean_token_accuracy": 0.2159707650542259,
      "num_tokens": 56962512.0,
      "step": 30970
    },
    {
      "entropy": 5.106261730194092,
      "epoch": 2.5872870030070163,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004328564508954719,
      "loss": 4.7164,
      "mean_token_accuracy": 0.22348330914974213,
      "num_tokens": 56972048.0,
      "step": 30975
    },
    {
      "entropy": 5.0327537059783936,
      "epoch": 2.5877046441697296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004328352358285706,
      "loss": 4.6436,
      "mean_token_accuracy": 0.23878000676631927,
      "num_tokens": 56981505.0,
      "step": 30980
    },
    {
      "entropy": 4.960415172576904,
      "epoch": 2.5881222853324424,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043281401799857356,
      "loss": 4.7286,
      "mean_token_accuracy": 0.22676901370286942,
      "num_tokens": 56990970.0,
      "step": 30985
    },
    {
      "entropy": 4.996281385421753,
      "epoch": 2.588539926495155,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004327927974058521,
      "loss": 4.7741,
      "mean_token_accuracy": 0.22176868021488189,
      "num_tokens": 57000776.0,
      "step": 30990
    },
    {
      "entropy": 5.024972343444825,
      "epoch": 2.5889575676578684,
      "grad_norm": 1.25,
      "learning_rate": 0.0004327715740507778,
      "loss": 4.5435,
      "mean_token_accuracy": 0.24957334101200104,
      "num_tokens": 57008677.0,
      "step": 30995
    },
    {
      "entropy": 4.982035398483276,
      "epoch": 2.5893752088205813,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004327503479337222,
      "loss": 4.7089,
      "mean_token_accuracy": 0.23157088905572892,
      "num_tokens": 57018229.0,
      "step": 31000
    },
    {
      "entropy": 4.9338303565979,
      "epoch": 2.5897928499832945,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043272911905505696,
      "loss": 4.6221,
      "mean_token_accuracy": 0.23077331781387328,
      "num_tokens": 57026814.0,
      "step": 31005
    },
    {
      "entropy": 4.970081853866577,
      "epoch": 2.5902104911460073,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004327078874151535,
      "loss": 4.6066,
      "mean_token_accuracy": 0.2288644477725029,
      "num_tokens": 57036213.0,
      "step": 31010
    },
    {
      "entropy": 5.022891521453857,
      "epoch": 2.59062813230872,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004326866530143838,
      "loss": 4.7613,
      "mean_token_accuracy": 0.22086622267961503,
      "num_tokens": 57045427.0,
      "step": 31015
    },
    {
      "entropy": 5.055294513702393,
      "epoch": 2.5910457734714334,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043266541585311924,
      "loss": 4.7083,
      "mean_token_accuracy": 0.2244478017091751,
      "num_tokens": 57055041.0,
      "step": 31020
    },
    {
      "entropy": 5.062456321716309,
      "epoch": 2.591463414634146,
      "grad_norm": 1.0,
      "learning_rate": 0.0004326441759317319,
      "loss": 4.623,
      "mean_token_accuracy": 0.23365342170000075,
      "num_tokens": 57065187.0,
      "step": 31025
    },
    {
      "entropy": 4.9776304244995115,
      "epoch": 2.5918810557968595,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004326229332505934,
      "loss": 4.6244,
      "mean_token_accuracy": 0.23490720689296724,
      "num_tokens": 57075601.0,
      "step": 31030
    },
    {
      "entropy": 4.919039869308472,
      "epoch": 2.5922986969595723,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004326016878100759,
      "loss": 4.6584,
      "mean_token_accuracy": 0.22454389184713364,
      "num_tokens": 57084968.0,
      "step": 31035
    },
    {
      "entropy": 4.974134302139282,
      "epoch": 2.592716338122285,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043258043961055104,
      "loss": 4.7447,
      "mean_token_accuracy": 0.2351214200258255,
      "num_tokens": 57093836.0,
      "step": 31040
    },
    {
      "entropy": 5.055296277999878,
      "epoch": 2.5931339792849983,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043255918865239097,
      "loss": 4.7617,
      "mean_token_accuracy": 0.22430101186037063,
      "num_tokens": 57103799.0,
      "step": 31045
    },
    {
      "entropy": 5.0953648567199705,
      "epoch": 2.5935516204477116,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043253793493596755,
      "loss": 4.719,
      "mean_token_accuracy": 0.22776830345392227,
      "num_tokens": 57113478.0,
      "step": 31050
    },
    {
      "entropy": 5.007707977294922,
      "epoch": 2.5939692616104244,
      "grad_norm": 1.046875,
      "learning_rate": 0.000432516678461653,
      "loss": 4.7012,
      "mean_token_accuracy": 0.2286467120051384,
      "num_tokens": 57122532.0,
      "step": 31055
    },
    {
      "entropy": 4.992282247543335,
      "epoch": 2.594386902773137,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004324954192298194,
      "loss": 4.6714,
      "mean_token_accuracy": 0.22801124006509782,
      "num_tokens": 57131159.0,
      "step": 31060
    },
    {
      "entropy": 5.002359867095947,
      "epoch": 2.5948045439358505,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043247415724083886,
      "loss": 4.6545,
      "mean_token_accuracy": 0.22596696019172668,
      "num_tokens": 57140182.0,
      "step": 31065
    },
    {
      "entropy": 4.978307628631592,
      "epoch": 2.5952221850985633,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043245289249508375,
      "loss": 4.5578,
      "mean_token_accuracy": 0.24537574201822282,
      "num_tokens": 57148364.0,
      "step": 31070
    },
    {
      "entropy": 5.006487846374512,
      "epoch": 2.5956398262612765,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004324316249929261,
      "loss": 4.7725,
      "mean_token_accuracy": 0.22230904698371887,
      "num_tokens": 57157475.0,
      "step": 31075
    },
    {
      "entropy": 5.047992849349976,
      "epoch": 2.5960574674239894,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004324103547347384,
      "loss": 4.7097,
      "mean_token_accuracy": 0.22771449238061905,
      "num_tokens": 57166343.0,
      "step": 31080
    },
    {
      "entropy": 4.978711414337158,
      "epoch": 2.596475108586702,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043238908172089306,
      "loss": 4.6496,
      "mean_token_accuracy": 0.24144928604364396,
      "num_tokens": 57175133.0,
      "step": 31085
    },
    {
      "entropy": 5.06573600769043,
      "epoch": 2.5968927497494154,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043236780595176227,
      "loss": 4.7208,
      "mean_token_accuracy": 0.22674540877342225,
      "num_tokens": 57183825.0,
      "step": 31090
    },
    {
      "entropy": 4.99486231803894,
      "epoch": 2.5973103909121282,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004323465274277187,
      "loss": 4.6852,
      "mean_token_accuracy": 0.22262825220823287,
      "num_tokens": 57193370.0,
      "step": 31095
    },
    {
      "entropy": 4.863487434387207,
      "epoch": 2.5977280320748415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004323252461491347,
      "loss": 4.5042,
      "mean_token_accuracy": 0.2398502692580223,
      "num_tokens": 57202366.0,
      "step": 31100
    },
    {
      "entropy": 5.007775068283081,
      "epoch": 2.5981456732375543,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004323039621163829,
      "loss": 4.6753,
      "mean_token_accuracy": 0.24120412170886993,
      "num_tokens": 57210756.0,
      "step": 31105
    },
    {
      "entropy": 4.9983562469482425,
      "epoch": 2.598563314400267,
      "grad_norm": 1.09375,
      "learning_rate": 0.000432282675329836,
      "loss": 4.6827,
      "mean_token_accuracy": 0.2323143482208252,
      "num_tokens": 57220327.0,
      "step": 31110
    },
    {
      "entropy": 5.133691263198853,
      "epoch": 2.5989809555629804,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043226138578986644,
      "loss": 4.8008,
      "mean_token_accuracy": 0.21788949370384217,
      "num_tokens": 57230152.0,
      "step": 31115
    },
    {
      "entropy": 5.034553003311157,
      "epoch": 2.599398596725693,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043224009349684706,
      "loss": 4.7002,
      "mean_token_accuracy": 0.22103979885578157,
      "num_tokens": 57239863.0,
      "step": 31120
    },
    {
      "entropy": 5.038334131240845,
      "epoch": 2.5998162378884064,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004322187984511506,
      "loss": 4.7071,
      "mean_token_accuracy": 0.2216392382979393,
      "num_tokens": 57249633.0,
      "step": 31125
    },
    {
      "entropy": 4.994322776794434,
      "epoch": 2.6002338790511192,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004321975006531498,
      "loss": 4.7082,
      "mean_token_accuracy": 0.23236584961414336,
      "num_tokens": 57259251.0,
      "step": 31130
    },
    {
      "entropy": 5.067329549789429,
      "epoch": 2.600651520213832,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004321762001032175,
      "loss": 4.7773,
      "mean_token_accuracy": 0.22271537333726882,
      "num_tokens": 57268576.0,
      "step": 31135
    },
    {
      "entropy": 5.099293804168701,
      "epoch": 2.6010691613765453,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043215489680172665,
      "loss": 4.7427,
      "mean_token_accuracy": 0.22436641305685043,
      "num_tokens": 57277823.0,
      "step": 31140
    },
    {
      "entropy": 5.072965812683106,
      "epoch": 2.601486802539258,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004321335907490502,
      "loss": 4.693,
      "mean_token_accuracy": 0.2245291531085968,
      "num_tokens": 57287929.0,
      "step": 31145
    },
    {
      "entropy": 4.96817193031311,
      "epoch": 2.6019044437019714,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004321122819455611,
      "loss": 4.6703,
      "mean_token_accuracy": 0.22761801779270172,
      "num_tokens": 57296584.0,
      "step": 31150
    },
    {
      "entropy": 4.981159591674805,
      "epoch": 2.602322084864684,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043209097039163233,
      "loss": 4.6933,
      "mean_token_accuracy": 0.2241129145026207,
      "num_tokens": 57305367.0,
      "step": 31155
    },
    {
      "entropy": 4.907794857025147,
      "epoch": 2.602739726027397,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043206965608763706,
      "loss": 4.5025,
      "mean_token_accuracy": 0.24229822903871537,
      "num_tokens": 57314480.0,
      "step": 31160
    },
    {
      "entropy": 4.92397723197937,
      "epoch": 2.6031573671901103,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004320483390339484,
      "loss": 4.5291,
      "mean_token_accuracy": 0.23623008280992508,
      "num_tokens": 57322989.0,
      "step": 31165
    },
    {
      "entropy": 5.02284951210022,
      "epoch": 2.6035750083528235,
      "grad_norm": 1.125,
      "learning_rate": 0.0004320270192309395,
      "loss": 4.768,
      "mean_token_accuracy": 0.21785923093557358,
      "num_tokens": 57332213.0,
      "step": 31170
    },
    {
      "entropy": 5.035300779342651,
      "epoch": 2.6039926495155363,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004320056966789836,
      "loss": 4.692,
      "mean_token_accuracy": 0.23087239414453506,
      "num_tokens": 57341374.0,
      "step": 31175
    },
    {
      "entropy": 5.158086156845092,
      "epoch": 2.604410290678249,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043198437137845403,
      "loss": 4.7704,
      "mean_token_accuracy": 0.22714982479810714,
      "num_tokens": 57351175.0,
      "step": 31180
    },
    {
      "entropy": 5.091256713867187,
      "epoch": 2.6048279318409624,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043196304332972414,
      "loss": 4.7661,
      "mean_token_accuracy": 0.22489510327577591,
      "num_tokens": 57360131.0,
      "step": 31185
    },
    {
      "entropy": 5.019409227371216,
      "epoch": 2.605245573003675,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043194171253316716,
      "loss": 4.7535,
      "mean_token_accuracy": 0.22638872116804123,
      "num_tokens": 57368457.0,
      "step": 31190
    },
    {
      "entropy": 5.026563930511474,
      "epoch": 2.6056632141663885,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043192037898915665,
      "loss": 4.6746,
      "mean_token_accuracy": 0.22511655390262603,
      "num_tokens": 57378171.0,
      "step": 31195
    },
    {
      "entropy": 5.07507290840149,
      "epoch": 2.6060808553291013,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000431899042698066,
      "loss": 4.7124,
      "mean_token_accuracy": 0.22362456619739532,
      "num_tokens": 57387165.0,
      "step": 31200
    },
    {
      "entropy": 5.045816230773926,
      "epoch": 2.606498496491814,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004318777036602687,
      "loss": 4.7098,
      "mean_token_accuracy": 0.2256873592734337,
      "num_tokens": 57395903.0,
      "step": 31205
    },
    {
      "entropy": 5.041414785385132,
      "epoch": 2.6069161376545273,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043185636187613837,
      "loss": 4.6978,
      "mean_token_accuracy": 0.22440752238035203,
      "num_tokens": 57405734.0,
      "step": 31210
    },
    {
      "entropy": 5.007499265670776,
      "epoch": 2.60733377881724,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004318350173460487,
      "loss": 4.6025,
      "mean_token_accuracy": 0.23205055445432662,
      "num_tokens": 57414383.0,
      "step": 31215
    },
    {
      "entropy": 5.065976238250732,
      "epoch": 2.6077514199799534,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004318136700703731,
      "loss": 4.756,
      "mean_token_accuracy": 0.22142608165740968,
      "num_tokens": 57424194.0,
      "step": 31220
    },
    {
      "entropy": 5.038768911361695,
      "epoch": 2.6081690611426662,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004317923200494856,
      "loss": 4.7193,
      "mean_token_accuracy": 0.21773310005664825,
      "num_tokens": 57433686.0,
      "step": 31225
    },
    {
      "entropy": 5.108309173583985,
      "epoch": 2.608586702305379,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043177096728375976,
      "loss": 4.703,
      "mean_token_accuracy": 0.22721367925405503,
      "num_tokens": 57443010.0,
      "step": 31230
    },
    {
      "entropy": 4.912920951843262,
      "epoch": 2.6090043434680923,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043174961177356946,
      "loss": 4.5518,
      "mean_token_accuracy": 0.23414430767297745,
      "num_tokens": 57451248.0,
      "step": 31235
    },
    {
      "entropy": 5.006478214263916,
      "epoch": 2.609421984630805,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004317282535192885,
      "loss": 4.6867,
      "mean_token_accuracy": 0.22368112504482268,
      "num_tokens": 57459672.0,
      "step": 31240
    },
    {
      "entropy": 5.025891923904419,
      "epoch": 2.6098396257935184,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043170689252129083,
      "loss": 4.6868,
      "mean_token_accuracy": 0.23004495352506638,
      "num_tokens": 57468343.0,
      "step": 31245
    },
    {
      "entropy": 4.912509727478027,
      "epoch": 2.610257266956231,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004316855287799504,
      "loss": 4.6112,
      "mean_token_accuracy": 0.23303083777427674,
      "num_tokens": 57477270.0,
      "step": 31250
    },
    {
      "entropy": 5.1069016456604,
      "epoch": 2.610674908118944,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043166416229564107,
      "loss": 4.8485,
      "mean_token_accuracy": 0.21716298311948776,
      "num_tokens": 57486343.0,
      "step": 31255
    },
    {
      "entropy": 5.20506329536438,
      "epoch": 2.6110925492816572,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043164279306873713,
      "loss": 4.8479,
      "mean_token_accuracy": 0.21905565708875657,
      "num_tokens": 57497700.0,
      "step": 31260
    },
    {
      "entropy": 4.925485372543335,
      "epoch": 2.61151019044437,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043162142109961246,
      "loss": 4.5682,
      "mean_token_accuracy": 0.23274968564510345,
      "num_tokens": 57506977.0,
      "step": 31265
    },
    {
      "entropy": 4.981057453155517,
      "epoch": 2.6119278316070833,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004316000463886414,
      "loss": 4.7223,
      "mean_token_accuracy": 0.22701816856861115,
      "num_tokens": 57516328.0,
      "step": 31270
    },
    {
      "entropy": 4.9953052520751955,
      "epoch": 2.612345472769796,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004315786689361979,
      "loss": 4.6205,
      "mean_token_accuracy": 0.23841443210840224,
      "num_tokens": 57524949.0,
      "step": 31275
    },
    {
      "entropy": 5.062626934051513,
      "epoch": 2.612763113932509,
      "grad_norm": 1.25,
      "learning_rate": 0.00043155728874265636,
      "loss": 4.7526,
      "mean_token_accuracy": 0.21875284910202025,
      "num_tokens": 57534214.0,
      "step": 31280
    },
    {
      "entropy": 5.054227447509765,
      "epoch": 2.613180755095222,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043153590580839106,
      "loss": 4.7587,
      "mean_token_accuracy": 0.22521868497133254,
      "num_tokens": 57544067.0,
      "step": 31285
    },
    {
      "entropy": 4.942282438278198,
      "epoch": 2.6135983962579354,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043151452013377636,
      "loss": 4.651,
      "mean_token_accuracy": 0.22800569385290145,
      "num_tokens": 57553386.0,
      "step": 31290
    },
    {
      "entropy": 5.004814338684082,
      "epoch": 2.6140160374206483,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043149313171918647,
      "loss": 4.7482,
      "mean_token_accuracy": 0.2198150023818016,
      "num_tokens": 57562003.0,
      "step": 31295
    },
    {
      "entropy": 5.082624769210815,
      "epoch": 2.614433678583361,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043147174056499605,
      "loss": 4.6501,
      "mean_token_accuracy": 0.23630136102437974,
      "num_tokens": 57571401.0,
      "step": 31300
    },
    {
      "entropy": 5.0295267581939695,
      "epoch": 2.6148513197460743,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043145034667157935,
      "loss": 4.7655,
      "mean_token_accuracy": 0.21964960992336274,
      "num_tokens": 57580837.0,
      "step": 31305
    },
    {
      "entropy": 4.9379647731781,
      "epoch": 2.615268960908787,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004314289500393111,
      "loss": 4.5596,
      "mean_token_accuracy": 0.23633454293012618,
      "num_tokens": 57589529.0,
      "step": 31310
    },
    {
      "entropy": 5.0807794570922855,
      "epoch": 2.6156866020715004,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043140755066856573,
      "loss": 4.7875,
      "mean_token_accuracy": 0.21996869295835494,
      "num_tokens": 57600225.0,
      "step": 31315
    },
    {
      "entropy": 4.996999454498291,
      "epoch": 2.616104243234213,
      "grad_norm": 1.09375,
      "learning_rate": 0.000431386148559718,
      "loss": 4.6991,
      "mean_token_accuracy": 0.23087942898273467,
      "num_tokens": 57609212.0,
      "step": 31320
    },
    {
      "entropy": 5.046552276611328,
      "epoch": 2.616521884396926,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043136474371314245,
      "loss": 4.695,
      "mean_token_accuracy": 0.22206660360097885,
      "num_tokens": 57618397.0,
      "step": 31325
    },
    {
      "entropy": 5.102052068710327,
      "epoch": 2.6169395255596393,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004313433361292138,
      "loss": 4.7356,
      "mean_token_accuracy": 0.2276801884174347,
      "num_tokens": 57627618.0,
      "step": 31330
    },
    {
      "entropy": 4.884241390228271,
      "epoch": 2.617357166722352,
      "grad_norm": 1.125,
      "learning_rate": 0.00043132192580830694,
      "loss": 4.5219,
      "mean_token_accuracy": 0.23722368031740187,
      "num_tokens": 57636252.0,
      "step": 31335
    },
    {
      "entropy": 4.892522382736206,
      "epoch": 2.6177748078850653,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043130051275079654,
      "loss": 4.6262,
      "mean_token_accuracy": 0.23618213534355165,
      "num_tokens": 57644681.0,
      "step": 31340
    },
    {
      "entropy": 5.06971287727356,
      "epoch": 2.618192449047778,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004312790969570575,
      "loss": 4.711,
      "mean_token_accuracy": 0.2204664960503578,
      "num_tokens": 57653633.0,
      "step": 31345
    },
    {
      "entropy": 4.96303071975708,
      "epoch": 2.618610090210491,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004312576784274648,
      "loss": 4.5527,
      "mean_token_accuracy": 0.24260056167840957,
      "num_tokens": 57662238.0,
      "step": 31350
    },
    {
      "entropy": 5.093283319473267,
      "epoch": 2.619027731373204,
      "grad_norm": 1.125,
      "learning_rate": 0.0004312362571623934,
      "loss": 4.8408,
      "mean_token_accuracy": 0.21462129354476928,
      "num_tokens": 57671534.0,
      "step": 31355
    },
    {
      "entropy": 5.104843425750732,
      "epoch": 2.619445372535917,
      "grad_norm": 1.0,
      "learning_rate": 0.0004312148331622182,
      "loss": 4.7502,
      "mean_token_accuracy": 0.21774016320705414,
      "num_tokens": 57681975.0,
      "step": 31360
    },
    {
      "entropy": 5.012953186035157,
      "epoch": 2.6198630136986303,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004311934064273143,
      "loss": 4.6718,
      "mean_token_accuracy": 0.22820666581392288,
      "num_tokens": 57691301.0,
      "step": 31365
    },
    {
      "entropy": 5.019132518768311,
      "epoch": 2.620280654861343,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004311719769580569,
      "loss": 4.6814,
      "mean_token_accuracy": 0.2302519738674164,
      "num_tokens": 57700337.0,
      "step": 31370
    },
    {
      "entropy": 5.108987140655517,
      "epoch": 2.620698296024056,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043115054475482094,
      "loss": 4.7611,
      "mean_token_accuracy": 0.2201029196381569,
      "num_tokens": 57708920.0,
      "step": 31375
    },
    {
      "entropy": 5.048565196990967,
      "epoch": 2.621115937186769,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004311291098179818,
      "loss": 4.7103,
      "mean_token_accuracy": 0.22457481920719147,
      "num_tokens": 57717827.0,
      "step": 31380
    },
    {
      "entropy": 5.099157762527466,
      "epoch": 2.621533578349482,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043110767214791466,
      "loss": 4.8051,
      "mean_token_accuracy": 0.21778713911771774,
      "num_tokens": 57727639.0,
      "step": 31385
    },
    {
      "entropy": 4.979329586029053,
      "epoch": 2.6219512195121952,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004310862317449948,
      "loss": 4.6456,
      "mean_token_accuracy": 0.2268392935395241,
      "num_tokens": 57736985.0,
      "step": 31390
    },
    {
      "entropy": 5.061885261535645,
      "epoch": 2.622368860674908,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004310647886095976,
      "loss": 4.7639,
      "mean_token_accuracy": 0.21782049536705017,
      "num_tokens": 57746735.0,
      "step": 31395
    },
    {
      "entropy": 4.933950328826905,
      "epoch": 2.622786501837621,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004310433427420984,
      "loss": 4.6621,
      "mean_token_accuracy": 0.2332362115383148,
      "num_tokens": 57755753.0,
      "step": 31400
    },
    {
      "entropy": 4.9991155624389645,
      "epoch": 2.623204143000334,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043102189414287255,
      "loss": 4.6425,
      "mean_token_accuracy": 0.2362857475876808,
      "num_tokens": 57765476.0,
      "step": 31405
    },
    {
      "entropy": 5.021741771697998,
      "epoch": 2.6236217841630474,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004310004428122958,
      "loss": 4.6823,
      "mean_token_accuracy": 0.23249199390411376,
      "num_tokens": 57775202.0,
      "step": 31410
    },
    {
      "entropy": 4.965254640579223,
      "epoch": 2.62403942532576,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043097898875074344,
      "loss": 4.6572,
      "mean_token_accuracy": 0.2329782322049141,
      "num_tokens": 57783845.0,
      "step": 31415
    },
    {
      "entropy": 5.130081415176392,
      "epoch": 2.624457066488473,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004309575319585912,
      "loss": 4.7989,
      "mean_token_accuracy": 0.21572598367929458,
      "num_tokens": 57792938.0,
      "step": 31420
    },
    {
      "entropy": 5.118165779113769,
      "epoch": 2.6248747076511862,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043093607243621456,
      "loss": 4.7821,
      "mean_token_accuracy": 0.21934910863637924,
      "num_tokens": 57802438.0,
      "step": 31425
    },
    {
      "entropy": 5.0904090881347654,
      "epoch": 2.625292348813899,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004309146101839893,
      "loss": 4.7123,
      "mean_token_accuracy": 0.21962365210056306,
      "num_tokens": 57813051.0,
      "step": 31430
    },
    {
      "entropy": 5.013396263122559,
      "epoch": 2.6257099899766123,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004308931452022911,
      "loss": 4.6945,
      "mean_token_accuracy": 0.23170342147350312,
      "num_tokens": 57821987.0,
      "step": 31435
    },
    {
      "entropy": 5.0135071754455565,
      "epoch": 2.626127631139325,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004308716774914957,
      "loss": 4.7773,
      "mean_token_accuracy": 0.21706026196479797,
      "num_tokens": 57831347.0,
      "step": 31440
    },
    {
      "entropy": 4.965327405929566,
      "epoch": 2.626545272302038,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043085020705197906,
      "loss": 4.5727,
      "mean_token_accuracy": 0.2377288222312927,
      "num_tokens": 57840262.0,
      "step": 31445
    },
    {
      "entropy": 5.125121402740478,
      "epoch": 2.626962913464751,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043082873388411696,
      "loss": 4.7358,
      "mean_token_accuracy": 0.22883038073778153,
      "num_tokens": 57849143.0,
      "step": 31450
    },
    {
      "entropy": 5.012681865692139,
      "epoch": 2.627380554627464,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043080725798828526,
      "loss": 4.6963,
      "mean_token_accuracy": 0.22488999664783477,
      "num_tokens": 57857853.0,
      "step": 31455
    },
    {
      "entropy": 5.056921815872192,
      "epoch": 2.6277981957901773,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004307857793648599,
      "loss": 4.7293,
      "mean_token_accuracy": 0.22550771832466127,
      "num_tokens": 57866159.0,
      "step": 31460
    },
    {
      "entropy": 5.06134181022644,
      "epoch": 2.62821583695289,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004307642980142171,
      "loss": 4.7198,
      "mean_token_accuracy": 0.2357912018895149,
      "num_tokens": 57874829.0,
      "step": 31465
    },
    {
      "entropy": 4.969124269485474,
      "epoch": 2.628633478115603,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043074281393673265,
      "loss": 4.6545,
      "mean_token_accuracy": 0.23065817654132842,
      "num_tokens": 57883382.0,
      "step": 31470
    },
    {
      "entropy": 5.046387815475464,
      "epoch": 2.629051119278316,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043072132713278273,
      "loss": 4.6821,
      "mean_token_accuracy": 0.22416360527276993,
      "num_tokens": 57891711.0,
      "step": 31475
    },
    {
      "entropy": 5.014673948287964,
      "epoch": 2.629468760441029,
      "grad_norm": 1.125,
      "learning_rate": 0.00043069983760274367,
      "loss": 4.7033,
      "mean_token_accuracy": 0.2272723451256752,
      "num_tokens": 57901349.0,
      "step": 31480
    },
    {
      "entropy": 4.936488103866577,
      "epoch": 2.629886401603742,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043067834534699144,
      "loss": 4.5787,
      "mean_token_accuracy": 0.2240470051765442,
      "num_tokens": 57911395.0,
      "step": 31485
    },
    {
      "entropy": 5.010385656356812,
      "epoch": 2.630304042766455,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043065685036590244,
      "loss": 4.6855,
      "mean_token_accuracy": 0.22862072438001632,
      "num_tokens": 57920251.0,
      "step": 31490
    },
    {
      "entropy": 4.934814310073852,
      "epoch": 2.630721683929168,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043063535265985283,
      "loss": 4.6538,
      "mean_token_accuracy": 0.23174146562814713,
      "num_tokens": 57928627.0,
      "step": 31495
    },
    {
      "entropy": 5.062898349761963,
      "epoch": 2.631139325091881,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043061385222921915,
      "loss": 4.7469,
      "mean_token_accuracy": 0.2222701445221901,
      "num_tokens": 57938980.0,
      "step": 31500
    },
    {
      "entropy": 4.991429710388184,
      "epoch": 2.631556966254594,
      "grad_norm": 1.125,
      "learning_rate": 0.0004305923490743777,
      "loss": 4.5978,
      "mean_token_accuracy": 0.2317742273211479,
      "num_tokens": 57947739.0,
      "step": 31505
    },
    {
      "entropy": 4.960998153686523,
      "epoch": 2.631974607417307,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004305708431957048,
      "loss": 4.6565,
      "mean_token_accuracy": 0.22969881743192672,
      "num_tokens": 57957084.0,
      "step": 31510
    },
    {
      "entropy": 4.951314258575439,
      "epoch": 2.63239224858002,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004305493345935769,
      "loss": 4.5731,
      "mean_token_accuracy": 0.23404977172613145,
      "num_tokens": 57967413.0,
      "step": 31515
    },
    {
      "entropy": 5.0701210498809814,
      "epoch": 2.632809889742733,
      "grad_norm": 1.125,
      "learning_rate": 0.00043052782326837084,
      "loss": 4.661,
      "mean_token_accuracy": 0.22913548350334167,
      "num_tokens": 57976712.0,
      "step": 31520
    },
    {
      "entropy": 4.941903686523437,
      "epoch": 2.633227530905446,
      "grad_norm": 1.09375,
      "learning_rate": 0.000430506309220463,
      "loss": 4.6811,
      "mean_token_accuracy": 0.22686709463596344,
      "num_tokens": 57985959.0,
      "step": 31525
    },
    {
      "entropy": 5.021202802658081,
      "epoch": 2.6336451720681593,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004304847924502301,
      "loss": 4.6762,
      "mean_token_accuracy": 0.21851125359535217,
      "num_tokens": 57995500.0,
      "step": 31530
    },
    {
      "entropy": 5.006774044036865,
      "epoch": 2.634062813230872,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043046327295804865,
      "loss": 4.6871,
      "mean_token_accuracy": 0.22582811564207078,
      "num_tokens": 58004751.0,
      "step": 31535
    },
    {
      "entropy": 4.984838390350342,
      "epoch": 2.634480454393585,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004304417507442955,
      "loss": 4.6475,
      "mean_token_accuracy": 0.23989203423261643,
      "num_tokens": 58014712.0,
      "step": 31540
    },
    {
      "entropy": 5.033983373641968,
      "epoch": 2.634898095556298,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004304202258093473,
      "loss": 4.7497,
      "mean_token_accuracy": 0.22406793236732483,
      "num_tokens": 58024518.0,
      "step": 31545
    },
    {
      "entropy": 5.052891683578491,
      "epoch": 2.635315736719011,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043039869815358114,
      "loss": 4.7077,
      "mean_token_accuracy": 0.2222844362258911,
      "num_tokens": 58033800.0,
      "step": 31550
    },
    {
      "entropy": 4.977894687652588,
      "epoch": 2.6357333778817242,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004303771677773736,
      "loss": 4.6769,
      "mean_token_accuracy": 0.22886688709259034,
      "num_tokens": 58042800.0,
      "step": 31555
    },
    {
      "entropy": 5.008616399765015,
      "epoch": 2.636151019044437,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004303556346811018,
      "loss": 4.6627,
      "mean_token_accuracy": 0.2249346062541008,
      "num_tokens": 58051748.0,
      "step": 31560
    },
    {
      "entropy": 4.9973736763000485,
      "epoch": 2.63656866020715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004303340988651425,
      "loss": 4.6485,
      "mean_token_accuracy": 0.229887518286705,
      "num_tokens": 58061081.0,
      "step": 31565
    },
    {
      "entropy": 5.07425103187561,
      "epoch": 2.636986301369863,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004303125603298729,
      "loss": 4.7357,
      "mean_token_accuracy": 0.22765659391880036,
      "num_tokens": 58069709.0,
      "step": 31570
    },
    {
      "entropy": 5.037710189819336,
      "epoch": 2.637403942532576,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043029101907567004,
      "loss": 4.6636,
      "mean_token_accuracy": 0.22779324203729628,
      "num_tokens": 58079007.0,
      "step": 31575
    },
    {
      "entropy": 5.075628042221069,
      "epoch": 2.637821583695289,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004302694751029108,
      "loss": 4.7338,
      "mean_token_accuracy": 0.2223716989159584,
      "num_tokens": 58087673.0,
      "step": 31580
    },
    {
      "entropy": 5.008274364471435,
      "epoch": 2.638239224858002,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004302479284119727,
      "loss": 4.7292,
      "mean_token_accuracy": 0.22378365993499755,
      "num_tokens": 58097248.0,
      "step": 31585
    },
    {
      "entropy": 5.07074761390686,
      "epoch": 2.638656866020715,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043022637900323263,
      "loss": 4.754,
      "mean_token_accuracy": 0.22345616668462753,
      "num_tokens": 58106412.0,
      "step": 31590
    },
    {
      "entropy": 5.033340263366699,
      "epoch": 2.639074507183428,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004302048268770679,
      "loss": 4.7279,
      "mean_token_accuracy": 0.2182938724756241,
      "num_tokens": 58115432.0,
      "step": 31595
    },
    {
      "entropy": 5.059076261520386,
      "epoch": 2.639492148346141,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043018327203385594,
      "loss": 4.7314,
      "mean_token_accuracy": 0.2276083454489708,
      "num_tokens": 58123740.0,
      "step": 31600
    },
    {
      "entropy": 4.890625238418579,
      "epoch": 2.639909789508854,
      "grad_norm": 1.03125,
      "learning_rate": 0.000430161714473974,
      "loss": 4.5081,
      "mean_token_accuracy": 0.24340036362409592,
      "num_tokens": 58132743.0,
      "step": 31605
    },
    {
      "entropy": 4.954384803771973,
      "epoch": 2.640327430671567,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043014015419779947,
      "loss": 4.6502,
      "mean_token_accuracy": 0.23210780024528505,
      "num_tokens": 58143068.0,
      "step": 31610
    },
    {
      "entropy": 4.936807775497437,
      "epoch": 2.6407450718342798,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004301185912057097,
      "loss": 4.5886,
      "mean_token_accuracy": 0.2318621650338173,
      "num_tokens": 58151859.0,
      "step": 31615
    },
    {
      "entropy": 5.045790958404541,
      "epoch": 2.641162712996993,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043009702549808243,
      "loss": 4.7535,
      "mean_token_accuracy": 0.22102392315864564,
      "num_tokens": 58160295.0,
      "step": 31620
    },
    {
      "entropy": 5.08877215385437,
      "epoch": 2.641580354159706,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043007545707529496,
      "loss": 4.6688,
      "mean_token_accuracy": 0.23241938203573226,
      "num_tokens": 58169392.0,
      "step": 31625
    },
    {
      "entropy": 4.996350860595703,
      "epoch": 2.641997995322419,
      "grad_norm": 1.21875,
      "learning_rate": 0.000430053885937725,
      "loss": 4.7389,
      "mean_token_accuracy": 0.22466142177581788,
      "num_tokens": 58178609.0,
      "step": 31630
    },
    {
      "entropy": 5.049106216430664,
      "epoch": 2.642415636485132,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043003231208575004,
      "loss": 4.7561,
      "mean_token_accuracy": 0.22731226682662964,
      "num_tokens": 58187463.0,
      "step": 31635
    },
    {
      "entropy": 5.017570352554321,
      "epoch": 2.6428332776478447,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043001073551974796,
      "loss": 4.6282,
      "mean_token_accuracy": 0.23731303364038467,
      "num_tokens": 58195980.0,
      "step": 31640
    },
    {
      "entropy": 4.931308126449585,
      "epoch": 2.643250918810558,
      "grad_norm": 1.125,
      "learning_rate": 0.00042998915624009625,
      "loss": 4.5532,
      "mean_token_accuracy": 0.24440379291772843,
      "num_tokens": 58203860.0,
      "step": 31645
    },
    {
      "entropy": 4.963100004196167,
      "epoch": 2.643668559973271,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004299675742471729,
      "loss": 4.6489,
      "mean_token_accuracy": 0.23416529297828675,
      "num_tokens": 58213050.0,
      "step": 31650
    },
    {
      "entropy": 5.045368385314942,
      "epoch": 2.644086201135984,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042994598954135553,
      "loss": 4.6674,
      "mean_token_accuracy": 0.23965429663658142,
      "num_tokens": 58221893.0,
      "step": 31655
    },
    {
      "entropy": 4.953304719924927,
      "epoch": 2.644503842298697,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004299244021230222,
      "loss": 4.6948,
      "mean_token_accuracy": 0.23290190249681472,
      "num_tokens": 58231067.0,
      "step": 31660
    },
    {
      "entropy": 4.97906174659729,
      "epoch": 2.64492148346141,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004299028119925507,
      "loss": 4.672,
      "mean_token_accuracy": 0.23539190739393234,
      "num_tokens": 58239799.0,
      "step": 31665
    },
    {
      "entropy": 4.997708415985107,
      "epoch": 2.645339124624123,
      "grad_norm": 1.0,
      "learning_rate": 0.00042988121915031894,
      "loss": 4.6127,
      "mean_token_accuracy": 0.22944242805242537,
      "num_tokens": 58249856.0,
      "step": 31670
    },
    {
      "entropy": 4.972099924087525,
      "epoch": 2.645756765786836,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004298596235967051,
      "loss": 4.5863,
      "mean_token_accuracy": 0.23918354511260986,
      "num_tokens": 58258990.0,
      "step": 31675
    },
    {
      "entropy": 5.029197645187378,
      "epoch": 2.646174406949549,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004298380253320871,
      "loss": 4.6837,
      "mean_token_accuracy": 0.23449262976646423,
      "num_tokens": 58269226.0,
      "step": 31680
    },
    {
      "entropy": 4.992516279220581,
      "epoch": 2.646592048112262,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004298164243568431,
      "loss": 4.7623,
      "mean_token_accuracy": 0.21929428577423096,
      "num_tokens": 58278737.0,
      "step": 31685
    },
    {
      "entropy": 4.974767827987671,
      "epoch": 2.647009689274975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004297948206713513,
      "loss": 4.5723,
      "mean_token_accuracy": 0.23645229041576385,
      "num_tokens": 58286781.0,
      "step": 31690
    },
    {
      "entropy": 5.0062864303588865,
      "epoch": 2.647427330437688,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042977321427598974,
      "loss": 4.6538,
      "mean_token_accuracy": 0.22714734971523284,
      "num_tokens": 58296263.0,
      "step": 31695
    },
    {
      "entropy": 4.957058620452881,
      "epoch": 2.647844971600401,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004297516051711368,
      "loss": 4.6895,
      "mean_token_accuracy": 0.23211335390806198,
      "num_tokens": 58305421.0,
      "step": 31700
    },
    {
      "entropy": 5.00985689163208,
      "epoch": 2.648262612763114,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004297299933571708,
      "loss": 4.6274,
      "mean_token_accuracy": 0.23005106002092363,
      "num_tokens": 58313744.0,
      "step": 31705
    },
    {
      "entropy": 5.01157865524292,
      "epoch": 2.6486802539258267,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004297083788344699,
      "loss": 4.6618,
      "mean_token_accuracy": 0.23035388588905334,
      "num_tokens": 58322844.0,
      "step": 31710
    },
    {
      "entropy": 4.994074535369873,
      "epoch": 2.64909789508854,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042968676160341275,
      "loss": 4.696,
      "mean_token_accuracy": 0.22969970852136612,
      "num_tokens": 58331748.0,
      "step": 31715
    },
    {
      "entropy": 5.013241958618164,
      "epoch": 2.649515536251253,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004296651416643775,
      "loss": 4.6868,
      "mean_token_accuracy": 0.22230846136808396,
      "num_tokens": 58340929.0,
      "step": 31720
    },
    {
      "entropy": 5.07204041481018,
      "epoch": 2.649933177413966,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042964351901774285,
      "loss": 4.7519,
      "mean_token_accuracy": 0.22089627087116243,
      "num_tokens": 58350156.0,
      "step": 31725
    },
    {
      "entropy": 5.008192920684815,
      "epoch": 2.650350818576679,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004296218936638873,
      "loss": 4.6741,
      "mean_token_accuracy": 0.23323683589696884,
      "num_tokens": 58360179.0,
      "step": 31730
    },
    {
      "entropy": 4.892460632324219,
      "epoch": 2.6507684597393917,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042960026560318927,
      "loss": 4.5663,
      "mean_token_accuracy": 0.24099613279104232,
      "num_tokens": 58369310.0,
      "step": 31735
    },
    {
      "entropy": 5.040891456604004,
      "epoch": 2.651186100902105,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042957863483602756,
      "loss": 4.7048,
      "mean_token_accuracy": 0.22822246551513672,
      "num_tokens": 58378337.0,
      "step": 31740
    },
    {
      "entropy": 5.007733631134033,
      "epoch": 2.6516037420648177,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042955700136278077,
      "loss": 4.6443,
      "mean_token_accuracy": 0.23745195865631102,
      "num_tokens": 58386244.0,
      "step": 31745
    },
    {
      "entropy": 4.984401655197144,
      "epoch": 2.652021383227531,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004295353651838276,
      "loss": 4.7098,
      "mean_token_accuracy": 0.22947203665971755,
      "num_tokens": 58395283.0,
      "step": 31750
    },
    {
      "entropy": 5.039993047714233,
      "epoch": 2.652439024390244,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042951372629954695,
      "loss": 4.6868,
      "mean_token_accuracy": 0.22762035727500915,
      "num_tokens": 58404571.0,
      "step": 31755
    },
    {
      "entropy": 5.061487483978271,
      "epoch": 2.6528566655529566,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004294920847103174,
      "loss": 4.6713,
      "mean_token_accuracy": 0.2254022553563118,
      "num_tokens": 58413664.0,
      "step": 31760
    },
    {
      "entropy": 4.994629144668579,
      "epoch": 2.65327430671567,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004294704404165181,
      "loss": 4.6595,
      "mean_token_accuracy": 0.2291726991534233,
      "num_tokens": 58422185.0,
      "step": 31765
    },
    {
      "entropy": 4.988795948028565,
      "epoch": 2.653691947878383,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042944879341852763,
      "loss": 4.6746,
      "mean_token_accuracy": 0.22690875232219695,
      "num_tokens": 58431696.0,
      "step": 31770
    },
    {
      "entropy": 5.0529945373535154,
      "epoch": 2.654109589041096,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004294271437167252,
      "loss": 4.6894,
      "mean_token_accuracy": 0.23295030295848845,
      "num_tokens": 58441058.0,
      "step": 31775
    },
    {
      "entropy": 4.958313083648681,
      "epoch": 2.6545272302038088,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042940549131148976,
      "loss": 4.5982,
      "mean_token_accuracy": 0.23234693259000777,
      "num_tokens": 58450490.0,
      "step": 31780
    },
    {
      "entropy": 5.102494096755981,
      "epoch": 2.654944871366522,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004293838362032003,
      "loss": 4.7682,
      "mean_token_accuracy": 0.2260906383395195,
      "num_tokens": 58460235.0,
      "step": 31785
    },
    {
      "entropy": 5.046355104446411,
      "epoch": 2.655362512529235,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042936217839223596,
      "loss": 4.7454,
      "mean_token_accuracy": 0.2280333921313286,
      "num_tokens": 58469674.0,
      "step": 31790
    },
    {
      "entropy": 5.0374139785766605,
      "epoch": 2.655780153691948,
      "grad_norm": 0.9375,
      "learning_rate": 0.00042934051787897593,
      "loss": 4.7193,
      "mean_token_accuracy": 0.22499503046274186,
      "num_tokens": 58479426.0,
      "step": 31795
    },
    {
      "entropy": 5.039321613311768,
      "epoch": 2.656197794854661,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042931885466379935,
      "loss": 4.7062,
      "mean_token_accuracy": 0.23081884980201722,
      "num_tokens": 58489341.0,
      "step": 31800
    },
    {
      "entropy": 4.960751628875732,
      "epoch": 2.6566154360173737,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004292971887470854,
      "loss": 4.6163,
      "mean_token_accuracy": 0.23744630068540573,
      "num_tokens": 58497708.0,
      "step": 31805
    },
    {
      "entropy": 5.051169157028198,
      "epoch": 2.657033077180087,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042927552012921353,
      "loss": 4.7464,
      "mean_token_accuracy": 0.225825759768486,
      "num_tokens": 58507364.0,
      "step": 31810
    },
    {
      "entropy": 4.981136608123779,
      "epoch": 2.6574507183428,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042925384881056284,
      "loss": 4.6183,
      "mean_token_accuracy": 0.2331593304872513,
      "num_tokens": 58516064.0,
      "step": 31815
    },
    {
      "entropy": 5.005228805541992,
      "epoch": 2.657868359505513,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042923217479151296,
      "loss": 4.7304,
      "mean_token_accuracy": 0.22347741425037385,
      "num_tokens": 58526010.0,
      "step": 31820
    },
    {
      "entropy": 5.046904754638672,
      "epoch": 2.658286000668226,
      "grad_norm": 0.984375,
      "learning_rate": 0.00042921049807244317,
      "loss": 4.7252,
      "mean_token_accuracy": 0.2273346811532974,
      "num_tokens": 58535407.0,
      "step": 31825
    },
    {
      "entropy": 5.0631091594696045,
      "epoch": 2.6587036418309387,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000429188818653733,
      "loss": 4.7177,
      "mean_token_accuracy": 0.2231988325715065,
      "num_tokens": 58545017.0,
      "step": 31830
    },
    {
      "entropy": 5.010118818283081,
      "epoch": 2.659121282993652,
      "grad_norm": 1.125,
      "learning_rate": 0.0004291671365357619,
      "loss": 4.6296,
      "mean_token_accuracy": 0.23137523829936982,
      "num_tokens": 58553933.0,
      "step": 31835
    },
    {
      "entropy": 4.9676981449127195,
      "epoch": 2.6595389241563647,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00042914545171890964,
      "loss": 4.5851,
      "mean_token_accuracy": 0.2309714674949646,
      "num_tokens": 58563745.0,
      "step": 31840
    },
    {
      "entropy": 4.938228130340576,
      "epoch": 2.659956565319078,
      "grad_norm": 1.125,
      "learning_rate": 0.00042912376420355553,
      "loss": 4.6557,
      "mean_token_accuracy": 0.22769935429096222,
      "num_tokens": 58572567.0,
      "step": 31845
    },
    {
      "entropy": 5.001905679702759,
      "epoch": 2.660374206481791,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004291020739900796,
      "loss": 4.6816,
      "mean_token_accuracy": 0.22504477202892303,
      "num_tokens": 58581060.0,
      "step": 31850
    },
    {
      "entropy": 5.040688705444336,
      "epoch": 2.6607918476445036,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004290803810788612,
      "loss": 4.772,
      "mean_token_accuracy": 0.22557052969932556,
      "num_tokens": 58590037.0,
      "step": 31855
    },
    {
      "entropy": 5.075235223770141,
      "epoch": 2.661209488807217,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004290586854702804,
      "loss": 4.6864,
      "mean_token_accuracy": 0.23279941976070403,
      "num_tokens": 58598648.0,
      "step": 31860
    },
    {
      "entropy": 4.960077285766602,
      "epoch": 2.6616271299699297,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042903698716471677,
      "loss": 4.5597,
      "mean_token_accuracy": 0.2347632110118866,
      "num_tokens": 58608233.0,
      "step": 31865
    },
    {
      "entropy": 4.904622173309326,
      "epoch": 2.662044771132643,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004290152861625503,
      "loss": 4.5953,
      "mean_token_accuracy": 0.2304455369710922,
      "num_tokens": 58617682.0,
      "step": 31870
    },
    {
      "entropy": 4.957739734649659,
      "epoch": 2.6624624122953557,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042899358246416085,
      "loss": 4.6585,
      "mean_token_accuracy": 0.2328437864780426,
      "num_tokens": 58627673.0,
      "step": 31875
    },
    {
      "entropy": 5.051027584075928,
      "epoch": 2.6628800534580686,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042897187606992846,
      "loss": 4.6717,
      "mean_token_accuracy": 0.23006118685007096,
      "num_tokens": 58637066.0,
      "step": 31880
    },
    {
      "entropy": 5.012171268463135,
      "epoch": 2.663297694620782,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004289501669802329,
      "loss": 4.6957,
      "mean_token_accuracy": 0.22766315042972565,
      "num_tokens": 58645750.0,
      "step": 31885
    },
    {
      "entropy": 4.8727751731872555,
      "epoch": 2.663715335783495,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004289284551954545,
      "loss": 4.5537,
      "mean_token_accuracy": 0.2344193309545517,
      "num_tokens": 58655096.0,
      "step": 31890
    },
    {
      "entropy": 4.955721044540406,
      "epoch": 2.664132976946208,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004289067407159731,
      "loss": 4.5417,
      "mean_token_accuracy": 0.24526460617780685,
      "num_tokens": 58663596.0,
      "step": 31895
    },
    {
      "entropy": 5.086572599411011,
      "epoch": 2.6645506181089207,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042888502354216897,
      "loss": 4.8189,
      "mean_token_accuracy": 0.2141084000468254,
      "num_tokens": 58672110.0,
      "step": 31900
    },
    {
      "entropy": 5.080271434783936,
      "epoch": 2.664968259271634,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042886330367442234,
      "loss": 4.743,
      "mean_token_accuracy": 0.22953566610813142,
      "num_tokens": 58682471.0,
      "step": 31905
    },
    {
      "entropy": 4.990059852600098,
      "epoch": 2.6653859004343468,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042884158111311326,
      "loss": 4.6463,
      "mean_token_accuracy": 0.23794717490673065,
      "num_tokens": 58691664.0,
      "step": 31910
    },
    {
      "entropy": 4.9584673881530765,
      "epoch": 2.66580354159706,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004288198558586222,
      "loss": 4.588,
      "mean_token_accuracy": 0.23920825123786926,
      "num_tokens": 58700495.0,
      "step": 31915
    },
    {
      "entropy": 4.99829330444336,
      "epoch": 2.666221182759773,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004287981279113294,
      "loss": 4.6078,
      "mean_token_accuracy": 0.23292710930109023,
      "num_tokens": 58709793.0,
      "step": 31920
    },
    {
      "entropy": 5.022454118728637,
      "epoch": 2.6666388239224856,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004287763972716152,
      "loss": 4.7217,
      "mean_token_accuracy": 0.22750740200281144,
      "num_tokens": 58719296.0,
      "step": 31925
    },
    {
      "entropy": 4.981874752044678,
      "epoch": 2.667056465085199,
      "grad_norm": 1.0,
      "learning_rate": 0.00042875466393986017,
      "loss": 4.6664,
      "mean_token_accuracy": 0.2274172529578209,
      "num_tokens": 58728096.0,
      "step": 31930
    },
    {
      "entropy": 4.980859184265137,
      "epoch": 2.6674741062479117,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004287329279164445,
      "loss": 4.6846,
      "mean_token_accuracy": 0.226352758705616,
      "num_tokens": 58737589.0,
      "step": 31935
    },
    {
      "entropy": 4.9575258731842045,
      "epoch": 2.667891747410625,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042871118920174907,
      "loss": 4.5445,
      "mean_token_accuracy": 0.23622696995735168,
      "num_tokens": 58746124.0,
      "step": 31940
    },
    {
      "entropy": 5.061528873443604,
      "epoch": 2.6683093885733378,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004286894477961541,
      "loss": 4.7376,
      "mean_token_accuracy": 0.2250124618411064,
      "num_tokens": 58755629.0,
      "step": 31945
    },
    {
      "entropy": 5.055577659606934,
      "epoch": 2.6687270297360506,
      "grad_norm": 1.125,
      "learning_rate": 0.00042866770370004044,
      "loss": 4.7408,
      "mean_token_accuracy": 0.23156363219022752,
      "num_tokens": 58765694.0,
      "step": 31950
    },
    {
      "entropy": 5.023420619964599,
      "epoch": 2.669144670898764,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004286459569137886,
      "loss": 4.6738,
      "mean_token_accuracy": 0.230927774310112,
      "num_tokens": 58774461.0,
      "step": 31955
    },
    {
      "entropy": 4.982572841644287,
      "epoch": 2.6695623120614766,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004286242074377794,
      "loss": 4.6434,
      "mean_token_accuracy": 0.2352468639612198,
      "num_tokens": 58783632.0,
      "step": 31960
    },
    {
      "entropy": 4.919364070892334,
      "epoch": 2.66997995322419,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004286024552723935,
      "loss": 4.5813,
      "mean_token_accuracy": 0.23531865775585176,
      "num_tokens": 58792495.0,
      "step": 31965
    },
    {
      "entropy": 4.978773307800293,
      "epoch": 2.6703975943869027,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042858070041801175,
      "loss": 4.6156,
      "mean_token_accuracy": 0.23237486630678178,
      "num_tokens": 58802130.0,
      "step": 31970
    },
    {
      "entropy": 4.9888299942016605,
      "epoch": 2.6708152355496155,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042855894287501493,
      "loss": 4.6816,
      "mean_token_accuracy": 0.22972810566425322,
      "num_tokens": 58810841.0,
      "step": 31975
    },
    {
      "entropy": 5.07589898109436,
      "epoch": 2.671232876712329,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042853718264378404,
      "loss": 4.6377,
      "mean_token_accuracy": 0.2354937329888344,
      "num_tokens": 58819691.0,
      "step": 31980
    },
    {
      "entropy": 4.979425621032715,
      "epoch": 2.671650517875042,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042851541972469996,
      "loss": 4.6463,
      "mean_token_accuracy": 0.23036426901817322,
      "num_tokens": 58829024.0,
      "step": 31985
    },
    {
      "entropy": 4.978243017196656,
      "epoch": 2.672068159037755,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042849365411814355,
      "loss": 4.6529,
      "mean_token_accuracy": 0.22662848383188247,
      "num_tokens": 58838558.0,
      "step": 31990
    },
    {
      "entropy": 5.102135848999024,
      "epoch": 2.6724858002004677,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004284718858244961,
      "loss": 4.8579,
      "mean_token_accuracy": 0.2163931205868721,
      "num_tokens": 58848369.0,
      "step": 31995
    },
    {
      "entropy": 5.041555166244507,
      "epoch": 2.672903441363181,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042845011484413846,
      "loss": 4.6973,
      "mean_token_accuracy": 0.22763971537351607,
      "num_tokens": 58857793.0,
      "step": 32000
    },
    {
      "entropy": 5.056178236007691,
      "epoch": 2.6733210825258937,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004284283411774519,
      "loss": 4.7116,
      "mean_token_accuracy": 0.22712981849908828,
      "num_tokens": 58867108.0,
      "step": 32005
    },
    {
      "entropy": 5.0347308158874515,
      "epoch": 2.673738723688607,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042840656482481755,
      "loss": 4.7084,
      "mean_token_accuracy": 0.22272755950689316,
      "num_tokens": 58877136.0,
      "step": 32010
    },
    {
      "entropy": 5.017215204238892,
      "epoch": 2.67415636485132,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004283847857866167,
      "loss": 4.6552,
      "mean_token_accuracy": 0.23394251763820648,
      "num_tokens": 58885437.0,
      "step": 32015
    },
    {
      "entropy": 5.029575443267822,
      "epoch": 2.6745740060140326,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042836300406323035,
      "loss": 4.7277,
      "mean_token_accuracy": 0.22775544077157975,
      "num_tokens": 58895392.0,
      "step": 32020
    },
    {
      "entropy": 5.1315598487854,
      "epoch": 2.674991647176746,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00042834121965504014,
      "loss": 4.7549,
      "mean_token_accuracy": 0.22714708298444747,
      "num_tokens": 58904441.0,
      "step": 32025
    },
    {
      "entropy": 5.033288240432739,
      "epoch": 2.6754092883394587,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042831943256242726,
      "loss": 4.5973,
      "mean_token_accuracy": 0.23784442991018295,
      "num_tokens": 58913789.0,
      "step": 32030
    },
    {
      "entropy": 5.034477472305298,
      "epoch": 2.675826929502172,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004282976427857732,
      "loss": 4.7298,
      "mean_token_accuracy": 0.21970774233341217,
      "num_tokens": 58923152.0,
      "step": 32035
    },
    {
      "entropy": 5.055186223983765,
      "epoch": 2.6762445706648847,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004282758503254594,
      "loss": 4.7562,
      "mean_token_accuracy": 0.2280259445309639,
      "num_tokens": 58933845.0,
      "step": 32040
    },
    {
      "entropy": 4.9882896900177,
      "epoch": 2.6766622118275976,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042825405518186725,
      "loss": 4.5735,
      "mean_token_accuracy": 0.23169414699077606,
      "num_tokens": 58943307.0,
      "step": 32045
    },
    {
      "entropy": 5.015697622299195,
      "epoch": 2.677079852990311,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042823225735537845,
      "loss": 4.6493,
      "mean_token_accuracy": 0.23097945898771285,
      "num_tokens": 58952068.0,
      "step": 32050
    },
    {
      "entropy": 4.973033285140991,
      "epoch": 2.6774974941530236,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004282104568463745,
      "loss": 4.6236,
      "mean_token_accuracy": 0.22839795053005219,
      "num_tokens": 58961000.0,
      "step": 32055
    },
    {
      "entropy": 5.005033254623413,
      "epoch": 2.677915135315737,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004281886536552372,
      "loss": 4.7322,
      "mean_token_accuracy": 0.2215066909790039,
      "num_tokens": 58970254.0,
      "step": 32060
    },
    {
      "entropy": 5.019870185852051,
      "epoch": 2.6783327764784497,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000428166847782348,
      "loss": 4.6743,
      "mean_token_accuracy": 0.23006365001201629,
      "num_tokens": 58980104.0,
      "step": 32065
    },
    {
      "entropy": 4.977225875854492,
      "epoch": 2.6787504176411625,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004281450392280889,
      "loss": 4.5557,
      "mean_token_accuracy": 0.2427832841873169,
      "num_tokens": 58988653.0,
      "step": 32070
    },
    {
      "entropy": 4.96872034072876,
      "epoch": 2.6791680588038758,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004281232279928414,
      "loss": 4.6468,
      "mean_token_accuracy": 0.23366237431764603,
      "num_tokens": 58997657.0,
      "step": 32075
    },
    {
      "entropy": 5.067616319656372,
      "epoch": 2.6795856999665886,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004281014140769875,
      "loss": 4.749,
      "mean_token_accuracy": 0.2212056875228882,
      "num_tokens": 59007024.0,
      "step": 32080
    },
    {
      "entropy": 5.0817896842956545,
      "epoch": 2.680003341129302,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042807959748090914,
      "loss": 4.8055,
      "mean_token_accuracy": 0.2143508329987526,
      "num_tokens": 59016162.0,
      "step": 32085
    },
    {
      "entropy": 5.120092105865479,
      "epoch": 2.6804209822920146,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004280577782049881,
      "loss": 4.6957,
      "mean_token_accuracy": 0.22959620058536528,
      "num_tokens": 59025029.0,
      "step": 32090
    },
    {
      "entropy": 5.04509744644165,
      "epoch": 2.6808386234547275,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042803595624960647,
      "loss": 4.7706,
      "mean_token_accuracy": 0.22432608604431153,
      "num_tokens": 59034575.0,
      "step": 32095
    },
    {
      "entropy": 5.012066173553467,
      "epoch": 2.6812562646174407,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042801413161514613,
      "loss": 4.6962,
      "mean_token_accuracy": 0.22658999711275102,
      "num_tokens": 59044208.0,
      "step": 32100
    },
    {
      "entropy": 5.009447526931763,
      "epoch": 2.681673905780154,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042799230430198935,
      "loss": 4.7286,
      "mean_token_accuracy": 0.22786771953105928,
      "num_tokens": 59053528.0,
      "step": 32105
    },
    {
      "entropy": 5.098279285430908,
      "epoch": 2.6820915469428668,
      "grad_norm": 1.046875,
      "learning_rate": 0.000427970474310518,
      "loss": 4.7413,
      "mean_token_accuracy": 0.22218701541423796,
      "num_tokens": 59062766.0,
      "step": 32110
    },
    {
      "entropy": 4.977794504165649,
      "epoch": 2.6825091881055796,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042794864164111447,
      "loss": 4.7228,
      "mean_token_accuracy": 0.2303025335073471,
      "num_tokens": 59071639.0,
      "step": 32115
    },
    {
      "entropy": 5.064857435226441,
      "epoch": 2.682926829268293,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004279268062941608,
      "loss": 4.7142,
      "mean_token_accuracy": 0.2234805181622505,
      "num_tokens": 59081023.0,
      "step": 32120
    },
    {
      "entropy": 5.075453424453736,
      "epoch": 2.6833444704310057,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004279049682700394,
      "loss": 4.6304,
      "mean_token_accuracy": 0.2341720089316368,
      "num_tokens": 59089916.0,
      "step": 32125
    },
    {
      "entropy": 4.992135906219483,
      "epoch": 2.683762111593719,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042788312756913234,
      "loss": 4.6406,
      "mean_token_accuracy": 0.22941638231277467,
      "num_tokens": 59099530.0,
      "step": 32130
    },
    {
      "entropy": 5.01830267906189,
      "epoch": 2.6841797527564317,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004278612841918222,
      "loss": 4.7135,
      "mean_token_accuracy": 0.2316283643245697,
      "num_tokens": 59108341.0,
      "step": 32135
    },
    {
      "entropy": 5.085480785369873,
      "epoch": 2.6845973939191445,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004278394381384912,
      "loss": 4.8589,
      "mean_token_accuracy": 0.2157691404223442,
      "num_tokens": 59118483.0,
      "step": 32140
    },
    {
      "entropy": 4.969301557540893,
      "epoch": 2.685015035081858,
      "grad_norm": 1.125,
      "learning_rate": 0.00042781758940952194,
      "loss": 4.5736,
      "mean_token_accuracy": 0.23873230814933777,
      "num_tokens": 59127630.0,
      "step": 32145
    },
    {
      "entropy": 5.054428052902222,
      "epoch": 2.6854326762445706,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004277957380052968,
      "loss": 4.7114,
      "mean_token_accuracy": 0.22313012480735778,
      "num_tokens": 59136213.0,
      "step": 32150
    },
    {
      "entropy": 5.035315656661988,
      "epoch": 2.685850317407284,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042777388392619835,
      "loss": 4.6907,
      "mean_token_accuracy": 0.2231716975569725,
      "num_tokens": 59145915.0,
      "step": 32155
    },
    {
      "entropy": 4.973000001907349,
      "epoch": 2.6862679585699967,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004277520271726092,
      "loss": 4.5654,
      "mean_token_accuracy": 0.23558054715394974,
      "num_tokens": 59154011.0,
      "step": 32160
    },
    {
      "entropy": 4.923845100402832,
      "epoch": 2.6866855997327095,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004277301677449119,
      "loss": 4.6276,
      "mean_token_accuracy": 0.2358328104019165,
      "num_tokens": 59162664.0,
      "step": 32165
    },
    {
      "entropy": 5.049215793609619,
      "epoch": 2.6871032408954227,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042770830564348915,
      "loss": 4.77,
      "mean_token_accuracy": 0.22882067710161208,
      "num_tokens": 59171819.0,
      "step": 32170
    },
    {
      "entropy": 5.019270706176758,
      "epoch": 2.6875208820581356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042768644086872384,
      "loss": 4.5951,
      "mean_token_accuracy": 0.22965466529130935,
      "num_tokens": 59180753.0,
      "step": 32175
    },
    {
      "entropy": 4.98876371383667,
      "epoch": 2.687938523220849,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004276645734209984,
      "loss": 4.6224,
      "mean_token_accuracy": 0.231885227560997,
      "num_tokens": 59189012.0,
      "step": 32180
    },
    {
      "entropy": 5.03987979888916,
      "epoch": 2.6883561643835616,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000427642703300696,
      "loss": 4.7826,
      "mean_token_accuracy": 0.22370702922344207,
      "num_tokens": 59199006.0,
      "step": 32185
    },
    {
      "entropy": 5.1082813262939455,
      "epoch": 2.6887738055462744,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004276208305081993,
      "loss": 4.7242,
      "mean_token_accuracy": 0.2255721554160118,
      "num_tokens": 59209938.0,
      "step": 32190
    },
    {
      "entropy": 5.106615686416626,
      "epoch": 2.6891914467089877,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004275989550438912,
      "loss": 4.7689,
      "mean_token_accuracy": 0.22149083912372589,
      "num_tokens": 59218843.0,
      "step": 32195
    },
    {
      "entropy": 4.927493476867676,
      "epoch": 2.6896090878717005,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004275770769081547,
      "loss": 4.5346,
      "mean_token_accuracy": 0.23575957864522934,
      "num_tokens": 59227581.0,
      "step": 32200
    },
    {
      "entropy": 5.020770645141601,
      "epoch": 2.6900267290344138,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004275551961013729,
      "loss": 4.7027,
      "mean_token_accuracy": 0.23305142521858216,
      "num_tokens": 59236416.0,
      "step": 32205
    },
    {
      "entropy": 4.981162357330322,
      "epoch": 2.6904443701971266,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042753331262392866,
      "loss": 4.5837,
      "mean_token_accuracy": 0.2377652794122696,
      "num_tokens": 59245780.0,
      "step": 32210
    },
    {
      "entropy": 4.950992774963379,
      "epoch": 2.6908620113598394,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042751142647620515,
      "loss": 4.56,
      "mean_token_accuracy": 0.2367936357855797,
      "num_tokens": 59254325.0,
      "step": 32215
    },
    {
      "entropy": 4.931535816192627,
      "epoch": 2.6912796525225526,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042748953765858565,
      "loss": 4.6618,
      "mean_token_accuracy": 0.22213598042726518,
      "num_tokens": 59263819.0,
      "step": 32220
    },
    {
      "entropy": 4.970761823654175,
      "epoch": 2.691697293685266,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004274676461714532,
      "loss": 4.594,
      "mean_token_accuracy": 0.22691826224327089,
      "num_tokens": 59274138.0,
      "step": 32225
    },
    {
      "entropy": 5.038632440567016,
      "epoch": 2.6921149348479787,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042744575201519115,
      "loss": 4.6283,
      "mean_token_accuracy": 0.22580550611019135,
      "num_tokens": 59283437.0,
      "step": 32230
    },
    {
      "entropy": 4.92136025428772,
      "epoch": 2.6925325760106915,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042742385519018263,
      "loss": 4.5973,
      "mean_token_accuracy": 0.23923926800489426,
      "num_tokens": 59292430.0,
      "step": 32235
    },
    {
      "entropy": 4.918774127960205,
      "epoch": 2.6929502171734048,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004274019556968111,
      "loss": 4.6073,
      "mean_token_accuracy": 0.23648997396230698,
      "num_tokens": 59302067.0,
      "step": 32240
    },
    {
      "entropy": 5.012346649169922,
      "epoch": 2.6933678583361176,
      "grad_norm": 1.125,
      "learning_rate": 0.00042738005353545986,
      "loss": 4.6632,
      "mean_token_accuracy": 0.22227419912815094,
      "num_tokens": 59310332.0,
      "step": 32245
    },
    {
      "entropy": 5.142246103286743,
      "epoch": 2.693785499498831,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042735814870651235,
      "loss": 4.787,
      "mean_token_accuracy": 0.22174075841903687,
      "num_tokens": 59319705.0,
      "step": 32250
    },
    {
      "entropy": 5.068542385101319,
      "epoch": 2.6942031406615436,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004273362412103521,
      "loss": 4.7793,
      "mean_token_accuracy": 0.22073817104101182,
      "num_tokens": 59329544.0,
      "step": 32255
    },
    {
      "entropy": 4.925544023513794,
      "epoch": 2.6946207818242565,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004273143310473625,
      "loss": 4.5329,
      "mean_token_accuracy": 0.23729879558086395,
      "num_tokens": 59337717.0,
      "step": 32260
    },
    {
      "entropy": 4.882941865921021,
      "epoch": 2.6950384229869697,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042729241821792726,
      "loss": 4.5845,
      "mean_token_accuracy": 0.23156876266002654,
      "num_tokens": 59347331.0,
      "step": 32265
    },
    {
      "entropy": 5.021655035018921,
      "epoch": 2.6954560641496825,
      "grad_norm": 1.0,
      "learning_rate": 0.0004272705027224299,
      "loss": 4.7316,
      "mean_token_accuracy": 0.22750356495380403,
      "num_tokens": 59357099.0,
      "step": 32270
    },
    {
      "entropy": 5.081203937530518,
      "epoch": 2.695873705312396,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004272485845612541,
      "loss": 4.6772,
      "mean_token_accuracy": 0.226670603454113,
      "num_tokens": 59366758.0,
      "step": 32275
    },
    {
      "entropy": 4.9944569110870365,
      "epoch": 2.6962913464751086,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042722666373478356,
      "loss": 4.5863,
      "mean_token_accuracy": 0.23724969923496247,
      "num_tokens": 59376016.0,
      "step": 32280
    },
    {
      "entropy": 5.058715724945069,
      "epoch": 2.6967089876378214,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042720474024340204,
      "loss": 4.7216,
      "mean_token_accuracy": 0.22601791024208068,
      "num_tokens": 59384915.0,
      "step": 32285
    },
    {
      "entropy": 5.051024007797241,
      "epoch": 2.6971266288005347,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004271828140874934,
      "loss": 4.8105,
      "mean_token_accuracy": 0.22225821614265442,
      "num_tokens": 59394825.0,
      "step": 32290
    },
    {
      "entropy": 5.022205543518067,
      "epoch": 2.6975442699632475,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042716088526744127,
      "loss": 4.726,
      "mean_token_accuracy": 0.22599008977413176,
      "num_tokens": 59404011.0,
      "step": 32295
    },
    {
      "entropy": 4.957924842834473,
      "epoch": 2.6979619111259607,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004271389537836299,
      "loss": 4.5921,
      "mean_token_accuracy": 0.23726934045553208,
      "num_tokens": 59412842.0,
      "step": 32300
    },
    {
      "entropy": 4.889598751068116,
      "epoch": 2.6983795522886735,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042711701963644283,
      "loss": 4.5911,
      "mean_token_accuracy": 0.24042753577232362,
      "num_tokens": 59420760.0,
      "step": 32305
    },
    {
      "entropy": 5.0087439060211185,
      "epoch": 2.6987971934513864,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042709508282626424,
      "loss": 4.6505,
      "mean_token_accuracy": 0.22889013588428497,
      "num_tokens": 59430134.0,
      "step": 32310
    },
    {
      "entropy": 5.03376407623291,
      "epoch": 2.6992148346140996,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042707314335347827,
      "loss": 4.7256,
      "mean_token_accuracy": 0.22932415157556535,
      "num_tokens": 59440078.0,
      "step": 32315
    },
    {
      "entropy": 4.9876610279083256,
      "epoch": 2.6996324757768124,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042705120121846877,
      "loss": 4.6505,
      "mean_token_accuracy": 0.23569780588150024,
      "num_tokens": 59449019.0,
      "step": 32320
    },
    {
      "entropy": 5.01930513381958,
      "epoch": 2.7000501169395257,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042702925642162,
      "loss": 4.6764,
      "mean_token_accuracy": 0.22599948495626448,
      "num_tokens": 59458498.0,
      "step": 32325
    },
    {
      "entropy": 5.019077157974243,
      "epoch": 2.7004677581022385,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042700730896331606,
      "loss": 4.6457,
      "mean_token_accuracy": 0.23289308249950408,
      "num_tokens": 59468286.0,
      "step": 32330
    },
    {
      "entropy": 5.006918334960938,
      "epoch": 2.7008853992649513,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004269853588439413,
      "loss": 4.5909,
      "mean_token_accuracy": 0.23185401260852814,
      "num_tokens": 59477083.0,
      "step": 32335
    },
    {
      "entropy": 4.976197290420532,
      "epoch": 2.7013030404276646,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042696340606387977,
      "loss": 4.6219,
      "mean_token_accuracy": 0.23109484910964967,
      "num_tokens": 59486380.0,
      "step": 32340
    },
    {
      "entropy": 4.952698659896851,
      "epoch": 2.701720681590378,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000426941450623516,
      "loss": 4.5874,
      "mean_token_accuracy": 0.24006399214267732,
      "num_tokens": 59495674.0,
      "step": 32345
    },
    {
      "entropy": 5.028787279129029,
      "epoch": 2.7021383227530906,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042691949252323427,
      "loss": 4.7015,
      "mean_token_accuracy": 0.21981089413166047,
      "num_tokens": 59504328.0,
      "step": 32350
    },
    {
      "entropy": 5.006440877914429,
      "epoch": 2.7025559639158034,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004268975317634188,
      "loss": 4.7091,
      "mean_token_accuracy": 0.22918360084295272,
      "num_tokens": 59513006.0,
      "step": 32355
    },
    {
      "entropy": 5.062955713272094,
      "epoch": 2.7029736050785167,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042687556834445435,
      "loss": 4.7701,
      "mean_token_accuracy": 0.22454809844493867,
      "num_tokens": 59522752.0,
      "step": 32360
    },
    {
      "entropy": 4.976297807693482,
      "epoch": 2.7033912462412295,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004268536022667252,
      "loss": 4.5812,
      "mean_token_accuracy": 0.23662036210298537,
      "num_tokens": 59532027.0,
      "step": 32365
    },
    {
      "entropy": 5.0068408966064455,
      "epoch": 2.7038088874039428,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042683163353061595,
      "loss": 4.735,
      "mean_token_accuracy": 0.22674973756074907,
      "num_tokens": 59540956.0,
      "step": 32370
    },
    {
      "entropy": 4.981815910339355,
      "epoch": 2.7042265285666556,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042680966213651126,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2233647122979164,
      "num_tokens": 59550821.0,
      "step": 32375
    },
    {
      "entropy": 4.9816389083862305,
      "epoch": 2.7046441697293684,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004267876880847956,
      "loss": 4.565,
      "mean_token_accuracy": 0.24181223213672637,
      "num_tokens": 59559484.0,
      "step": 32380
    },
    {
      "entropy": 5.039239931106567,
      "epoch": 2.7050618108920816,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004267657113758538,
      "loss": 4.699,
      "mean_token_accuracy": 0.22594345659017562,
      "num_tokens": 59568485.0,
      "step": 32385
    },
    {
      "entropy": 5.150718688964844,
      "epoch": 2.7054794520547945,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042674373201007055,
      "loss": 4.8173,
      "mean_token_accuracy": 0.21419045329093933,
      "num_tokens": 59578991.0,
      "step": 32390
    },
    {
      "entropy": 5.017487096786499,
      "epoch": 2.7058970932175077,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004267217499878306,
      "loss": 4.6517,
      "mean_token_accuracy": 0.22870827317237855,
      "num_tokens": 59588164.0,
      "step": 32395
    },
    {
      "entropy": 5.018115758895874,
      "epoch": 2.7063147343802205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042669976530951883,
      "loss": 4.8052,
      "mean_token_accuracy": 0.22953962087631224,
      "num_tokens": 59597149.0,
      "step": 32400
    },
    {
      "entropy": 5.06482629776001,
      "epoch": 2.7067323755429333,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042667777797552003,
      "loss": 4.724,
      "mean_token_accuracy": 0.22789444774389267,
      "num_tokens": 59606110.0,
      "step": 32405
    },
    {
      "entropy": 5.055538654327393,
      "epoch": 2.7071500167056466,
      "grad_norm": 1.125,
      "learning_rate": 0.00042665578798621915,
      "loss": 4.7221,
      "mean_token_accuracy": 0.2294059306383133,
      "num_tokens": 59614822.0,
      "step": 32410
    },
    {
      "entropy": 5.0169206142425535,
      "epoch": 2.7075676578683594,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042663379534200113,
      "loss": 4.7091,
      "mean_token_accuracy": 0.23214015662670134,
      "num_tokens": 59623095.0,
      "step": 32415
    },
    {
      "entropy": 4.98992805480957,
      "epoch": 2.7079852990310727,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004266118000432511,
      "loss": 4.6257,
      "mean_token_accuracy": 0.22829171717166902,
      "num_tokens": 59632899.0,
      "step": 32420
    },
    {
      "entropy": 4.945273971557617,
      "epoch": 2.7084029401937855,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042658980209035394,
      "loss": 4.6546,
      "mean_token_accuracy": 0.2330899342894554,
      "num_tokens": 59642049.0,
      "step": 32425
    },
    {
      "entropy": 5.064625883102417,
      "epoch": 2.7088205813564983,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042656780148369477,
      "loss": 4.7676,
      "mean_token_accuracy": 0.215540412068367,
      "num_tokens": 59651716.0,
      "step": 32430
    },
    {
      "entropy": 5.025696659088135,
      "epoch": 2.7092382225192115,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004265457982236589,
      "loss": 4.7385,
      "mean_token_accuracy": 0.2358606591820717,
      "num_tokens": 59661785.0,
      "step": 32435
    },
    {
      "entropy": 5.057707929611206,
      "epoch": 2.7096558636819243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004265237923106314,
      "loss": 4.6581,
      "mean_token_accuracy": 0.23093579560518265,
      "num_tokens": 59671907.0,
      "step": 32440
    },
    {
      "entropy": 4.9310078620910645,
      "epoch": 2.7100735048446376,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042650178374499753,
      "loss": 4.6309,
      "mean_token_accuracy": 0.23653926104307174,
      "num_tokens": 59680429.0,
      "step": 32445
    },
    {
      "entropy": 5.018417167663574,
      "epoch": 2.7104911460073504,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042647977252714247,
      "loss": 4.7573,
      "mean_token_accuracy": 0.22600536197423934,
      "num_tokens": 59690530.0,
      "step": 32450
    },
    {
      "entropy": 5.066075992584229,
      "epoch": 2.7109087871700632,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004264577586574518,
      "loss": 4.6335,
      "mean_token_accuracy": 0.2315348729491234,
      "num_tokens": 59699315.0,
      "step": 32455
    },
    {
      "entropy": 5.05024471282959,
      "epoch": 2.7113264283327765,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042643574213631065,
      "loss": 4.6394,
      "mean_token_accuracy": 0.23484901636838912,
      "num_tokens": 59708314.0,
      "step": 32460
    },
    {
      "entropy": 4.98932991027832,
      "epoch": 2.7117440694954897,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004264137229641046,
      "loss": 4.6479,
      "mean_token_accuracy": 0.23488865345716475,
      "num_tokens": 59718597.0,
      "step": 32465
    },
    {
      "entropy": 5.05079984664917,
      "epoch": 2.7121617106582026,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000426391701141219,
      "loss": 4.7883,
      "mean_token_accuracy": 0.2173192471265793,
      "num_tokens": 59728345.0,
      "step": 32470
    },
    {
      "entropy": 4.999870634078979,
      "epoch": 2.7125793518209154,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042636967666803954,
      "loss": 4.5596,
      "mean_token_accuracy": 0.2407701402902603,
      "num_tokens": 59736370.0,
      "step": 32475
    },
    {
      "entropy": 4.961795091629028,
      "epoch": 2.7129969929836286,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042634764954495167,
      "loss": 4.7676,
      "mean_token_accuracy": 0.22041718810796737,
      "num_tokens": 59745628.0,
      "step": 32480
    },
    {
      "entropy": 4.9953492164611815,
      "epoch": 2.7134146341463414,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042632561977234095,
      "loss": 4.6745,
      "mean_token_accuracy": 0.22960835248231887,
      "num_tokens": 59754280.0,
      "step": 32485
    },
    {
      "entropy": 5.178605079650879,
      "epoch": 2.7138322753090547,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004263035873505932,
      "loss": 4.8105,
      "mean_token_accuracy": 0.21688320934772493,
      "num_tokens": 59763528.0,
      "step": 32490
    },
    {
      "entropy": 5.068901062011719,
      "epoch": 2.7142499164717675,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004262815522800939,
      "loss": 4.6735,
      "mean_token_accuracy": 0.2304905205965042,
      "num_tokens": 59773375.0,
      "step": 32495
    },
    {
      "entropy": 4.9983748435974125,
      "epoch": 2.7146675576344803,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004262595145612291,
      "loss": 4.6732,
      "mean_token_accuracy": 0.23166654706001283,
      "num_tokens": 59782795.0,
      "step": 32500
    },
    {
      "entropy": 4.9074183940887455,
      "epoch": 2.7150851987971936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042623747419438427,
      "loss": 4.6069,
      "mean_token_accuracy": 0.23501864820718765,
      "num_tokens": 59791999.0,
      "step": 32505
    },
    {
      "entropy": 5.046203565597534,
      "epoch": 2.7155028399599064,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004262154311799455,
      "loss": 4.7327,
      "mean_token_accuracy": 0.23317863345146178,
      "num_tokens": 59802547.0,
      "step": 32510
    },
    {
      "entropy": 5.024926948547363,
      "epoch": 2.7159204811226196,
      "grad_norm": 1.0,
      "learning_rate": 0.0004261933855182985,
      "loss": 4.6252,
      "mean_token_accuracy": 0.2313526377081871,
      "num_tokens": 59811570.0,
      "step": 32515
    },
    {
      "entropy": 4.9905863285064695,
      "epoch": 2.7163381222853324,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042617133720982943,
      "loss": 4.6533,
      "mean_token_accuracy": 0.22336269170045853,
      "num_tokens": 59820258.0,
      "step": 32520
    },
    {
      "entropy": 5.0244231700897215,
      "epoch": 2.7167557634480453,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042614928625492397,
      "loss": 4.6968,
      "mean_token_accuracy": 0.22272728234529496,
      "num_tokens": 59831089.0,
      "step": 32525
    },
    {
      "entropy": 4.991362905502319,
      "epoch": 2.7171734046107585,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042612723265396846,
      "loss": 4.6695,
      "mean_token_accuracy": 0.23201386332511903,
      "num_tokens": 59840276.0,
      "step": 32530
    },
    {
      "entropy": 5.035800886154175,
      "epoch": 2.7175910457734713,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004261051764073488,
      "loss": 4.725,
      "mean_token_accuracy": 0.22778797745704651,
      "num_tokens": 59849571.0,
      "step": 32535
    },
    {
      "entropy": 4.9997601985931395,
      "epoch": 2.7180086869361846,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042608311751545115,
      "loss": 4.6351,
      "mean_token_accuracy": 0.23124848753213884,
      "num_tokens": 59858092.0,
      "step": 32540
    },
    {
      "entropy": 4.984669923782349,
      "epoch": 2.7184263280988974,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042606105597866163,
      "loss": 4.578,
      "mean_token_accuracy": 0.2315923348069191,
      "num_tokens": 59866824.0,
      "step": 32545
    },
    {
      "entropy": 4.985531663894653,
      "epoch": 2.71884396926161,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004260389917973665,
      "loss": 4.7173,
      "mean_token_accuracy": 0.2250983700156212,
      "num_tokens": 59876209.0,
      "step": 32550
    },
    {
      "entropy": 5.043226289749145,
      "epoch": 2.7192616104243235,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000426016924971952,
      "loss": 4.7521,
      "mean_token_accuracy": 0.2174416035413742,
      "num_tokens": 59885588.0,
      "step": 32555
    },
    {
      "entropy": 4.995805120468139,
      "epoch": 2.7196792515870363,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042599485550280446,
      "loss": 4.6172,
      "mean_token_accuracy": 0.23766996711492538,
      "num_tokens": 59894507.0,
      "step": 32560
    },
    {
      "entropy": 5.026371622085572,
      "epoch": 2.7200968927497495,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042597278339031024,
      "loss": 4.7315,
      "mean_token_accuracy": 0.22127582430839537,
      "num_tokens": 59903657.0,
      "step": 32565
    },
    {
      "entropy": 5.122867107391357,
      "epoch": 2.7205145339124623,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004259507086348557,
      "loss": 4.7791,
      "mean_token_accuracy": 0.21356049329042434,
      "num_tokens": 59914062.0,
      "step": 32570
    },
    {
      "entropy": 4.963083982467651,
      "epoch": 2.720932175075175,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042592863123682735,
      "loss": 4.6024,
      "mean_token_accuracy": 0.2380631983280182,
      "num_tokens": 59922571.0,
      "step": 32575
    },
    {
      "entropy": 4.961720085144043,
      "epoch": 2.7213498162378884,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004259065511966116,
      "loss": 4.6888,
      "mean_token_accuracy": 0.23227466344833375,
      "num_tokens": 59932763.0,
      "step": 32580
    },
    {
      "entropy": 4.9776535987854,
      "epoch": 2.7217674574006017,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000425884468514595,
      "loss": 4.6486,
      "mean_token_accuracy": 0.2281242161989212,
      "num_tokens": 59942312.0,
      "step": 32585
    },
    {
      "entropy": 4.944695281982422,
      "epoch": 2.7221850985633145,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042586238319116423,
      "loss": 4.6057,
      "mean_token_accuracy": 0.23336426615715028,
      "num_tokens": 59951102.0,
      "step": 32590
    },
    {
      "entropy": 4.887078475952149,
      "epoch": 2.7226027397260273,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042584029522670574,
      "loss": 4.5493,
      "mean_token_accuracy": 0.23379187732934953,
      "num_tokens": 59960160.0,
      "step": 32595
    },
    {
      "entropy": 4.99014720916748,
      "epoch": 2.7230203808887405,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004258182046216064,
      "loss": 4.6224,
      "mean_token_accuracy": 0.2313872754573822,
      "num_tokens": 59969376.0,
      "step": 32600
    },
    {
      "entropy": 5.019026899337769,
      "epoch": 2.7234380220514534,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004257961113762528,
      "loss": 4.7293,
      "mean_token_accuracy": 0.21924201399087906,
      "num_tokens": 59978605.0,
      "step": 32605
    },
    {
      "entropy": 4.90643720626831,
      "epoch": 2.7238556632141666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004257740154910318,
      "loss": 4.5517,
      "mean_token_accuracy": 0.23919350653886795,
      "num_tokens": 59987471.0,
      "step": 32610
    },
    {
      "entropy": 5.00980396270752,
      "epoch": 2.7242733043768794,
      "grad_norm": 1.125,
      "learning_rate": 0.0004257519169663302,
      "loss": 4.6889,
      "mean_token_accuracy": 0.22887759655714035,
      "num_tokens": 59995750.0,
      "step": 32615
    },
    {
      "entropy": 4.968307542800903,
      "epoch": 2.7246909455395922,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042572981580253475,
      "loss": 4.5953,
      "mean_token_accuracy": 0.24106734991073608,
      "num_tokens": 60004871.0,
      "step": 32620
    },
    {
      "entropy": 4.90307126045227,
      "epoch": 2.7251085867023055,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042570771200003255,
      "loss": 4.5986,
      "mean_token_accuracy": 0.23933605700731278,
      "num_tokens": 60013751.0,
      "step": 32625
    },
    {
      "entropy": 5.0673065185546875,
      "epoch": 2.7255262278650183,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042568560555921033,
      "loss": 4.7268,
      "mean_token_accuracy": 0.2232468992471695,
      "num_tokens": 60023019.0,
      "step": 32630
    },
    {
      "entropy": 4.932082462310791,
      "epoch": 2.7259438690277316,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004256634964804553,
      "loss": 4.5076,
      "mean_token_accuracy": 0.24247115850448608,
      "num_tokens": 60032531.0,
      "step": 32635
    },
    {
      "entropy": 4.997700452804565,
      "epoch": 2.7263615101904444,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042564138476415433,
      "loss": 4.7335,
      "mean_token_accuracy": 0.21963980942964553,
      "num_tokens": 60042540.0,
      "step": 32640
    },
    {
      "entropy": 5.037789916992187,
      "epoch": 2.726779151353157,
      "grad_norm": 1.125,
      "learning_rate": 0.00042561927041069467,
      "loss": 4.6932,
      "mean_token_accuracy": 0.22886464446783067,
      "num_tokens": 60051280.0,
      "step": 32645
    },
    {
      "entropy": 5.01505126953125,
      "epoch": 2.7271967925158704,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042559715342046334,
      "loss": 4.6403,
      "mean_token_accuracy": 0.22908636480569838,
      "num_tokens": 60060891.0,
      "step": 32650
    },
    {
      "entropy": 5.0045982837677006,
      "epoch": 2.7276144336785833,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004255750337938475,
      "loss": 4.7206,
      "mean_token_accuracy": 0.22465914487838745,
      "num_tokens": 60070011.0,
      "step": 32655
    },
    {
      "entropy": 5.072884035110474,
      "epoch": 2.7280320748412965,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042555291153123455,
      "loss": 4.7974,
      "mean_token_accuracy": 0.21866463720798493,
      "num_tokens": 60080255.0,
      "step": 32660
    },
    {
      "entropy": 5.077678394317627,
      "epoch": 2.7284497160040093,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042553078663301163,
      "loss": 4.6955,
      "mean_token_accuracy": 0.22412512451410294,
      "num_tokens": 60089907.0,
      "step": 32665
    },
    {
      "entropy": 5.120520162582397,
      "epoch": 2.728867357166722,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042550865909956606,
      "loss": 4.7593,
      "mean_token_accuracy": 0.22327315658330918,
      "num_tokens": 60099427.0,
      "step": 32670
    },
    {
      "entropy": 4.997345685958862,
      "epoch": 2.7292849983294354,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042548652893128526,
      "loss": 4.6263,
      "mean_token_accuracy": 0.23574073612689972,
      "num_tokens": 60107197.0,
      "step": 32675
    },
    {
      "entropy": 4.932960271835327,
      "epoch": 2.729702639492148,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004254643961285567,
      "loss": 4.7414,
      "mean_token_accuracy": 0.22109083235263824,
      "num_tokens": 60116267.0,
      "step": 32680
    },
    {
      "entropy": 5.124262189865112,
      "epoch": 2.7301202806548615,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042544226069176765,
      "loss": 4.7737,
      "mean_token_accuracy": 0.2265182614326477,
      "num_tokens": 60125772.0,
      "step": 32685
    },
    {
      "entropy": 5.133836698532105,
      "epoch": 2.7305379218175743,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004254201226213059,
      "loss": 4.7929,
      "mean_token_accuracy": 0.21590634137392045,
      "num_tokens": 60134727.0,
      "step": 32690
    },
    {
      "entropy": 5.005365514755249,
      "epoch": 2.730955562980287,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004253979819175587,
      "loss": 4.7616,
      "mean_token_accuracy": 0.22935348600149155,
      "num_tokens": 60144332.0,
      "step": 32695
    },
    {
      "entropy": 5.023284482955932,
      "epoch": 2.7313732041430003,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042537583858091383,
      "loss": 4.7076,
      "mean_token_accuracy": 0.22768335938453674,
      "num_tokens": 60154620.0,
      "step": 32700
    },
    {
      "entropy": 5.0025464534759525,
      "epoch": 2.7317908453057136,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004253536926117589,
      "loss": 4.6353,
      "mean_token_accuracy": 0.2326085940003395,
      "num_tokens": 60164442.0,
      "step": 32705
    },
    {
      "entropy": 4.971075057983398,
      "epoch": 2.7322084864684264,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004253315440104816,
      "loss": 4.6236,
      "mean_token_accuracy": 0.23705478012561798,
      "num_tokens": 60172688.0,
      "step": 32710
    },
    {
      "entropy": 4.972922849655151,
      "epoch": 2.732626127631139,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042530939277746964,
      "loss": 4.6862,
      "mean_token_accuracy": 0.22586483955383302,
      "num_tokens": 60182730.0,
      "step": 32715
    },
    {
      "entropy": 5.0698377132415775,
      "epoch": 2.7330437687938525,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004252872389131109,
      "loss": 4.708,
      "mean_token_accuracy": 0.22386007159948348,
      "num_tokens": 60192571.0,
      "step": 32720
    },
    {
      "entropy": 5.110123968124389,
      "epoch": 2.7334614099565653,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004252650824177931,
      "loss": 4.7084,
      "mean_token_accuracy": 0.22898343652486802,
      "num_tokens": 60202857.0,
      "step": 32725
    },
    {
      "entropy": 5.105692291259766,
      "epoch": 2.7338790511192785,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004252429232919042,
      "loss": 4.7277,
      "mean_token_accuracy": 0.22808421701192855,
      "num_tokens": 60211981.0,
      "step": 32730
    },
    {
      "entropy": 4.986785459518432,
      "epoch": 2.7342966922819913,
      "grad_norm": 1.09375,
      "learning_rate": 0.000425220761535832,
      "loss": 4.5705,
      "mean_token_accuracy": 0.2387503743171692,
      "num_tokens": 60220904.0,
      "step": 32735
    },
    {
      "entropy": 4.939090919494629,
      "epoch": 2.734714333444704,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004251985971499647,
      "loss": 4.6382,
      "mean_token_accuracy": 0.23323111236095428,
      "num_tokens": 60230239.0,
      "step": 32740
    },
    {
      "entropy": 4.964633989334106,
      "epoch": 2.7351319746074174,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042517643013469,
      "loss": 4.6971,
      "mean_token_accuracy": 0.23518436700105666,
      "num_tokens": 60239991.0,
      "step": 32745
    },
    {
      "entropy": 5.070864534378051,
      "epoch": 2.7355496157701302,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004251542604903962,
      "loss": 4.65,
      "mean_token_accuracy": 0.23257865756750107,
      "num_tokens": 60248826.0,
      "step": 32750
    },
    {
      "entropy": 4.965701055526734,
      "epoch": 2.7359672569328435,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004251320882174713,
      "loss": 4.6233,
      "mean_token_accuracy": 0.2363791808485985,
      "num_tokens": 60257033.0,
      "step": 32755
    },
    {
      "entropy": 5.119166088104248,
      "epoch": 2.7363848980955563,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004251099133163035,
      "loss": 4.8256,
      "mean_token_accuracy": 0.21936115026473998,
      "num_tokens": 60265973.0,
      "step": 32760
    },
    {
      "entropy": 5.0588319301605225,
      "epoch": 2.736802539258269,
      "grad_norm": 1.015625,
      "learning_rate": 0.000425087735787281,
      "loss": 4.6216,
      "mean_token_accuracy": 0.2321573957800865,
      "num_tokens": 60275065.0,
      "step": 32765
    },
    {
      "entropy": 5.032834815979004,
      "epoch": 2.7372201804209824,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042506555563079194,
      "loss": 4.6993,
      "mean_token_accuracy": 0.23218956142663955,
      "num_tokens": 60285158.0,
      "step": 32770
    },
    {
      "entropy": 4.94688606262207,
      "epoch": 2.737637821583695,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004250433728472248,
      "loss": 4.5503,
      "mean_token_accuracy": 0.24663546234369277,
      "num_tokens": 60294258.0,
      "step": 32775
    },
    {
      "entropy": 4.994598293304444,
      "epoch": 2.7380554627464084,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004250211874369678,
      "loss": 4.6122,
      "mean_token_accuracy": 0.24339343905448912,
      "num_tokens": 60302809.0,
      "step": 32780
    },
    {
      "entropy": 4.897377300262451,
      "epoch": 2.7384731039091212,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004249989994004093,
      "loss": 4.5443,
      "mean_token_accuracy": 0.24793113470077516,
      "num_tokens": 60311291.0,
      "step": 32785
    },
    {
      "entropy": 5.047710990905761,
      "epoch": 2.738890745071834,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042497680873793776,
      "loss": 4.7422,
      "mean_token_accuracy": 0.22948309928178787,
      "num_tokens": 60320460.0,
      "step": 32790
    },
    {
      "entropy": 5.010602235794067,
      "epoch": 2.7393083862345473,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004249546154499416,
      "loss": 4.6142,
      "mean_token_accuracy": 0.2349050909280777,
      "num_tokens": 60328538.0,
      "step": 32795
    },
    {
      "entropy": 4.98354811668396,
      "epoch": 2.73972602739726,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042493241953680947,
      "loss": 4.6752,
      "mean_token_accuracy": 0.23084231913089753,
      "num_tokens": 60337182.0,
      "step": 32800
    },
    {
      "entropy": 4.900496530532837,
      "epoch": 2.7401436685599734,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042491022099892985,
      "loss": 4.5474,
      "mean_token_accuracy": 0.23764651119709015,
      "num_tokens": 60344962.0,
      "step": 32805
    },
    {
      "entropy": 5.071768808364868,
      "epoch": 2.740561309722686,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004248880198366913,
      "loss": 4.7214,
      "mean_token_accuracy": 0.22391000241041184,
      "num_tokens": 60355211.0,
      "step": 32810
    },
    {
      "entropy": 5.02468638420105,
      "epoch": 2.740978950885399,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004248658160504827,
      "loss": 4.6634,
      "mean_token_accuracy": 0.2297162875533104,
      "num_tokens": 60364976.0,
      "step": 32815
    },
    {
      "entropy": 5.002512216567993,
      "epoch": 2.7413965920481123,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042484360964069247,
      "loss": 4.6709,
      "mean_token_accuracy": 0.23238217681646348,
      "num_tokens": 60374605.0,
      "step": 32820
    },
    {
      "entropy": 4.965276622772217,
      "epoch": 2.7418142332108255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004248214006077095,
      "loss": 4.6402,
      "mean_token_accuracy": 0.23510123342275618,
      "num_tokens": 60383911.0,
      "step": 32825
    },
    {
      "entropy": 5.066047859191895,
      "epoch": 2.7422318743735383,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042479918895192253,
      "loss": 4.6266,
      "mean_token_accuracy": 0.23344129472970962,
      "num_tokens": 60392896.0,
      "step": 32830
    },
    {
      "entropy": 5.004463338851929,
      "epoch": 2.742649515536251,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042477697467372056,
      "loss": 4.6517,
      "mean_token_accuracy": 0.2257180020213127,
      "num_tokens": 60401670.0,
      "step": 32835
    },
    {
      "entropy": 4.928977966308594,
      "epoch": 2.7430671566989644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042475475777349234,
      "loss": 4.6069,
      "mean_token_accuracy": 0.24008703827857972,
      "num_tokens": 60410130.0,
      "step": 32840
    },
    {
      "entropy": 4.98634967803955,
      "epoch": 2.743484797861677,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042473253825162674,
      "loss": 4.69,
      "mean_token_accuracy": 0.23158777356147767,
      "num_tokens": 60419223.0,
      "step": 32845
    },
    {
      "entropy": 5.06792631149292,
      "epoch": 2.7439024390243905,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004247103161085129,
      "loss": 4.6381,
      "mean_token_accuracy": 0.23268176913261412,
      "num_tokens": 60428165.0,
      "step": 32850
    },
    {
      "entropy": 5.014925336837768,
      "epoch": 2.7443200801871033,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042468809134453973,
      "loss": 4.725,
      "mean_token_accuracy": 0.22361446022987366,
      "num_tokens": 60436633.0,
      "step": 32855
    },
    {
      "entropy": 5.018450832366943,
      "epoch": 2.744737721349816,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004246658639600963,
      "loss": 4.7377,
      "mean_token_accuracy": 0.2284717693924904,
      "num_tokens": 60445691.0,
      "step": 32860
    },
    {
      "entropy": 5.04145655632019,
      "epoch": 2.7451553625125293,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004246436339555719,
      "loss": 4.7044,
      "mean_token_accuracy": 0.22270301431417466,
      "num_tokens": 60454692.0,
      "step": 32865
    },
    {
      "entropy": 5.036565637588501,
      "epoch": 2.745573003675242,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042462140133135544,
      "loss": 4.6926,
      "mean_token_accuracy": 0.23215589672327042,
      "num_tokens": 60463161.0,
      "step": 32870
    },
    {
      "entropy": 5.013756656646729,
      "epoch": 2.7459906448379554,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004245991660878363,
      "loss": 4.657,
      "mean_token_accuracy": 0.23055509775876998,
      "num_tokens": 60471838.0,
      "step": 32875
    },
    {
      "entropy": 5.073283386230469,
      "epoch": 2.746408286000668,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004245769282254037,
      "loss": 4.6634,
      "mean_token_accuracy": 0.22395621240139008,
      "num_tokens": 60481016.0,
      "step": 32880
    },
    {
      "entropy": 5.043162441253662,
      "epoch": 2.746825927163381,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042455468774444695,
      "loss": 4.7495,
      "mean_token_accuracy": 0.23009424209594725,
      "num_tokens": 60490046.0,
      "step": 32885
    },
    {
      "entropy": 5.008092880249023,
      "epoch": 2.7472435683260943,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042453244464535524,
      "loss": 4.6629,
      "mean_token_accuracy": 0.22944066822528839,
      "num_tokens": 60499042.0,
      "step": 32890
    },
    {
      "entropy": 4.952996349334716,
      "epoch": 2.747661209488807,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004245101989285182,
      "loss": 4.6473,
      "mean_token_accuracy": 0.23037095218896866,
      "num_tokens": 60507469.0,
      "step": 32895
    },
    {
      "entropy": 4.940918874740601,
      "epoch": 2.7480788506515204,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004244879505943252,
      "loss": 4.6505,
      "mean_token_accuracy": 0.23247134536504746,
      "num_tokens": 60516122.0,
      "step": 32900
    },
    {
      "entropy": 5.072495651245117,
      "epoch": 2.748496491814233,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042446569964316557,
      "loss": 4.7957,
      "mean_token_accuracy": 0.21823005080223085,
      "num_tokens": 60526534.0,
      "step": 32905
    },
    {
      "entropy": 5.007509517669678,
      "epoch": 2.748914132976946,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000424443446075429,
      "loss": 4.6203,
      "mean_token_accuracy": 0.24088159054517747,
      "num_tokens": 60535530.0,
      "step": 32910
    },
    {
      "entropy": 5.153663778305054,
      "epoch": 2.7493317741396592,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004244211898915051,
      "loss": 4.823,
      "mean_token_accuracy": 0.21942740678787231,
      "num_tokens": 60545639.0,
      "step": 32915
    },
    {
      "entropy": 5.109315204620361,
      "epoch": 2.749749415302372,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004243989310917833,
      "loss": 4.7273,
      "mean_token_accuracy": 0.22569006383419038,
      "num_tokens": 60555229.0,
      "step": 32920
    },
    {
      "entropy": 5.053382539749146,
      "epoch": 2.7501670564650853,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042437666967665345,
      "loss": 4.638,
      "mean_token_accuracy": 0.23162438571453095,
      "num_tokens": 60564981.0,
      "step": 32925
    },
    {
      "entropy": 4.940986251831054,
      "epoch": 2.750584697627798,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042435440564650526,
      "loss": 4.5896,
      "mean_token_accuracy": 0.2333654135465622,
      "num_tokens": 60574023.0,
      "step": 32930
    },
    {
      "entropy": 4.941262531280517,
      "epoch": 2.751002338790511,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004243321390017283,
      "loss": 4.6847,
      "mean_token_accuracy": 0.23046448081731796,
      "num_tokens": 60583535.0,
      "step": 32935
    },
    {
      "entropy": 4.937882566452027,
      "epoch": 2.751419979953224,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004243098697427125,
      "loss": 4.548,
      "mean_token_accuracy": 0.24022651165723802,
      "num_tokens": 60592102.0,
      "step": 32940
    },
    {
      "entropy": 5.002734231948852,
      "epoch": 2.7518376211159374,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042428759786984773,
      "loss": 4.7023,
      "mean_token_accuracy": 0.22348754703998566,
      "num_tokens": 60601502.0,
      "step": 32945
    },
    {
      "entropy": 5.078869581222534,
      "epoch": 2.7522552622786502,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004242653233835239,
      "loss": 4.6475,
      "mean_token_accuracy": 0.23341959863901138,
      "num_tokens": 60610302.0,
      "step": 32950
    },
    {
      "entropy": 5.070456552505493,
      "epoch": 2.752672903441363,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00042424304628413086,
      "loss": 4.7551,
      "mean_token_accuracy": 0.2278181791305542,
      "num_tokens": 60620014.0,
      "step": 32955
    },
    {
      "entropy": 4.953239917755127,
      "epoch": 2.7530905446040763,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004242207665720587,
      "loss": 4.5396,
      "mean_token_accuracy": 0.24480929225683212,
      "num_tokens": 60628324.0,
      "step": 32960
    },
    {
      "entropy": 4.972448301315308,
      "epoch": 2.753508185766789,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042419848424769737,
      "loss": 4.6602,
      "mean_token_accuracy": 0.23185937255620956,
      "num_tokens": 60638250.0,
      "step": 32965
    },
    {
      "entropy": 4.8910231590271,
      "epoch": 2.7539258269295024,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000424176199311437,
      "loss": 4.5454,
      "mean_token_accuracy": 0.24307310581207275,
      "num_tokens": 60646278.0,
      "step": 32970
    },
    {
      "entropy": 5.014189291000366,
      "epoch": 2.754343468092215,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042415391176366763,
      "loss": 4.6312,
      "mean_token_accuracy": 0.2436388537287712,
      "num_tokens": 60654589.0,
      "step": 32975
    },
    {
      "entropy": 4.983237600326538,
      "epoch": 2.754761109254928,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004241316216047796,
      "loss": 4.6415,
      "mean_token_accuracy": 0.23502724170684813,
      "num_tokens": 60663824.0,
      "step": 32980
    },
    {
      "entropy": 5.026510620117188,
      "epoch": 2.7551787504176413,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042410932883516297,
      "loss": 4.7196,
      "mean_token_accuracy": 0.23006887137889862,
      "num_tokens": 60673973.0,
      "step": 32985
    },
    {
      "entropy": 5.048143339157105,
      "epoch": 2.755596391580354,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004240870334552081,
      "loss": 4.6826,
      "mean_token_accuracy": 0.23382201045751572,
      "num_tokens": 60683291.0,
      "step": 32990
    },
    {
      "entropy": 5.010144376754761,
      "epoch": 2.7560140327430673,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042406473546530523,
      "loss": 4.6763,
      "mean_token_accuracy": 0.2213860645890236,
      "num_tokens": 60692829.0,
      "step": 32995
    },
    {
      "entropy": 5.0675023078918455,
      "epoch": 2.75643167390578,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042404243486584463,
      "loss": 4.6998,
      "mean_token_accuracy": 0.21850354373455047,
      "num_tokens": 60701535.0,
      "step": 33000
    },
    {
      "epoch": 2.75643167390578,
      "eval_entropy": 4.7675416095356296,
      "eval_loss": 4.708503246307373,
      "eval_mean_token_accuracy": 0.23877504833628127,
      "eval_num_tokens": 60701535.0,
      "eval_runtime": 27.7615,
      "eval_samples_per_second": 1360.263,
      "eval_steps_per_second": 170.055,
      "step": 33000
    },
    {
      "entropy": 5.026006650924683,
      "epoch": 2.756849315068493,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00042402013165721697,
      "loss": 4.6585,
      "mean_token_accuracy": 0.22339597195386887,
      "num_tokens": 60710434.0,
      "step": 33005
    },
    {
      "entropy": 4.9279882431030275,
      "epoch": 2.757266956231206,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042399782583981246,
      "loss": 4.6389,
      "mean_token_accuracy": 0.2335094094276428,
      "num_tokens": 60719862.0,
      "step": 33010
    },
    {
      "entropy": 5.003027248382568,
      "epoch": 2.757684597393919,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004239755174140216,
      "loss": 4.7648,
      "mean_token_accuracy": 0.22345812767744064,
      "num_tokens": 60729549.0,
      "step": 33015
    },
    {
      "entropy": 5.088839817047119,
      "epoch": 2.7581022385566323,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042395320638023503,
      "loss": 4.7244,
      "mean_token_accuracy": 0.2273315519094467,
      "num_tokens": 60740855.0,
      "step": 33020
    },
    {
      "entropy": 5.061381387710571,
      "epoch": 2.758519879719345,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042393089273884334,
      "loss": 4.6017,
      "mean_token_accuracy": 0.23512290865182878,
      "num_tokens": 60749274.0,
      "step": 33025
    },
    {
      "entropy": 5.099486064910889,
      "epoch": 2.758937520882058,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004239085764902371,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2255618989467621,
      "num_tokens": 60758347.0,
      "step": 33030
    },
    {
      "entropy": 5.02245602607727,
      "epoch": 2.759355162044771,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000423886257634807,
      "loss": 4.6822,
      "mean_token_accuracy": 0.23505840003490447,
      "num_tokens": 60766454.0,
      "step": 33035
    },
    {
      "entropy": 5.014695453643799,
      "epoch": 2.759772803207484,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004238639361729436,
      "loss": 4.7407,
      "mean_token_accuracy": 0.2251971036195755,
      "num_tokens": 60775659.0,
      "step": 33040
    },
    {
      "entropy": 5.018243265151978,
      "epoch": 2.7601904443701972,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042384161210503793,
      "loss": 4.5973,
      "mean_token_accuracy": 0.22732291519641876,
      "num_tokens": 60785471.0,
      "step": 33045
    },
    {
      "entropy": 5.0334851264953615,
      "epoch": 2.76060808553291,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042381928543148067,
      "loss": 4.6963,
      "mean_token_accuracy": 0.22563712298870087,
      "num_tokens": 60795424.0,
      "step": 33050
    },
    {
      "entropy": 5.007568502426148,
      "epoch": 2.761025726695623,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004237969561526627,
      "loss": 4.6457,
      "mean_token_accuracy": 0.23271620124578477,
      "num_tokens": 60804096.0,
      "step": 33055
    },
    {
      "entropy": 5.030919313430786,
      "epoch": 2.761443367858336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042377462426897486,
      "loss": 4.6122,
      "mean_token_accuracy": 0.22494628876447678,
      "num_tokens": 60812335.0,
      "step": 33060
    },
    {
      "entropy": 4.883628845214844,
      "epoch": 2.7618610090210494,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042375228978080813,
      "loss": 4.5823,
      "mean_token_accuracy": 0.23950987607240676,
      "num_tokens": 60821693.0,
      "step": 33065
    },
    {
      "entropy": 4.922558164596557,
      "epoch": 2.762278650183762,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042372995268855356,
      "loss": 4.5895,
      "mean_token_accuracy": 0.23637034744024277,
      "num_tokens": 60830097.0,
      "step": 33070
    },
    {
      "entropy": 4.991423511505127,
      "epoch": 2.762696291346475,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004237076129926022,
      "loss": 4.664,
      "mean_token_accuracy": 0.23715830296278,
      "num_tokens": 60838760.0,
      "step": 33075
    },
    {
      "entropy": 5.0418556213378904,
      "epoch": 2.7631139325091882,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042368527069334494,
      "loss": 4.66,
      "mean_token_accuracy": 0.22401974648237227,
      "num_tokens": 60847580.0,
      "step": 33080
    },
    {
      "entropy": 5.042490863800049,
      "epoch": 2.763531573671901,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004236629257911731,
      "loss": 4.6784,
      "mean_token_accuracy": 0.23831814676523208,
      "num_tokens": 60856367.0,
      "step": 33085
    },
    {
      "entropy": 5.011930894851685,
      "epoch": 2.7639492148346143,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004236405782864778,
      "loss": 4.6876,
      "mean_token_accuracy": 0.2315009593963623,
      "num_tokens": 60865825.0,
      "step": 33090
    },
    {
      "entropy": 4.890200185775757,
      "epoch": 2.764366855997327,
      "grad_norm": 1.0,
      "learning_rate": 0.0004236182281796503,
      "loss": 4.6568,
      "mean_token_accuracy": 0.23009939789772033,
      "num_tokens": 60874953.0,
      "step": 33095
    },
    {
      "entropy": 4.87938756942749,
      "epoch": 2.76478449716004,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004235958754710818,
      "loss": 4.5923,
      "mean_token_accuracy": 0.23511407077312468,
      "num_tokens": 60884719.0,
      "step": 33100
    },
    {
      "entropy": 5.134215402603149,
      "epoch": 2.765202138322753,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004235735201611636,
      "loss": 4.772,
      "mean_token_accuracy": 0.22092864215373992,
      "num_tokens": 60894294.0,
      "step": 33105
    },
    {
      "entropy": 5.029180431365967,
      "epoch": 2.765619779485466,
      "grad_norm": 1.125,
      "learning_rate": 0.0004235511622502871,
      "loss": 4.6362,
      "mean_token_accuracy": 0.22946296036243438,
      "num_tokens": 60902111.0,
      "step": 33110
    },
    {
      "entropy": 5.033327960968018,
      "epoch": 2.7660374206481793,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004235288017388437,
      "loss": 4.587,
      "mean_token_accuracy": 0.23181069493293763,
      "num_tokens": 60911704.0,
      "step": 33115
    },
    {
      "entropy": 4.875524950027466,
      "epoch": 2.766455061810892,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042350643862722485,
      "loss": 4.5835,
      "mean_token_accuracy": 0.24605355262756348,
      "num_tokens": 60920662.0,
      "step": 33120
    },
    {
      "entropy": 4.89171781539917,
      "epoch": 2.766872702973605,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042348407291582216,
      "loss": 4.5841,
      "mean_token_accuracy": 0.2421452894806862,
      "num_tokens": 60929586.0,
      "step": 33125
    },
    {
      "entropy": 4.990018272399903,
      "epoch": 2.767290344136318,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004234617046050268,
      "loss": 4.6557,
      "mean_token_accuracy": 0.23621955811977385,
      "num_tokens": 60938887.0,
      "step": 33130
    },
    {
      "entropy": 4.976779317855835,
      "epoch": 2.767707985299031,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004234393336952308,
      "loss": 4.6141,
      "mean_token_accuracy": 0.22798570543527602,
      "num_tokens": 60948235.0,
      "step": 33135
    },
    {
      "entropy": 4.911331748962402,
      "epoch": 2.768125626461744,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004234169601868255,
      "loss": 4.5697,
      "mean_token_accuracy": 0.23988277167081834,
      "num_tokens": 60956369.0,
      "step": 33140
    },
    {
      "entropy": 4.960474634170533,
      "epoch": 2.768543267624457,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042339458408020267,
      "loss": 4.6459,
      "mean_token_accuracy": 0.23800712525844575,
      "num_tokens": 60964670.0,
      "step": 33145
    },
    {
      "entropy": 4.936977529525757,
      "epoch": 2.76896090878717,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042337220537575393,
      "loss": 4.5983,
      "mean_token_accuracy": 0.2299118682742119,
      "num_tokens": 60973291.0,
      "step": 33150
    },
    {
      "entropy": 4.989633989334107,
      "epoch": 2.769378549949883,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004233498240738713,
      "loss": 4.6017,
      "mean_token_accuracy": 0.2305191919207573,
      "num_tokens": 60982047.0,
      "step": 33155
    },
    {
      "entropy": 5.014838075637817,
      "epoch": 2.769796191112596,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042332744017494633,
      "loss": 4.7027,
      "mean_token_accuracy": 0.22411265075206757,
      "num_tokens": 60990812.0,
      "step": 33160
    },
    {
      "entropy": 4.983539295196533,
      "epoch": 2.770213832275309,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000423305053679371,
      "loss": 4.6306,
      "mean_token_accuracy": 0.22359195500612258,
      "num_tokens": 61000250.0,
      "step": 33165
    },
    {
      "entropy": 5.020107412338257,
      "epoch": 2.770631473438022,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004232826645875372,
      "loss": 4.7426,
      "mean_token_accuracy": 0.22630233615636824,
      "num_tokens": 61011219.0,
      "step": 33170
    },
    {
      "entropy": 5.039358854293823,
      "epoch": 2.7710491146007348,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042326027289983685,
      "loss": 4.6778,
      "mean_token_accuracy": 0.22408032268285752,
      "num_tokens": 61020219.0,
      "step": 33175
    },
    {
      "entropy": 5.164390563964844,
      "epoch": 2.771466755763448,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042323787861666204,
      "loss": 4.8388,
      "mean_token_accuracy": 0.2144281417131424,
      "num_tokens": 61029172.0,
      "step": 33180
    },
    {
      "entropy": 5.0431581974029545,
      "epoch": 2.7718843969261613,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042321548173840465,
      "loss": 4.6036,
      "mean_token_accuracy": 0.22717739194631575,
      "num_tokens": 61038621.0,
      "step": 33185
    },
    {
      "entropy": 4.995883655548096,
      "epoch": 2.772302038088874,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004231930822654569,
      "loss": 4.6801,
      "mean_token_accuracy": 0.23457446247339248,
      "num_tokens": 61048279.0,
      "step": 33190
    },
    {
      "entropy": 4.952418994903565,
      "epoch": 2.772719679251587,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042317068019821075,
      "loss": 4.6028,
      "mean_token_accuracy": 0.2358264908194542,
      "num_tokens": 61057459.0,
      "step": 33195
    },
    {
      "entropy": 5.022450590133667,
      "epoch": 2.7731373204143,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004231482755370586,
      "loss": 4.7224,
      "mean_token_accuracy": 0.2263111725449562,
      "num_tokens": 61066399.0,
      "step": 33200
    },
    {
      "entropy": 4.916226005554199,
      "epoch": 2.773554961577013,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042312586828239254,
      "loss": 4.6162,
      "mean_token_accuracy": 0.23588029593229293,
      "num_tokens": 61074930.0,
      "step": 33205
    },
    {
      "entropy": 4.991524934768677,
      "epoch": 2.7739726027397262,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004231034584346048,
      "loss": 4.5807,
      "mean_token_accuracy": 0.2368966519832611,
      "num_tokens": 61084693.0,
      "step": 33210
    },
    {
      "entropy": 5.060964155197143,
      "epoch": 2.774390243902439,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042308104599408783,
      "loss": 4.6593,
      "mean_token_accuracy": 0.22727951854467393,
      "num_tokens": 61093393.0,
      "step": 33215
    },
    {
      "entropy": 4.990995931625366,
      "epoch": 2.774807885065152,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004230586309612339,
      "loss": 4.664,
      "mean_token_accuracy": 0.23528939634561538,
      "num_tokens": 61102185.0,
      "step": 33220
    },
    {
      "entropy": 5.13645076751709,
      "epoch": 2.775225526227865,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004230362133364354,
      "loss": 4.7001,
      "mean_token_accuracy": 0.22883555442094802,
      "num_tokens": 61111382.0,
      "step": 33225
    },
    {
      "entropy": 5.0038759231567385,
      "epoch": 2.775643167390578,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042301379312008477,
      "loss": 4.7272,
      "mean_token_accuracy": 0.21865395158529283,
      "num_tokens": 61121230.0,
      "step": 33230
    },
    {
      "entropy": 4.981807899475098,
      "epoch": 2.776060808553291,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004229913703125745,
      "loss": 4.6711,
      "mean_token_accuracy": 0.23589764088392257,
      "num_tokens": 61131033.0,
      "step": 33235
    },
    {
      "entropy": 4.986028432846069,
      "epoch": 2.776478449716004,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042296894491429716,
      "loss": 4.6765,
      "mean_token_accuracy": 0.23348462581634521,
      "num_tokens": 61140093.0,
      "step": 33240
    },
    {
      "entropy": 5.012242364883423,
      "epoch": 2.776896090878717,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004229465169256454,
      "loss": 4.7573,
      "mean_token_accuracy": 0.22277370989322662,
      "num_tokens": 61149719.0,
      "step": 33245
    },
    {
      "entropy": 5.010865688323975,
      "epoch": 2.77731373204143,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042292408634701166,
      "loss": 4.633,
      "mean_token_accuracy": 0.23134703934192657,
      "num_tokens": 61158475.0,
      "step": 33250
    },
    {
      "entropy": 4.9808409214019775,
      "epoch": 2.777731373204143,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004229016531787889,
      "loss": 4.6261,
      "mean_token_accuracy": 0.23251782208681107,
      "num_tokens": 61167870.0,
      "step": 33255
    },
    {
      "entropy": 4.968017053604126,
      "epoch": 2.778149014366856,
      "grad_norm": 1.125,
      "learning_rate": 0.00042287921742136957,
      "loss": 4.7354,
      "mean_token_accuracy": 0.2311794564127922,
      "num_tokens": 61176757.0,
      "step": 33260
    },
    {
      "entropy": 5.027201700210571,
      "epoch": 2.778566655529569,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042285677907514656,
      "loss": 4.719,
      "mean_token_accuracy": 0.22520701736211776,
      "num_tokens": 61186524.0,
      "step": 33265
    },
    {
      "entropy": 5.094524812698364,
      "epoch": 2.7789842966922818,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042283433814051263,
      "loss": 4.7268,
      "mean_token_accuracy": 0.22213528007268907,
      "num_tokens": 61195649.0,
      "step": 33270
    },
    {
      "entropy": 5.013278675079346,
      "epoch": 2.779401937854995,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042281189461786074,
      "loss": 4.6768,
      "mean_token_accuracy": 0.22866119295358658,
      "num_tokens": 61205176.0,
      "step": 33275
    },
    {
      "entropy": 5.0670726776123045,
      "epoch": 2.779819579017708,
      "grad_norm": 1.125,
      "learning_rate": 0.0004227894485075837,
      "loss": 4.7513,
      "mean_token_accuracy": 0.22471499890089036,
      "num_tokens": 61214229.0,
      "step": 33280
    },
    {
      "entropy": 5.039881181716919,
      "epoch": 2.780237220180421,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004227669998100744,
      "loss": 4.6483,
      "mean_token_accuracy": 0.23223035782575607,
      "num_tokens": 61223587.0,
      "step": 33285
    },
    {
      "entropy": 5.106830501556397,
      "epoch": 2.780654861343134,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042274454852572583,
      "loss": 4.8232,
      "mean_token_accuracy": 0.21988152116537094,
      "num_tokens": 61232693.0,
      "step": 33290
    },
    {
      "entropy": 5.022715139389038,
      "epoch": 2.7810725025058467,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042272209465493124,
      "loss": 4.6519,
      "mean_token_accuracy": 0.22852616012096405,
      "num_tokens": 61241790.0,
      "step": 33295
    },
    {
      "entropy": 4.919955587387085,
      "epoch": 2.78149014366856,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004226996381980835,
      "loss": 4.6708,
      "mean_token_accuracy": 0.232130728662014,
      "num_tokens": 61252074.0,
      "step": 33300
    },
    {
      "entropy": 4.944873714447022,
      "epoch": 2.781907784831273,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004226771791555758,
      "loss": 4.5231,
      "mean_token_accuracy": 0.23348711729049682,
      "num_tokens": 61261065.0,
      "step": 33305
    },
    {
      "entropy": 4.939263820648193,
      "epoch": 2.782325425993986,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004226547175278014,
      "loss": 4.6078,
      "mean_token_accuracy": 0.23546002060174942,
      "num_tokens": 61270178.0,
      "step": 33310
    },
    {
      "entropy": 5.043076705932617,
      "epoch": 2.782743067156699,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004226322533151534,
      "loss": 4.6921,
      "mean_token_accuracy": 0.22847207486629487,
      "num_tokens": 61279635.0,
      "step": 33315
    },
    {
      "entropy": 5.057018089294433,
      "epoch": 2.783160708319412,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004226097865180251,
      "loss": 4.7356,
      "mean_token_accuracy": 0.2227129340171814,
      "num_tokens": 61289301.0,
      "step": 33320
    },
    {
      "entropy": 4.948888635635376,
      "epoch": 2.783578349482125,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042258731713680974,
      "loss": 4.5882,
      "mean_token_accuracy": 0.23512272536754608,
      "num_tokens": 61298027.0,
      "step": 33325
    },
    {
      "entropy": 5.011192798614502,
      "epoch": 2.783995990644838,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004225648451719008,
      "loss": 4.6957,
      "mean_token_accuracy": 0.2241205096244812,
      "num_tokens": 61306719.0,
      "step": 33330
    },
    {
      "entropy": 4.96929783821106,
      "epoch": 2.784413631807551,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004225423706236917,
      "loss": 4.6442,
      "mean_token_accuracy": 0.2355051040649414,
      "num_tokens": 61315455.0,
      "step": 33335
    },
    {
      "entropy": 5.0418352603912355,
      "epoch": 2.784831272970264,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004225198934925757,
      "loss": 4.6863,
      "mean_token_accuracy": 0.23341869413852692,
      "num_tokens": 61326082.0,
      "step": 33340
    },
    {
      "entropy": 5.003683137893677,
      "epoch": 2.785248914132977,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004224974137789464,
      "loss": 4.7048,
      "mean_token_accuracy": 0.22992815673351288,
      "num_tokens": 61335652.0,
      "step": 33345
    },
    {
      "entropy": 5.078162384033203,
      "epoch": 2.78566655529569,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004224749314831973,
      "loss": 4.7455,
      "mean_token_accuracy": 0.2282559886574745,
      "num_tokens": 61343728.0,
      "step": 33350
    },
    {
      "entropy": 5.028096914291382,
      "epoch": 2.786084196458403,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004224524466057221,
      "loss": 4.7423,
      "mean_token_accuracy": 0.23155664056539535,
      "num_tokens": 61353534.0,
      "step": 33355
    },
    {
      "entropy": 4.970406579971313,
      "epoch": 2.786501837621116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004224299591469143,
      "loss": 4.6286,
      "mean_token_accuracy": 0.23388165682554246,
      "num_tokens": 61362189.0,
      "step": 33360
    },
    {
      "entropy": 5.000257110595703,
      "epoch": 2.7869194787838287,
      "grad_norm": 1.125,
      "learning_rate": 0.0004224074691071676,
      "loss": 4.6451,
      "mean_token_accuracy": 0.23775570392608641,
      "num_tokens": 61370877.0,
      "step": 33365
    },
    {
      "entropy": 4.944062662124634,
      "epoch": 2.787337119946542,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042238497648687567,
      "loss": 4.6141,
      "mean_token_accuracy": 0.2351052388548851,
      "num_tokens": 61379593.0,
      "step": 33370
    },
    {
      "entropy": 4.951638317108154,
      "epoch": 2.787754761109255,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042236248128643244,
      "loss": 4.6507,
      "mean_token_accuracy": 0.22774874567985534,
      "num_tokens": 61387552.0,
      "step": 33375
    },
    {
      "entropy": 5.037874603271485,
      "epoch": 2.788172402271968,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004223399835062315,
      "loss": 4.677,
      "mean_token_accuracy": 0.2322031781077385,
      "num_tokens": 61396329.0,
      "step": 33380
    },
    {
      "entropy": 5.010325622558594,
      "epoch": 2.788590043434681,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042231748314666684,
      "loss": 4.6198,
      "mean_token_accuracy": 0.22839533984661103,
      "num_tokens": 61405134.0,
      "step": 33385
    },
    {
      "entropy": 5.04682035446167,
      "epoch": 2.7890076845973937,
      "grad_norm": 1.125,
      "learning_rate": 0.0004222949802081323,
      "loss": 4.7626,
      "mean_token_accuracy": 0.21292010992765426,
      "num_tokens": 61415201.0,
      "step": 33390
    },
    {
      "entropy": 5.005916929244995,
      "epoch": 2.789425325760107,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042227247469102175,
      "loss": 4.674,
      "mean_token_accuracy": 0.2333976373076439,
      "num_tokens": 61423429.0,
      "step": 33395
    },
    {
      "entropy": 5.04067177772522,
      "epoch": 2.78984296692282,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004222499665957294,
      "loss": 4.6299,
      "mean_token_accuracy": 0.22395915538072586,
      "num_tokens": 61432712.0,
      "step": 33400
    },
    {
      "entropy": 5.076129579544068,
      "epoch": 2.790260608085533,
      "grad_norm": 1.0,
      "learning_rate": 0.00042222745592264903,
      "loss": 4.7298,
      "mean_token_accuracy": 0.22781918346881866,
      "num_tokens": 61442410.0,
      "step": 33405
    },
    {
      "entropy": 4.92684907913208,
      "epoch": 2.790678249248246,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042220494267217483,
      "loss": 4.6116,
      "mean_token_accuracy": 0.23674662113189698,
      "num_tokens": 61452027.0,
      "step": 33410
    },
    {
      "entropy": 4.973494529724121,
      "epoch": 2.791095890410959,
      "grad_norm": 1.0625,
      "learning_rate": 0.000422182426844701,
      "loss": 4.5913,
      "mean_token_accuracy": 0.2385205090045929,
      "num_tokens": 61460288.0,
      "step": 33415
    },
    {
      "entropy": 4.957191038131714,
      "epoch": 2.791513531573672,
      "grad_norm": 1.125,
      "learning_rate": 0.0004221599084406216,
      "loss": 4.6565,
      "mean_token_accuracy": 0.24169424772262574,
      "num_tokens": 61469624.0,
      "step": 33420
    },
    {
      "entropy": 4.928747940063476,
      "epoch": 2.791931172736385,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004221373874603308,
      "loss": 4.5744,
      "mean_token_accuracy": 0.2409015953540802,
      "num_tokens": 61479278.0,
      "step": 33425
    },
    {
      "entropy": 5.025892066955566,
      "epoch": 2.792348813899098,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042211486390422297,
      "loss": 4.6634,
      "mean_token_accuracy": 0.22837887853384017,
      "num_tokens": 61488796.0,
      "step": 33430
    },
    {
      "entropy": 5.063781070709228,
      "epoch": 2.7927664550618108,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042209233777269235,
      "loss": 4.7492,
      "mean_token_accuracy": 0.22267863005399705,
      "num_tokens": 61498559.0,
      "step": 33435
    },
    {
      "entropy": 5.062264108657837,
      "epoch": 2.793184096224524,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042206980906613333,
      "loss": 4.7053,
      "mean_token_accuracy": 0.22696684449911117,
      "num_tokens": 61507506.0,
      "step": 33440
    },
    {
      "entropy": 4.938253784179688,
      "epoch": 2.793601737387237,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004220472777849403,
      "loss": 4.5974,
      "mean_token_accuracy": 0.23799720108509065,
      "num_tokens": 61516437.0,
      "step": 33445
    },
    {
      "entropy": 5.029847812652588,
      "epoch": 2.79401937854995,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042202474392950763,
      "loss": 4.6673,
      "mean_token_accuracy": 0.2314683601260185,
      "num_tokens": 61524480.0,
      "step": 33450
    },
    {
      "entropy": 4.9742269039154055,
      "epoch": 2.794437019712663,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004220022075002298,
      "loss": 4.6751,
      "mean_token_accuracy": 0.23304740637540816,
      "num_tokens": 61535130.0,
      "step": 33455
    },
    {
      "entropy": 5.068253660202027,
      "epoch": 2.7948546608753757,
      "grad_norm": 1.125,
      "learning_rate": 0.0004219796684975014,
      "loss": 4.7266,
      "mean_token_accuracy": 0.22079967111349105,
      "num_tokens": 61544448.0,
      "step": 33460
    },
    {
      "entropy": 4.940319299697876,
      "epoch": 2.795272302038089,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042195712692171705,
      "loss": 4.6042,
      "mean_token_accuracy": 0.2406189978122711,
      "num_tokens": 61552810.0,
      "step": 33465
    },
    {
      "entropy": 5.082782459259033,
      "epoch": 2.7956899432008018,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004219345827732712,
      "loss": 4.7589,
      "mean_token_accuracy": 0.22861257940530777,
      "num_tokens": 61561476.0,
      "step": 33470
    },
    {
      "entropy": 5.000978326797485,
      "epoch": 2.796107584363515,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004219120360525588,
      "loss": 4.621,
      "mean_token_accuracy": 0.2317455843091011,
      "num_tokens": 61570896.0,
      "step": 33475
    },
    {
      "entropy": 4.956778144836425,
      "epoch": 2.796525225526228,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004218894867599742,
      "loss": 4.6742,
      "mean_token_accuracy": 0.2374252811074257,
      "num_tokens": 61579625.0,
      "step": 33480
    },
    {
      "entropy": 5.026124572753906,
      "epoch": 2.7969428666889407,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042186693489591244,
      "loss": 4.6582,
      "mean_token_accuracy": 0.22601830810308457,
      "num_tokens": 61588882.0,
      "step": 33485
    },
    {
      "entropy": 4.982210540771485,
      "epoch": 2.797360507851654,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004218443804607681,
      "loss": 4.6025,
      "mean_token_accuracy": 0.23413151502609253,
      "num_tokens": 61598053.0,
      "step": 33490
    },
    {
      "entropy": 4.982962656021118,
      "epoch": 2.7977781490143667,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042182182345493614,
      "loss": 4.589,
      "mean_token_accuracy": 0.23558966368436812,
      "num_tokens": 61607418.0,
      "step": 33495
    },
    {
      "entropy": 4.939707851409912,
      "epoch": 2.79819579017708,
      "grad_norm": 1.125,
      "learning_rate": 0.0004217992638788114,
      "loss": 4.6219,
      "mean_token_accuracy": 0.2261250749230385,
      "num_tokens": 61616881.0,
      "step": 33500
    },
    {
      "entropy": 5.002894306182862,
      "epoch": 2.798613431339793,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004217767017327889,
      "loss": 4.732,
      "mean_token_accuracy": 0.22632152140140532,
      "num_tokens": 61625225.0,
      "step": 33505
    },
    {
      "entropy": 5.000752305984497,
      "epoch": 2.7990310725025056,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004217541370172635,
      "loss": 4.6696,
      "mean_token_accuracy": 0.23220040053129196,
      "num_tokens": 61633551.0,
      "step": 33510
    },
    {
      "entropy": 5.1217796325683596,
      "epoch": 2.799448713665219,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042173156973263044,
      "loss": 4.7635,
      "mean_token_accuracy": 0.21858031898736954,
      "num_tokens": 61643531.0,
      "step": 33515
    },
    {
      "entropy": 5.042950630187988,
      "epoch": 2.799866354827932,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004217089998792845,
      "loss": 4.7928,
      "mean_token_accuracy": 0.22511079609394075,
      "num_tokens": 61653112.0,
      "step": 33520
    },
    {
      "entropy": 5.000666761398316,
      "epoch": 2.800283995990645,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004216864274576209,
      "loss": 4.6106,
      "mean_token_accuracy": 0.2265822172164917,
      "num_tokens": 61662089.0,
      "step": 33525
    },
    {
      "entropy": 4.997918176651001,
      "epoch": 2.8007016371533577,
      "grad_norm": 1.125,
      "learning_rate": 0.0004216638524680349,
      "loss": 4.6266,
      "mean_token_accuracy": 0.2352616861462593,
      "num_tokens": 61671255.0,
      "step": 33530
    },
    {
      "entropy": 5.008117532730102,
      "epoch": 2.801119278316071,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042164127491092155,
      "loss": 4.7028,
      "mean_token_accuracy": 0.21910753697156907,
      "num_tokens": 61680619.0,
      "step": 33535
    },
    {
      "entropy": 5.055190229415894,
      "epoch": 2.801536919478784,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004216186947866762,
      "loss": 4.7559,
      "mean_token_accuracy": 0.24034977257251738,
      "num_tokens": 61691093.0,
      "step": 33540
    },
    {
      "entropy": 5.102552843093872,
      "epoch": 2.801954560641497,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042159611209569416,
      "loss": 4.7465,
      "mean_token_accuracy": 0.22331393808126448,
      "num_tokens": 61700783.0,
      "step": 33545
    },
    {
      "entropy": 5.033966207504273,
      "epoch": 2.80237220180421,
      "grad_norm": 1.125,
      "learning_rate": 0.00042157352683837063,
      "loss": 4.6874,
      "mean_token_accuracy": 0.23086962401866912,
      "num_tokens": 61710156.0,
      "step": 33550
    },
    {
      "entropy": 5.00388240814209,
      "epoch": 2.8027898429669227,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004215509390151012,
      "loss": 4.6601,
      "mean_token_accuracy": 0.2292836755514145,
      "num_tokens": 61719130.0,
      "step": 33555
    },
    {
      "entropy": 4.976269245147705,
      "epoch": 2.803207484129636,
      "grad_norm": 1.078125,
      "learning_rate": 0.000421528348626281,
      "loss": 4.5661,
      "mean_token_accuracy": 0.24413691610097885,
      "num_tokens": 61727396.0,
      "step": 33560
    },
    {
      "entropy": 4.9913167476654055,
      "epoch": 2.8036251252923488,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042150575567230584,
      "loss": 4.6554,
      "mean_token_accuracy": 0.23793882131576538,
      "num_tokens": 61736152.0,
      "step": 33565
    },
    {
      "entropy": 5.001427698135376,
      "epoch": 2.804042766455062,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042148316015357107,
      "loss": 4.6458,
      "mean_token_accuracy": 0.22277824878692626,
      "num_tokens": 61745663.0,
      "step": 33570
    },
    {
      "entropy": 5.033577537536621,
      "epoch": 2.804460407617775,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004214605620704722,
      "loss": 4.7519,
      "mean_token_accuracy": 0.22196636497974395,
      "num_tokens": 61754608.0,
      "step": 33575
    },
    {
      "entropy": 4.9960136890411375,
      "epoch": 2.8048780487804876,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042143796142340496,
      "loss": 4.5887,
      "mean_token_accuracy": 0.23449108004570007,
      "num_tokens": 61764355.0,
      "step": 33580
    },
    {
      "entropy": 5.004236078262329,
      "epoch": 2.805295689943201,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042141535821276497,
      "loss": 4.633,
      "mean_token_accuracy": 0.23511847257614135,
      "num_tokens": 61773108.0,
      "step": 33585
    },
    {
      "entropy": 5.155463075637817,
      "epoch": 2.8057133311059137,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042139275243894777,
      "loss": 4.8339,
      "mean_token_accuracy": 0.2164183184504509,
      "num_tokens": 61782757.0,
      "step": 33590
    },
    {
      "entropy": 4.920201826095581,
      "epoch": 2.806130972268627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042137014410234926,
      "loss": 4.5597,
      "mean_token_accuracy": 0.24397081434726714,
      "num_tokens": 61791684.0,
      "step": 33595
    },
    {
      "entropy": 5.019274282455444,
      "epoch": 2.8065486134313398,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042134753320336517,
      "loss": 4.7129,
      "mean_token_accuracy": 0.22117989659309387,
      "num_tokens": 61801303.0,
      "step": 33600
    },
    {
      "entropy": 5.042372703552246,
      "epoch": 2.8069662545940526,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004213249197423914,
      "loss": 4.6197,
      "mean_token_accuracy": 0.23251387625932693,
      "num_tokens": 61810640.0,
      "step": 33605
    },
    {
      "entropy": 4.959527921676636,
      "epoch": 2.807383895756766,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004213023037198238,
      "loss": 4.5838,
      "mean_token_accuracy": 0.23715462982654573,
      "num_tokens": 61818976.0,
      "step": 33610
    },
    {
      "entropy": 4.931808757781982,
      "epoch": 2.8078015369194786,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042127968513605823,
      "loss": 4.4999,
      "mean_token_accuracy": 0.23966427147388458,
      "num_tokens": 61827253.0,
      "step": 33615
    },
    {
      "entropy": 4.900357341766357,
      "epoch": 2.808219178082192,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042125706399149073,
      "loss": 4.6349,
      "mean_token_accuracy": 0.2340223267674446,
      "num_tokens": 61836083.0,
      "step": 33620
    },
    {
      "entropy": 5.062822437286377,
      "epoch": 2.8086368192449047,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042123444028651725,
      "loss": 4.7193,
      "mean_token_accuracy": 0.23058853596448897,
      "num_tokens": 61845835.0,
      "step": 33625
    },
    {
      "entropy": 5.058974409103394,
      "epoch": 2.8090544604076175,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042121181402153394,
      "loss": 4.6817,
      "mean_token_accuracy": 0.2283428966999054,
      "num_tokens": 61856075.0,
      "step": 33630
    },
    {
      "entropy": 4.982777214050293,
      "epoch": 2.809472101570331,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042118918519693665,
      "loss": 4.6181,
      "mean_token_accuracy": 0.236976358294487,
      "num_tokens": 61864607.0,
      "step": 33635
    },
    {
      "entropy": 4.980930423736572,
      "epoch": 2.809889742733044,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042116655381312193,
      "loss": 4.6578,
      "mean_token_accuracy": 0.22689546197652816,
      "num_tokens": 61873602.0,
      "step": 33640
    },
    {
      "entropy": 5.023687744140625,
      "epoch": 2.810307383895757,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042114391987048557,
      "loss": 4.6462,
      "mean_token_accuracy": 0.2274514004588127,
      "num_tokens": 61882936.0,
      "step": 33645
    },
    {
      "entropy": 4.965233325958252,
      "epoch": 2.8107250250584697,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042112128336942406,
      "loss": 4.5767,
      "mean_token_accuracy": 0.2424253210425377,
      "num_tokens": 61892131.0,
      "step": 33650
    },
    {
      "entropy": 4.976123428344726,
      "epoch": 2.811142666221183,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004210986443103335,
      "loss": 4.6715,
      "mean_token_accuracy": 0.2300476998090744,
      "num_tokens": 61901953.0,
      "step": 33655
    },
    {
      "entropy": 4.913595533370971,
      "epoch": 2.8115603073838957,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042107600269361046,
      "loss": 4.5908,
      "mean_token_accuracy": 0.23441161215305328,
      "num_tokens": 61911779.0,
      "step": 33660
    },
    {
      "entropy": 5.019628286361694,
      "epoch": 2.811977948546609,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042105335851965104,
      "loss": 4.6669,
      "mean_token_accuracy": 0.235475555062294,
      "num_tokens": 61920860.0,
      "step": 33665
    },
    {
      "entropy": 5.056917953491211,
      "epoch": 2.812395589709322,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004210307117888519,
      "loss": 4.6255,
      "mean_token_accuracy": 0.22988687455654144,
      "num_tokens": 61930480.0,
      "step": 33670
    },
    {
      "entropy": 5.052816247940063,
      "epoch": 2.8128132308720346,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042100806250160925,
      "loss": 4.7178,
      "mean_token_accuracy": 0.22406836450099946,
      "num_tokens": 61940368.0,
      "step": 33675
    },
    {
      "entropy": 5.045366811752319,
      "epoch": 2.813230872034748,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004209854106583198,
      "loss": 4.6772,
      "mean_token_accuracy": 0.22537895888090134,
      "num_tokens": 61950310.0,
      "step": 33680
    },
    {
      "entropy": 4.957862710952758,
      "epoch": 2.8136485131974607,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042096275625938,
      "loss": 4.5536,
      "mean_token_accuracy": 0.24285958558321,
      "num_tokens": 61959846.0,
      "step": 33685
    },
    {
      "entropy": 4.920775985717773,
      "epoch": 2.814066154360174,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042094009930518643,
      "loss": 4.5636,
      "mean_token_accuracy": 0.23462893664836884,
      "num_tokens": 61969559.0,
      "step": 33690
    },
    {
      "entropy": 4.959434986114502,
      "epoch": 2.8144837955228867,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042091743979613574,
      "loss": 4.572,
      "mean_token_accuracy": 0.23189371079206467,
      "num_tokens": 61978684.0,
      "step": 33695
    },
    {
      "entropy": 4.954928970336914,
      "epoch": 2.8149014366855996,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004208947777326246,
      "loss": 4.6071,
      "mean_token_accuracy": 0.23424306511878967,
      "num_tokens": 61987070.0,
      "step": 33700
    },
    {
      "entropy": 4.947752571105957,
      "epoch": 2.815319077848313,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042087211311504986,
      "loss": 4.587,
      "mean_token_accuracy": 0.22694578915834426,
      "num_tokens": 61995813.0,
      "step": 33705
    },
    {
      "entropy": 4.925452661514282,
      "epoch": 2.8157367190110256,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042084944594380813,
      "loss": 4.6393,
      "mean_token_accuracy": 0.2330278977751732,
      "num_tokens": 62004638.0,
      "step": 33710
    },
    {
      "entropy": 5.100624084472656,
      "epoch": 2.816154360173739,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042082677621929626,
      "loss": 4.7514,
      "mean_token_accuracy": 0.23159761875867843,
      "num_tokens": 62013699.0,
      "step": 33715
    },
    {
      "entropy": 5.1406022071838375,
      "epoch": 2.8165720013364517,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004208041039419113,
      "loss": 4.7353,
      "mean_token_accuracy": 0.23349197059869767,
      "num_tokens": 62023081.0,
      "step": 33720
    },
    {
      "entropy": 4.994840145111084,
      "epoch": 2.8169896424991645,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042078142911204983,
      "loss": 4.6203,
      "mean_token_accuracy": 0.24223596453666688,
      "num_tokens": 62032077.0,
      "step": 33725
    },
    {
      "entropy": 4.987251901626587,
      "epoch": 2.8174072836618778,
      "grad_norm": 1.09375,
      "learning_rate": 0.000420758751730109,
      "loss": 4.6362,
      "mean_token_accuracy": 0.23655566424131394,
      "num_tokens": 62041714.0,
      "step": 33730
    },
    {
      "entropy": 4.93692569732666,
      "epoch": 2.8178249248245906,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042073607179648575,
      "loss": 4.5892,
      "mean_token_accuracy": 0.23998921364545822,
      "num_tokens": 62049946.0,
      "step": 33735
    },
    {
      "entropy": 4.959801292419433,
      "epoch": 2.818242565987304,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004207133893115772,
      "loss": 4.6444,
      "mean_token_accuracy": 0.23112045228481293,
      "num_tokens": 62058758.0,
      "step": 33740
    },
    {
      "entropy": 5.045012140274048,
      "epoch": 2.8186602071500166,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004206907042757803,
      "loss": 4.7081,
      "mean_token_accuracy": 0.22714383751153946,
      "num_tokens": 62068424.0,
      "step": 33745
    },
    {
      "entropy": 4.999184322357178,
      "epoch": 2.8190778483127295,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042066801668949236,
      "loss": 4.6489,
      "mean_token_accuracy": 0.22967298179864884,
      "num_tokens": 62078658.0,
      "step": 33750
    },
    {
      "entropy": 5.065019989013672,
      "epoch": 2.8194954894754427,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004206453265531104,
      "loss": 4.6625,
      "mean_token_accuracy": 0.2319467097520828,
      "num_tokens": 62087956.0,
      "step": 33755
    },
    {
      "entropy": 4.956689119338989,
      "epoch": 2.819913130638156,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042062263386703167,
      "loss": 4.6335,
      "mean_token_accuracy": 0.23316727578639984,
      "num_tokens": 62097380.0,
      "step": 33760
    },
    {
      "entropy": 5.003720140457153,
      "epoch": 2.8203307718008688,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042059993863165343,
      "loss": 4.7035,
      "mean_token_accuracy": 0.2301735609769821,
      "num_tokens": 62106148.0,
      "step": 33765
    },
    {
      "entropy": 5.06308274269104,
      "epoch": 2.8207484129635816,
      "grad_norm": 1.15625,
      "learning_rate": 0.000420577240847373,
      "loss": 4.7373,
      "mean_token_accuracy": 0.2300790876150131,
      "num_tokens": 62115617.0,
      "step": 33770
    },
    {
      "entropy": 5.064489555358887,
      "epoch": 2.821166054126295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004205545405145878,
      "loss": 4.7014,
      "mean_token_accuracy": 0.22867980897426604,
      "num_tokens": 62124923.0,
      "step": 33775
    },
    {
      "entropy": 4.992897415161133,
      "epoch": 2.8215836952890077,
      "grad_norm": 1.125,
      "learning_rate": 0.0004205318376336952,
      "loss": 4.6811,
      "mean_token_accuracy": 0.22869728952646257,
      "num_tokens": 62133703.0,
      "step": 33780
    },
    {
      "entropy": 4.985577392578125,
      "epoch": 2.822001336451721,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042050913220509253,
      "loss": 4.6453,
      "mean_token_accuracy": 0.22533738911151885,
      "num_tokens": 62143036.0,
      "step": 33785
    },
    {
      "entropy": 5.094156932830811,
      "epoch": 2.8224189776144337,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004204864242291773,
      "loss": 4.7291,
      "mean_token_accuracy": 0.22509319484233856,
      "num_tokens": 62152318.0,
      "step": 33790
    },
    {
      "entropy": 5.121952533721924,
      "epoch": 2.8228366187771465,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042046371370634716,
      "loss": 4.7702,
      "mean_token_accuracy": 0.2225964769721031,
      "num_tokens": 62162190.0,
      "step": 33795
    },
    {
      "entropy": 4.947400379180908,
      "epoch": 2.82325425993986,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042044100063699964,
      "loss": 4.608,
      "mean_token_accuracy": 0.24017363637685776,
      "num_tokens": 62171094.0,
      "step": 33800
    },
    {
      "entropy": 4.984966039657593,
      "epoch": 2.8236719011025726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004204182850215323,
      "loss": 4.6397,
      "mean_token_accuracy": 0.23560561388731002,
      "num_tokens": 62179624.0,
      "step": 33805
    },
    {
      "entropy": 5.069425392150879,
      "epoch": 2.824089542265286,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042039556686034285,
      "loss": 4.7653,
      "mean_token_accuracy": 0.22194967716932296,
      "num_tokens": 62189497.0,
      "step": 33810
    },
    {
      "entropy": 5.020899486541748,
      "epoch": 2.8245071834279987,
      "grad_norm": 1.125,
      "learning_rate": 0.000420372846153829,
      "loss": 4.6604,
      "mean_token_accuracy": 0.23849330246448516,
      "num_tokens": 62198671.0,
      "step": 33815
    },
    {
      "entropy": 5.0212485790252686,
      "epoch": 2.8249248245907115,
      "grad_norm": 1.125,
      "learning_rate": 0.00042035012290238846,
      "loss": 4.6838,
      "mean_token_accuracy": 0.23689756840467452,
      "num_tokens": 62207617.0,
      "step": 33820
    },
    {
      "entropy": 4.93328275680542,
      "epoch": 2.8253424657534247,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004203273971064191,
      "loss": 4.5658,
      "mean_token_accuracy": 0.23504294008016585,
      "num_tokens": 62215921.0,
      "step": 33825
    },
    {
      "entropy": 5.00967926979065,
      "epoch": 2.8257601069161375,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004203046687663187,
      "loss": 4.6915,
      "mean_token_accuracy": 0.2301537424325943,
      "num_tokens": 62225859.0,
      "step": 33830
    },
    {
      "entropy": 4.993383741378784,
      "epoch": 2.826177748078851,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004202819378824851,
      "loss": 4.6492,
      "mean_token_accuracy": 0.2352867141366005,
      "num_tokens": 62234295.0,
      "step": 33835
    },
    {
      "entropy": 4.986025714874268,
      "epoch": 2.8265953892415636,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004202592044553165,
      "loss": 4.6567,
      "mean_token_accuracy": 0.22906930297613143,
      "num_tokens": 62243886.0,
      "step": 33840
    },
    {
      "entropy": 4.990660715103149,
      "epoch": 2.8270130304042764,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042023646848521047,
      "loss": 4.6122,
      "mean_token_accuracy": 0.23647975027561188,
      "num_tokens": 62252188.0,
      "step": 33845
    },
    {
      "entropy": 4.959005546569824,
      "epoch": 2.8274306715669897,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042021372997256537,
      "loss": 4.6571,
      "mean_token_accuracy": 0.2290931150317192,
      "num_tokens": 62261481.0,
      "step": 33850
    },
    {
      "entropy": 4.959955024719238,
      "epoch": 2.8278483127297025,
      "grad_norm": 1.015625,
      "learning_rate": 0.000420190988917779,
      "loss": 4.6171,
      "mean_token_accuracy": 0.23301629573106766,
      "num_tokens": 62271059.0,
      "step": 33855
    },
    {
      "entropy": 4.94107608795166,
      "epoch": 2.8282659538924158,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004201682453212497,
      "loss": 4.6182,
      "mean_token_accuracy": 0.23103617429733275,
      "num_tokens": 62279956.0,
      "step": 33860
    },
    {
      "entropy": 5.0439262866973875,
      "epoch": 2.8286835950551286,
      "grad_norm": 1.125,
      "learning_rate": 0.00042014549918337554,
      "loss": 4.6991,
      "mean_token_accuracy": 0.2266998693346977,
      "num_tokens": 62289114.0,
      "step": 33865
    },
    {
      "entropy": 5.013571691513062,
      "epoch": 2.8291012362178414,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042012275050455465,
      "loss": 4.6499,
      "mean_token_accuracy": 0.2286590278148651,
      "num_tokens": 62299458.0,
      "step": 33870
    },
    {
      "entropy": 5.033198547363281,
      "epoch": 2.8295188773805546,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004200999992851853,
      "loss": 4.6667,
      "mean_token_accuracy": 0.22941222935914993,
      "num_tokens": 62308658.0,
      "step": 33875
    },
    {
      "entropy": 5.030980777740479,
      "epoch": 2.829936518543268,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042007724552566597,
      "loss": 4.7406,
      "mean_token_accuracy": 0.22692743092775344,
      "num_tokens": 62318282.0,
      "step": 33880
    },
    {
      "entropy": 5.101660633087159,
      "epoch": 2.8303541597059807,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042005448922639464,
      "loss": 4.7256,
      "mean_token_accuracy": 0.22687510251998902,
      "num_tokens": 62328847.0,
      "step": 33885
    },
    {
      "entropy": 5.003559541702271,
      "epoch": 2.8307718008686935,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042003173038777,
      "loss": 4.6004,
      "mean_token_accuracy": 0.233925162255764,
      "num_tokens": 62338508.0,
      "step": 33890
    },
    {
      "entropy": 4.9201404571533205,
      "epoch": 2.8311894420314068,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004200089690101903,
      "loss": 4.5797,
      "mean_token_accuracy": 0.2382000595331192,
      "num_tokens": 62347462.0,
      "step": 33895
    },
    {
      "entropy": 5.0176185131072994,
      "epoch": 2.8316070831941196,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041998620509405406,
      "loss": 4.7432,
      "mean_token_accuracy": 0.22634306848049163,
      "num_tokens": 62356716.0,
      "step": 33900
    },
    {
      "entropy": 5.039562654495239,
      "epoch": 2.832024724356833,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004199634386397598,
      "loss": 4.7177,
      "mean_token_accuracy": 0.2281585454940796,
      "num_tokens": 62365712.0,
      "step": 33905
    },
    {
      "entropy": 4.978007030487061,
      "epoch": 2.8324423655195456,
      "grad_norm": 1.0625,
      "learning_rate": 0.000419940669647706,
      "loss": 4.682,
      "mean_token_accuracy": 0.22926387041807175,
      "num_tokens": 62375522.0,
      "step": 33910
    },
    {
      "entropy": 5.0038892269134525,
      "epoch": 2.8328600066822585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004199178981182913,
      "loss": 4.627,
      "mean_token_accuracy": 0.2326670080423355,
      "num_tokens": 62384440.0,
      "step": 33915
    },
    {
      "entropy": 4.983642959594727,
      "epoch": 2.8332776478449717,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004198951240519144,
      "loss": 4.6125,
      "mean_token_accuracy": 0.23948695212602616,
      "num_tokens": 62394013.0,
      "step": 33920
    },
    {
      "entropy": 4.9536127090454105,
      "epoch": 2.8336952890076845,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041987234744897395,
      "loss": 4.6415,
      "mean_token_accuracy": 0.22966720312833785,
      "num_tokens": 62403234.0,
      "step": 33925
    },
    {
      "entropy": 4.991685676574707,
      "epoch": 2.834112930170398,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041984956830986865,
      "loss": 4.6247,
      "mean_token_accuracy": 0.23374879509210586,
      "num_tokens": 62412496.0,
      "step": 33930
    },
    {
      "entropy": 4.993606042861939,
      "epoch": 2.8345305713331106,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004198267866349973,
      "loss": 4.6256,
      "mean_token_accuracy": 0.23374840915203093,
      "num_tokens": 62421360.0,
      "step": 33935
    },
    {
      "entropy": 4.937182283401489,
      "epoch": 2.8349482124958234,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004198040024247587,
      "loss": 4.6124,
      "mean_token_accuracy": 0.2326929271221161,
      "num_tokens": 62429694.0,
      "step": 33940
    },
    {
      "entropy": 4.960342073440552,
      "epoch": 2.8353658536585367,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004197812156795518,
      "loss": 4.7413,
      "mean_token_accuracy": 0.22911633551120758,
      "num_tokens": 62438411.0,
      "step": 33945
    },
    {
      "entropy": 5.019028282165527,
      "epoch": 2.8357834948212495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004197584263997754,
      "loss": 4.7036,
      "mean_token_accuracy": 0.22975700050592424,
      "num_tokens": 62448707.0,
      "step": 33950
    },
    {
      "entropy": 5.00363245010376,
      "epoch": 2.8362011359839627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004197356345858285,
      "loss": 4.6125,
      "mean_token_accuracy": 0.2397565945982933,
      "num_tokens": 62456964.0,
      "step": 33955
    },
    {
      "entropy": 4.991561317443848,
      "epoch": 2.8366187771466755,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041971284023811014,
      "loss": 4.6651,
      "mean_token_accuracy": 0.2353361204266548,
      "num_tokens": 62466892.0,
      "step": 33960
    },
    {
      "entropy": 4.982417392730713,
      "epoch": 2.8370364183093884,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00041969004335701927,
      "loss": 4.5905,
      "mean_token_accuracy": 0.23683094829320908,
      "num_tokens": 62475741.0,
      "step": 33965
    },
    {
      "entropy": 4.988082218170166,
      "epoch": 2.8374540594721016,
      "grad_norm": 1.0625,
      "learning_rate": 0.000419667243942955,
      "loss": 4.609,
      "mean_token_accuracy": 0.23507929146289824,
      "num_tokens": 62485083.0,
      "step": 33970
    },
    {
      "entropy": 5.014020395278931,
      "epoch": 2.8378717006348144,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004196444419963166,
      "loss": 4.7312,
      "mean_token_accuracy": 0.22137293666601182,
      "num_tokens": 62494432.0,
      "step": 33975
    },
    {
      "entropy": 4.870135402679443,
      "epoch": 2.8382893417975277,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004196216375175031,
      "loss": 4.4984,
      "mean_token_accuracy": 0.23982589989900588,
      "num_tokens": 62503758.0,
      "step": 33980
    },
    {
      "entropy": 5.005137395858765,
      "epoch": 2.8387069829602405,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004195988305069137,
      "loss": 4.7729,
      "mean_token_accuracy": 0.22393050491809846,
      "num_tokens": 62513799.0,
      "step": 33985
    },
    {
      "entropy": 5.110180616378784,
      "epoch": 2.8391246241229533,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004195760209649478,
      "loss": 4.7957,
      "mean_token_accuracy": 0.22091321498155594,
      "num_tokens": 62523244.0,
      "step": 33990
    },
    {
      "entropy": 5.052149534225464,
      "epoch": 2.8395422652856666,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004195532088920045,
      "loss": 4.6008,
      "mean_token_accuracy": 0.23119663298130036,
      "num_tokens": 62532387.0,
      "step": 33995
    },
    {
      "entropy": 5.059255743026734,
      "epoch": 2.83995990644838,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041953039428848336,
      "loss": 4.8206,
      "mean_token_accuracy": 0.21941456496715545,
      "num_tokens": 62542431.0,
      "step": 34000
    },
    {
      "entropy": 4.88815712928772,
      "epoch": 2.8403775476110926,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004195075771547837,
      "loss": 4.5336,
      "mean_token_accuracy": 0.24208628982305527,
      "num_tokens": 62551910.0,
      "step": 34005
    },
    {
      "entropy": 4.960805130004883,
      "epoch": 2.8407951887738054,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000419484757491305,
      "loss": 4.6324,
      "mean_token_accuracy": 0.23201049715280533,
      "num_tokens": 62561307.0,
      "step": 34010
    },
    {
      "entropy": 5.027277946472168,
      "epoch": 2.8412128299365187,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004194619352984467,
      "loss": 4.7049,
      "mean_token_accuracy": 0.22839016914367677,
      "num_tokens": 62571313.0,
      "step": 34015
    },
    {
      "entropy": 4.944858646392822,
      "epoch": 2.8416304710992315,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041943911057660817,
      "loss": 4.5521,
      "mean_token_accuracy": 0.24104329198598862,
      "num_tokens": 62580444.0,
      "step": 34020
    },
    {
      "entropy": 5.019084405899048,
      "epoch": 2.8420481122619448,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004194162833261893,
      "loss": 4.6306,
      "mean_token_accuracy": 0.2385693684220314,
      "num_tokens": 62590188.0,
      "step": 34025
    },
    {
      "entropy": 5.029233026504516,
      "epoch": 2.8424657534246576,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004193934535475895,
      "loss": 4.8186,
      "mean_token_accuracy": 0.21652020514011383,
      "num_tokens": 62599989.0,
      "step": 34030
    },
    {
      "entropy": 4.966252565383911,
      "epoch": 2.8428833945873704,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041937062124120856,
      "loss": 4.6537,
      "mean_token_accuracy": 0.23158881962299346,
      "num_tokens": 62609239.0,
      "step": 34035
    },
    {
      "entropy": 4.971028423309326,
      "epoch": 2.8433010357500836,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041934778640744603,
      "loss": 4.6584,
      "mean_token_accuracy": 0.23262715786695481,
      "num_tokens": 62618347.0,
      "step": 34040
    },
    {
      "entropy": 5.005687427520752,
      "epoch": 2.8437186769127965,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041932494904670174,
      "loss": 4.6143,
      "mean_token_accuracy": 0.23467119485139848,
      "num_tokens": 62626865.0,
      "step": 34045
    },
    {
      "entropy": 4.997774648666382,
      "epoch": 2.8441363180755097,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004193021091593756,
      "loss": 4.6984,
      "mean_token_accuracy": 0.2289646238088608,
      "num_tokens": 62636498.0,
      "step": 34050
    },
    {
      "entropy": 5.07376503944397,
      "epoch": 2.8445539592382225,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041927926674586725,
      "loss": 4.7853,
      "mean_token_accuracy": 0.22269437611103057,
      "num_tokens": 62646192.0,
      "step": 34055
    },
    {
      "entropy": 4.982354831695557,
      "epoch": 2.8449716004009353,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004192564218065768,
      "loss": 4.5551,
      "mean_token_accuracy": 0.23722055107355117,
      "num_tokens": 62654986.0,
      "step": 34060
    },
    {
      "entropy": 4.976088905334473,
      "epoch": 2.8453892415636486,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004192335743419039,
      "loss": 4.6325,
      "mean_token_accuracy": 0.23907106071710588,
      "num_tokens": 62664515.0,
      "step": 34065
    },
    {
      "entropy": 5.054563283920288,
      "epoch": 2.8458068827263614,
      "grad_norm": 1.125,
      "learning_rate": 0.0004192107243522487,
      "loss": 4.6504,
      "mean_token_accuracy": 0.2289798453450203,
      "num_tokens": 62673573.0,
      "step": 34070
    },
    {
      "entropy": 5.015541028976441,
      "epoch": 2.8462245238890747,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041918787183801114,
      "loss": 4.6485,
      "mean_token_accuracy": 0.23378919810056686,
      "num_tokens": 62682002.0,
      "step": 34075
    },
    {
      "entropy": 4.94172773361206,
      "epoch": 2.8466421650517875,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041916501679959144,
      "loss": 4.5799,
      "mean_token_accuracy": 0.2361077845096588,
      "num_tokens": 62690214.0,
      "step": 34080
    },
    {
      "entropy": 4.955369901657105,
      "epoch": 2.8470598062145003,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041914215923738945,
      "loss": 4.5977,
      "mean_token_accuracy": 0.23442363888025283,
      "num_tokens": 62698090.0,
      "step": 34085
    },
    {
      "entropy": 4.912707805633545,
      "epoch": 2.8474774473772135,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041911929915180556,
      "loss": 4.5695,
      "mean_token_accuracy": 0.24555558562278748,
      "num_tokens": 62706575.0,
      "step": 34090
    },
    {
      "entropy": 4.949275732040405,
      "epoch": 2.8478950885399263,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004190964365432398,
      "loss": 4.5934,
      "mean_token_accuracy": 0.23856433033943175,
      "num_tokens": 62715400.0,
      "step": 34095
    },
    {
      "entropy": 5.0280028820037845,
      "epoch": 2.8483127297026396,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041907357141209256,
      "loss": 4.73,
      "mean_token_accuracy": 0.22686291486024857,
      "num_tokens": 62726264.0,
      "step": 34100
    },
    {
      "entropy": 5.063887357711792,
      "epoch": 2.8487303708653524,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004190507037587639,
      "loss": 4.7181,
      "mean_token_accuracy": 0.22891027480363846,
      "num_tokens": 62734838.0,
      "step": 34105
    },
    {
      "entropy": 5.146815919876099,
      "epoch": 2.8491480120280652,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041902783358365445,
      "loss": 4.8306,
      "mean_token_accuracy": 0.21222046315670012,
      "num_tokens": 62745674.0,
      "step": 34110
    },
    {
      "entropy": 4.99925971031189,
      "epoch": 2.8495656531907785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041900496088716427,
      "loss": 4.7009,
      "mean_token_accuracy": 0.22908754497766495,
      "num_tokens": 62755921.0,
      "step": 34115
    },
    {
      "entropy": 5.032650947570801,
      "epoch": 2.8499832943534917,
      "grad_norm": 1.0625,
      "learning_rate": 0.000418982085669694,
      "loss": 4.7577,
      "mean_token_accuracy": 0.22695206552743913,
      "num_tokens": 62764988.0,
      "step": 34120
    },
    {
      "entropy": 4.972420883178711,
      "epoch": 2.8504009355162045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000418959207931644,
      "loss": 4.5494,
      "mean_token_accuracy": 0.23291562199592591,
      "num_tokens": 62774034.0,
      "step": 34125
    },
    {
      "entropy": 4.960363578796387,
      "epoch": 2.8508185766789174,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004189363276734147,
      "loss": 4.5806,
      "mean_token_accuracy": 0.2411369651556015,
      "num_tokens": 62783339.0,
      "step": 34130
    },
    {
      "entropy": 5.05671272277832,
      "epoch": 2.8512362178416306,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004189134448954068,
      "loss": 4.7023,
      "mean_token_accuracy": 0.23273497819900513,
      "num_tokens": 62793314.0,
      "step": 34135
    },
    {
      "entropy": 5.070263671875,
      "epoch": 2.8516538590043434,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041889055959802075,
      "loss": 4.6936,
      "mean_token_accuracy": 0.23070500791072845,
      "num_tokens": 62802096.0,
      "step": 34140
    },
    {
      "entropy": 5.007365369796753,
      "epoch": 2.8520715001670567,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004188676717816572,
      "loss": 4.6364,
      "mean_token_accuracy": 0.22693876177072525,
      "num_tokens": 62811065.0,
      "step": 34145
    },
    {
      "entropy": 4.999882698059082,
      "epoch": 2.8524891413297695,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041884478144671704,
      "loss": 4.6932,
      "mean_token_accuracy": 0.22778577953577042,
      "num_tokens": 62820425.0,
      "step": 34150
    },
    {
      "entropy": 4.970752143859864,
      "epoch": 2.8529067824924823,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00041882188859360077,
      "loss": 4.6161,
      "mean_token_accuracy": 0.23354039788246156,
      "num_tokens": 62829704.0,
      "step": 34155
    },
    {
      "entropy": 5.079357099533081,
      "epoch": 2.8533244236551956,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004187989932227092,
      "loss": 4.8172,
      "mean_token_accuracy": 0.216727177798748,
      "num_tokens": 62839823.0,
      "step": 34160
    },
    {
      "entropy": 5.054744243621826,
      "epoch": 2.8537420648179084,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041877609533444324,
      "loss": 4.6904,
      "mean_token_accuracy": 0.23427702635526657,
      "num_tokens": 62848739.0,
      "step": 34165
    },
    {
      "entropy": 5.079883432388305,
      "epoch": 2.8541597059806216,
      "grad_norm": 1.125,
      "learning_rate": 0.0004187531949292036,
      "loss": 4.6743,
      "mean_token_accuracy": 0.22138160616159439,
      "num_tokens": 62857716.0,
      "step": 34170
    },
    {
      "entropy": 4.9187901496887205,
      "epoch": 2.8545773471433344,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004187302920073912,
      "loss": 4.5357,
      "mean_token_accuracy": 0.23760683536529542,
      "num_tokens": 62866880.0,
      "step": 34175
    },
    {
      "entropy": 4.942369794845581,
      "epoch": 2.8549949883060473,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041870738656940717,
      "loss": 4.628,
      "mean_token_accuracy": 0.24599756598472594,
      "num_tokens": 62875582.0,
      "step": 34180
    },
    {
      "entropy": 4.991359281539917,
      "epoch": 2.8554126294687605,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004186844786156522,
      "loss": 4.6624,
      "mean_token_accuracy": 0.23386605232954025,
      "num_tokens": 62884465.0,
      "step": 34185
    },
    {
      "entropy": 5.032326078414917,
      "epoch": 2.8558302706314733,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004186615681465276,
      "loss": 4.6458,
      "mean_token_accuracy": 0.22616499662399292,
      "num_tokens": 62893527.0,
      "step": 34190
    },
    {
      "entropy": 5.036355018615723,
      "epoch": 2.8562479117941866,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041863865516243427,
      "loss": 4.7268,
      "mean_token_accuracy": 0.22407529056072234,
      "num_tokens": 62902233.0,
      "step": 34195
    },
    {
      "entropy": 4.97780032157898,
      "epoch": 2.8566655529568994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004186157396637734,
      "loss": 4.5005,
      "mean_token_accuracy": 0.23847667425870894,
      "num_tokens": 62910944.0,
      "step": 34200
    },
    {
      "entropy": 4.976170539855957,
      "epoch": 2.857083194119612,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041859282165094614,
      "loss": 4.6393,
      "mean_token_accuracy": 0.23225167542695999,
      "num_tokens": 62919223.0,
      "step": 34205
    },
    {
      "entropy": 4.898573493957519,
      "epoch": 2.8575008352823255,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041856990112435367,
      "loss": 4.5874,
      "mean_token_accuracy": 0.22590824514627456,
      "num_tokens": 62928429.0,
      "step": 34210
    },
    {
      "entropy": 4.997984266281128,
      "epoch": 2.8579184764450383,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004185469780843972,
      "loss": 4.6719,
      "mean_token_accuracy": 0.22943593114614486,
      "num_tokens": 62937635.0,
      "step": 34215
    },
    {
      "entropy": 5.063547372817993,
      "epoch": 2.8583361176077515,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004185240525314783,
      "loss": 4.76,
      "mean_token_accuracy": 0.22174013108015062,
      "num_tokens": 62946107.0,
      "step": 34220
    },
    {
      "entropy": 5.089448738098144,
      "epoch": 2.8587537587704643,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000418501124465998,
      "loss": 4.714,
      "mean_token_accuracy": 0.2290724113583565,
      "num_tokens": 62955273.0,
      "step": 34225
    },
    {
      "entropy": 5.006072759628296,
      "epoch": 2.859171399933177,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004184781938883577,
      "loss": 4.674,
      "mean_token_accuracy": 0.23638030439615249,
      "num_tokens": 62964318.0,
      "step": 34230
    },
    {
      "entropy": 4.9841429710388185,
      "epoch": 2.8595890410958904,
      "grad_norm": 1.15625,
      "learning_rate": 0.000418455260798959,
      "loss": 4.6443,
      "mean_token_accuracy": 0.233096344769001,
      "num_tokens": 62974102.0,
      "step": 34235
    },
    {
      "entropy": 5.038571357727051,
      "epoch": 2.8600066822586037,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041843232519820316,
      "loss": 4.7161,
      "mean_token_accuracy": 0.22975164651870728,
      "num_tokens": 62983804.0,
      "step": 34240
    },
    {
      "entropy": 4.9466980457305905,
      "epoch": 2.8604243234213165,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004184093870864919,
      "loss": 4.5719,
      "mean_token_accuracy": 0.2327027887105942,
      "num_tokens": 62992633.0,
      "step": 34245
    },
    {
      "entropy": 4.989398193359375,
      "epoch": 2.8608419645840293,
      "grad_norm": 1.0,
      "learning_rate": 0.0004183864464642267,
      "loss": 4.5569,
      "mean_token_accuracy": 0.23770961612462999,
      "num_tokens": 63001720.0,
      "step": 34250
    },
    {
      "entropy": 5.00641188621521,
      "epoch": 2.8612596057467425,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041836350333180917,
      "loss": 4.638,
      "mean_token_accuracy": 0.22936555594205857,
      "num_tokens": 63011610.0,
      "step": 34255
    },
    {
      "entropy": 5.007635688781738,
      "epoch": 2.8616772469094554,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041834055768964095,
      "loss": 4.6387,
      "mean_token_accuracy": 0.23572589308023453,
      "num_tokens": 63019580.0,
      "step": 34260
    },
    {
      "entropy": 5.037321090698242,
      "epoch": 2.8620948880721686,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004183176095381238,
      "loss": 4.7166,
      "mean_token_accuracy": 0.2291743665933609,
      "num_tokens": 63028326.0,
      "step": 34265
    },
    {
      "entropy": 4.9662854194641115,
      "epoch": 2.8625125292348814,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041829465887765924,
      "loss": 4.6036,
      "mean_token_accuracy": 0.23920480012893677,
      "num_tokens": 63037154.0,
      "step": 34270
    },
    {
      "entropy": 4.877361869812011,
      "epoch": 2.8629301703975942,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041827170570864924,
      "loss": 4.5843,
      "mean_token_accuracy": 0.23972396850585936,
      "num_tokens": 63045887.0,
      "step": 34275
    },
    {
      "entropy": 5.053529882431031,
      "epoch": 2.8633478115603075,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004182487500314956,
      "loss": 4.7572,
      "mean_token_accuracy": 0.22738512307405473,
      "num_tokens": 63055019.0,
      "step": 34280
    },
    {
      "entropy": 5.047768259048462,
      "epoch": 2.8637654527230203,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041822579184660016,
      "loss": 4.6465,
      "mean_token_accuracy": 0.23466170877218245,
      "num_tokens": 63063633.0,
      "step": 34285
    },
    {
      "entropy": 4.96454176902771,
      "epoch": 2.8641830938857336,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041820283115436485,
      "loss": 4.651,
      "mean_token_accuracy": 0.23781934529542922,
      "num_tokens": 63072292.0,
      "step": 34290
    },
    {
      "entropy": 5.070100021362305,
      "epoch": 2.8646007350484464,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004181798679551916,
      "loss": 4.7271,
      "mean_token_accuracy": 0.22635700404644013,
      "num_tokens": 63082020.0,
      "step": 34295
    },
    {
      "entropy": 4.971969938278198,
      "epoch": 2.865018376211159,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004181569022494824,
      "loss": 4.6134,
      "mean_token_accuracy": 0.23911563903093339,
      "num_tokens": 63090451.0,
      "step": 34300
    },
    {
      "entropy": 4.9451850891113285,
      "epoch": 2.8654360173738724,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004181339340376393,
      "loss": 4.5663,
      "mean_token_accuracy": 0.24314500093460084,
      "num_tokens": 63099059.0,
      "step": 34305
    },
    {
      "entropy": 5.059632205963135,
      "epoch": 2.8658536585365852,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004181109633200644,
      "loss": 4.737,
      "mean_token_accuracy": 0.2230019062757492,
      "num_tokens": 63108358.0,
      "step": 34310
    },
    {
      "entropy": 5.0801417350769045,
      "epoch": 2.8662712996992985,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004180879900971598,
      "loss": 4.7019,
      "mean_token_accuracy": 0.23192968517541884,
      "num_tokens": 63117504.0,
      "step": 34315
    },
    {
      "entropy": 4.967871856689453,
      "epoch": 2.8666889408620113,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041806501436932774,
      "loss": 4.6745,
      "mean_token_accuracy": 0.2295812785625458,
      "num_tokens": 63127046.0,
      "step": 34320
    },
    {
      "entropy": 5.01019926071167,
      "epoch": 2.867106582024724,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004180420361369704,
      "loss": 4.6521,
      "mean_token_accuracy": 0.24143767207860947,
      "num_tokens": 63135580.0,
      "step": 34325
    },
    {
      "entropy": 5.094692659378052,
      "epoch": 2.8675242231874374,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041801905540049,
      "loss": 4.7952,
      "mean_token_accuracy": 0.21466606110334396,
      "num_tokens": 63145275.0,
      "step": 34330
    },
    {
      "entropy": 5.060675144195557,
      "epoch": 2.86794186435015,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004179960721602889,
      "loss": 4.762,
      "mean_token_accuracy": 0.22556534707546233,
      "num_tokens": 63154861.0,
      "step": 34335
    },
    {
      "entropy": 5.107614326477051,
      "epoch": 2.8683595055128634,
      "grad_norm": 0.984375,
      "learning_rate": 0.00041797308641676944,
      "loss": 4.8108,
      "mean_token_accuracy": 0.2192470461130142,
      "num_tokens": 63163962.0,
      "step": 34340
    },
    {
      "entropy": 5.020327711105347,
      "epoch": 2.8687771466755763,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000417950098170334,
      "loss": 4.7033,
      "mean_token_accuracy": 0.23112028688192368,
      "num_tokens": 63174114.0,
      "step": 34345
    },
    {
      "entropy": 5.016961669921875,
      "epoch": 2.869194787838289,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000417927107421385,
      "loss": 4.6529,
      "mean_token_accuracy": 0.24032869338989257,
      "num_tokens": 63182913.0,
      "step": 34350
    },
    {
      "entropy": 4.962658023834228,
      "epoch": 2.8696124290010023,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041790411417032497,
      "loss": 4.5692,
      "mean_token_accuracy": 0.23414386510849,
      "num_tokens": 63192407.0,
      "step": 34355
    },
    {
      "entropy": 4.9288228988647464,
      "epoch": 2.8700300701637156,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041788111841755646,
      "loss": 4.6236,
      "mean_token_accuracy": 0.23019588142633438,
      "num_tokens": 63201164.0,
      "step": 34360
    },
    {
      "entropy": 4.989523267745971,
      "epoch": 2.8704477113264284,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004178581201634819,
      "loss": 4.7228,
      "mean_token_accuracy": 0.22432279139757155,
      "num_tokens": 63210654.0,
      "step": 34365
    },
    {
      "entropy": 5.0911292552948,
      "epoch": 2.870865352489141,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041783511940850405,
      "loss": 4.688,
      "mean_token_accuracy": 0.23014525324106216,
      "num_tokens": 63219171.0,
      "step": 34370
    },
    {
      "entropy": 5.104710054397583,
      "epoch": 2.8712829936518545,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041781211615302545,
      "loss": 4.692,
      "mean_token_accuracy": 0.23330830037593842,
      "num_tokens": 63228412.0,
      "step": 34375
    },
    {
      "entropy": 4.929451847076416,
      "epoch": 2.8717006348145673,
      "grad_norm": 1.125,
      "learning_rate": 0.0004177891103974489,
      "loss": 4.608,
      "mean_token_accuracy": 0.2430238515138626,
      "num_tokens": 63236929.0,
      "step": 34380
    },
    {
      "entropy": 5.062254333496094,
      "epoch": 2.8721182759772805,
      "grad_norm": 1.0,
      "learning_rate": 0.00041776610214217707,
      "loss": 4.7305,
      "mean_token_accuracy": 0.2251413345336914,
      "num_tokens": 63246179.0,
      "step": 34385
    },
    {
      "entropy": 5.093402671813965,
      "epoch": 2.8725359171399933,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004177430913876128,
      "loss": 4.7112,
      "mean_token_accuracy": 0.22646227180957795,
      "num_tokens": 63255186.0,
      "step": 34390
    },
    {
      "entropy": 5.020917892456055,
      "epoch": 2.872953558302706,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004177200781341588,
      "loss": 4.654,
      "mean_token_accuracy": 0.2313263088464737,
      "num_tokens": 63264954.0,
      "step": 34395
    },
    {
      "entropy": 5.007740497589111,
      "epoch": 2.8733711994654194,
      "grad_norm": 1.125,
      "learning_rate": 0.0004176970623822182,
      "loss": 4.5829,
      "mean_token_accuracy": 0.23482096791267396,
      "num_tokens": 63273586.0,
      "step": 34400
    },
    {
      "entropy": 5.066953563690186,
      "epoch": 2.8737888406281322,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004176740441321936,
      "loss": 4.6714,
      "mean_token_accuracy": 0.23746049106121064,
      "num_tokens": 63283339.0,
      "step": 34405
    },
    {
      "entropy": 5.030169725418091,
      "epoch": 2.8742064817908455,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041765102338448825,
      "loss": 4.6742,
      "mean_token_accuracy": 0.2209380656480789,
      "num_tokens": 63292974.0,
      "step": 34410
    },
    {
      "entropy": 5.024952936172485,
      "epoch": 2.8746241229535583,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041762800013950496,
      "loss": 4.6759,
      "mean_token_accuracy": 0.22923281341791152,
      "num_tokens": 63302924.0,
      "step": 34415
    },
    {
      "entropy": 5.023262405395508,
      "epoch": 2.875041764116271,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041760497439764685,
      "loss": 4.6471,
      "mean_token_accuracy": 0.22994728982448578,
      "num_tokens": 63311168.0,
      "step": 34420
    },
    {
      "entropy": 4.991069412231445,
      "epoch": 2.8754594052789844,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000417581946159317,
      "loss": 4.5825,
      "mean_token_accuracy": 0.23952683210372924,
      "num_tokens": 63320256.0,
      "step": 34425
    },
    {
      "entropy": 4.928984308242798,
      "epoch": 2.875877046441697,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041755891542491854,
      "loss": 4.5527,
      "mean_token_accuracy": 0.24543777257204055,
      "num_tokens": 63328548.0,
      "step": 34430
    },
    {
      "entropy": 4.962594223022461,
      "epoch": 2.8762946876044104,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041753588219485473,
      "loss": 4.6403,
      "mean_token_accuracy": 0.23544154763221742,
      "num_tokens": 63336787.0,
      "step": 34435
    },
    {
      "entropy": 5.08310170173645,
      "epoch": 2.8767123287671232,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004175128464695287,
      "loss": 4.7073,
      "mean_token_accuracy": 0.2279314622282982,
      "num_tokens": 63345674.0,
      "step": 34440
    },
    {
      "entropy": 5.1170634746551515,
      "epoch": 2.877129969929836,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041748980824934376,
      "loss": 4.7076,
      "mean_token_accuracy": 0.22399203181266786,
      "num_tokens": 63354888.0,
      "step": 34445
    },
    {
      "entropy": 4.954849624633789,
      "epoch": 2.8775476110925493,
      "grad_norm": 1.125,
      "learning_rate": 0.0004174667675347032,
      "loss": 4.6048,
      "mean_token_accuracy": 0.2322609931230545,
      "num_tokens": 63363941.0,
      "step": 34450
    },
    {
      "entropy": 4.977200174331665,
      "epoch": 2.877965252255262,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004174437243260104,
      "loss": 4.6097,
      "mean_token_accuracy": 0.23305184543132781,
      "num_tokens": 63372492.0,
      "step": 34455
    },
    {
      "entropy": 5.00695104598999,
      "epoch": 2.8783828934179754,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004174206786236687,
      "loss": 4.7394,
      "mean_token_accuracy": 0.2217859923839569,
      "num_tokens": 63382367.0,
      "step": 34460
    },
    {
      "entropy": 5.085033178329468,
      "epoch": 2.878800534580688,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041739763042808166,
      "loss": 4.7058,
      "mean_token_accuracy": 0.23238953649997712,
      "num_tokens": 63392046.0,
      "step": 34465
    },
    {
      "entropy": 5.035031747817993,
      "epoch": 2.879218175743401,
      "grad_norm": 1.125,
      "learning_rate": 0.00041737457973965264,
      "loss": 4.6901,
      "mean_token_accuracy": 0.22995828837156296,
      "num_tokens": 63401263.0,
      "step": 34470
    },
    {
      "entropy": 4.966222715377808,
      "epoch": 2.8796358169061143,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041735152655878523,
      "loss": 4.5667,
      "mean_token_accuracy": 0.23918326199054718,
      "num_tokens": 63410017.0,
      "step": 34475
    },
    {
      "entropy": 5.005432033538819,
      "epoch": 2.8800534580688275,
      "grad_norm": 1.125,
      "learning_rate": 0.000417328470885883,
      "loss": 4.6176,
      "mean_token_accuracy": 0.23722177743911743,
      "num_tokens": 63418661.0,
      "step": 34480
    },
    {
      "entropy": 4.981644964218139,
      "epoch": 2.8804710992315403,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041730541272134964,
      "loss": 4.6452,
      "mean_token_accuracy": 0.23488765954971313,
      "num_tokens": 63426847.0,
      "step": 34485
    },
    {
      "entropy": 5.112283754348755,
      "epoch": 2.880888740394253,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004172823520655887,
      "loss": 4.8956,
      "mean_token_accuracy": 0.21466597318649291,
      "num_tokens": 63436602.0,
      "step": 34490
    },
    {
      "entropy": 5.095676898956299,
      "epoch": 2.8813063815569664,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041725928891900386,
      "loss": 4.7402,
      "mean_token_accuracy": 0.22825533598661424,
      "num_tokens": 63446057.0,
      "step": 34495
    },
    {
      "entropy": 4.940972089767456,
      "epoch": 2.881724022719679,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041723622328199893,
      "loss": 4.5368,
      "mean_token_accuracy": 0.24192133098840712,
      "num_tokens": 63454612.0,
      "step": 34500
    },
    {
      "entropy": 5.058462762832642,
      "epoch": 2.8821416638823925,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004172131551549778,
      "loss": 4.717,
      "mean_token_accuracy": 0.22621480673551558,
      "num_tokens": 63463158.0,
      "step": 34505
    },
    {
      "entropy": 5.066804838180542,
      "epoch": 2.8825593050451053,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004171900845383441,
      "loss": 4.7718,
      "mean_token_accuracy": 0.2246847778558731,
      "num_tokens": 63473380.0,
      "step": 34510
    },
    {
      "entropy": 5.039558792114258,
      "epoch": 2.882976946207818,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041716701143250184,
      "loss": 4.7291,
      "mean_token_accuracy": 0.23117925822734833,
      "num_tokens": 63482610.0,
      "step": 34515
    },
    {
      "entropy": 5.048277473449707,
      "epoch": 2.8833945873705313,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000417143935837855,
      "loss": 4.6716,
      "mean_token_accuracy": 0.2195283755660057,
      "num_tokens": 63492547.0,
      "step": 34520
    },
    {
      "entropy": 5.004374837875366,
      "epoch": 2.883812228533244,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004171208577548074,
      "loss": 4.5574,
      "mean_token_accuracy": 0.2370736911892891,
      "num_tokens": 63501316.0,
      "step": 34525
    },
    {
      "entropy": 4.863378667831421,
      "epoch": 2.8842298696959574,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041709777718376315,
      "loss": 4.4561,
      "mean_token_accuracy": 0.24623901396989822,
      "num_tokens": 63509387.0,
      "step": 34530
    },
    {
      "entropy": 4.984460735321045,
      "epoch": 2.88464751085867,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041707469412512616,
      "loss": 4.6744,
      "mean_token_accuracy": 0.23377482295036317,
      "num_tokens": 63518144.0,
      "step": 34535
    },
    {
      "entropy": 5.032863473892212,
      "epoch": 2.885065152021383,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041705160857930077,
      "loss": 4.7235,
      "mean_token_accuracy": 0.22125530987977982,
      "num_tokens": 63528021.0,
      "step": 34540
    },
    {
      "entropy": 5.062039852142334,
      "epoch": 2.8854827931840963,
      "grad_norm": 1.125,
      "learning_rate": 0.0004170285205466909,
      "loss": 4.7028,
      "mean_token_accuracy": 0.22741710245609284,
      "num_tokens": 63536750.0,
      "step": 34545
    },
    {
      "entropy": 4.977202892303467,
      "epoch": 2.885900434346809,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004170054300277009,
      "loss": 4.5488,
      "mean_token_accuracy": 0.23960092663764954,
      "num_tokens": 63546312.0,
      "step": 34550
    },
    {
      "entropy": 4.989394569396973,
      "epoch": 2.8863180755095224,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004169823370227348,
      "loss": 4.6247,
      "mean_token_accuracy": 0.23450030833482743,
      "num_tokens": 63555449.0,
      "step": 34555
    },
    {
      "entropy": 4.996150732040405,
      "epoch": 2.886735716672235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041695924153219713,
      "loss": 4.554,
      "mean_token_accuracy": 0.23387723714113234,
      "num_tokens": 63564726.0,
      "step": 34560
    },
    {
      "entropy": 5.010378837585449,
      "epoch": 2.887153357834948,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004169361435564919,
      "loss": 4.6759,
      "mean_token_accuracy": 0.22454145550727844,
      "num_tokens": 63573448.0,
      "step": 34565
    },
    {
      "entropy": 5.032565975189209,
      "epoch": 2.8875709989976612,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004169130430960237,
      "loss": 4.6585,
      "mean_token_accuracy": 0.22526547759771348,
      "num_tokens": 63582459.0,
      "step": 34570
    },
    {
      "entropy": 5.049540996551514,
      "epoch": 2.887988640160374,
      "grad_norm": 1.0,
      "learning_rate": 0.0004168899401511969,
      "loss": 4.6492,
      "mean_token_accuracy": 0.23832936733961105,
      "num_tokens": 63591535.0,
      "step": 34575
    },
    {
      "entropy": 5.02357759475708,
      "epoch": 2.8884062813230873,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004168668347224158,
      "loss": 4.7017,
      "mean_token_accuracy": 0.22832900285720825,
      "num_tokens": 63600500.0,
      "step": 34580
    },
    {
      "entropy": 4.933320760726929,
      "epoch": 2.8888239224858,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041684372681008506,
      "loss": 4.5599,
      "mean_token_accuracy": 0.24078545570373536,
      "num_tokens": 63609365.0,
      "step": 34585
    },
    {
      "entropy": 5.073916053771972,
      "epoch": 2.889241563648513,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004168206164146092,
      "loss": 4.8648,
      "mean_token_accuracy": 0.2187473088502884,
      "num_tokens": 63618989.0,
      "step": 34590
    },
    {
      "entropy": 5.038974189758301,
      "epoch": 2.889659204811226,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041679750353639265,
      "loss": 4.5948,
      "mean_token_accuracy": 0.2399693727493286,
      "num_tokens": 63627938.0,
      "step": 34595
    },
    {
      "entropy": 4.939596700668335,
      "epoch": 2.8900768459739394,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041677438817584023,
      "loss": 4.6567,
      "mean_token_accuracy": 0.23064590394496917,
      "num_tokens": 63635938.0,
      "step": 34600
    },
    {
      "entropy": 5.122353553771973,
      "epoch": 2.8904944871366522,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041675127033335643,
      "loss": 4.7674,
      "mean_token_accuracy": 0.22491547763347625,
      "num_tokens": 63645265.0,
      "step": 34605
    },
    {
      "entropy": 5.058960723876953,
      "epoch": 2.890912128299365,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000416728150009346,
      "loss": 4.759,
      "mean_token_accuracy": 0.22746715545654297,
      "num_tokens": 63654677.0,
      "step": 34610
    },
    {
      "entropy": 4.980344629287719,
      "epoch": 2.8913297694620783,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041670502720421367,
      "loss": 4.6833,
      "mean_token_accuracy": 0.228813736140728,
      "num_tokens": 63664141.0,
      "step": 34615
    },
    {
      "entropy": 4.964523983001709,
      "epoch": 2.891747410624791,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004166819019183643,
      "loss": 4.548,
      "mean_token_accuracy": 0.23944003731012345,
      "num_tokens": 63672372.0,
      "step": 34620
    },
    {
      "entropy": 5.103826713562012,
      "epoch": 2.8921650517875044,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004166587741522027,
      "loss": 4.7862,
      "mean_token_accuracy": 0.2240980789065361,
      "num_tokens": 63682048.0,
      "step": 34625
    },
    {
      "entropy": 5.064325904846191,
      "epoch": 2.892582692950217,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004166356439061338,
      "loss": 4.7004,
      "mean_token_accuracy": 0.2292930379509926,
      "num_tokens": 63691953.0,
      "step": 34630
    },
    {
      "entropy": 4.937698602676392,
      "epoch": 2.89300033411293,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041661251118056236,
      "loss": 4.6223,
      "mean_token_accuracy": 0.2387334704399109,
      "num_tokens": 63700866.0,
      "step": 34635
    },
    {
      "entropy": 5.0404143810272215,
      "epoch": 2.8934179752756433,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004165893759758935,
      "loss": 4.6742,
      "mean_token_accuracy": 0.2305962011218071,
      "num_tokens": 63710606.0,
      "step": 34640
    },
    {
      "entropy": 4.994511985778809,
      "epoch": 2.893835616438356,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004165662382925322,
      "loss": 4.5823,
      "mean_token_accuracy": 0.23580061346292497,
      "num_tokens": 63720213.0,
      "step": 34645
    },
    {
      "entropy": 4.965139245986938,
      "epoch": 2.8942532576010693,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041654309813088353,
      "loss": 4.588,
      "mean_token_accuracy": 0.24403978139162064,
      "num_tokens": 63729472.0,
      "step": 34650
    },
    {
      "entropy": 5.020829916000366,
      "epoch": 2.894670898763782,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004165199554913525,
      "loss": 4.7663,
      "mean_token_accuracy": 0.2246737524867058,
      "num_tokens": 63738568.0,
      "step": 34655
    },
    {
      "entropy": 5.041161251068115,
      "epoch": 2.895088539926495,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004164968103743443,
      "loss": 4.7189,
      "mean_token_accuracy": 0.2356904372572899,
      "num_tokens": 63747598.0,
      "step": 34660
    },
    {
      "entropy": 4.974423503875732,
      "epoch": 2.895506181089208,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041647366278026415,
      "loss": 4.6015,
      "mean_token_accuracy": 0.23526650965213775,
      "num_tokens": 63756651.0,
      "step": 34665
    },
    {
      "entropy": 4.96504316329956,
      "epoch": 2.895923822251921,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004164505127095172,
      "loss": 4.5657,
      "mean_token_accuracy": 0.23249243348836898,
      "num_tokens": 63765930.0,
      "step": 34670
    },
    {
      "entropy": 5.024288511276245,
      "epoch": 2.8963414634146343,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004164273601625088,
      "loss": 4.7113,
      "mean_token_accuracy": 0.23146249055862428,
      "num_tokens": 63775161.0,
      "step": 34675
    },
    {
      "entropy": 5.05126051902771,
      "epoch": 2.896759104577347,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041640420513964414,
      "loss": 4.6748,
      "mean_token_accuracy": 0.23301206082105635,
      "num_tokens": 63783394.0,
      "step": 34680
    },
    {
      "entropy": 4.970368671417236,
      "epoch": 2.89717674574006,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004163810476413287,
      "loss": 4.6536,
      "mean_token_accuracy": 0.23391122370958328,
      "num_tokens": 63792939.0,
      "step": 34685
    },
    {
      "entropy": 4.977013731002808,
      "epoch": 2.897594386902773,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041635788766796795,
      "loss": 4.6476,
      "mean_token_accuracy": 0.22761960029602052,
      "num_tokens": 63802393.0,
      "step": 34690
    },
    {
      "entropy": 4.940495395660401,
      "epoch": 2.8980120280654864,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004163347252199671,
      "loss": 4.6227,
      "mean_token_accuracy": 0.23450478315353393,
      "num_tokens": 63810459.0,
      "step": 34695
    },
    {
      "entropy": 5.0357129096984865,
      "epoch": 2.8984296692281992,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004163115602977318,
      "loss": 4.6129,
      "mean_token_accuracy": 0.24143119752407075,
      "num_tokens": 63819285.0,
      "step": 34700
    },
    {
      "entropy": 5.115005540847778,
      "epoch": 2.898847310390912,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041628839290166754,
      "loss": 4.7843,
      "mean_token_accuracy": 0.21811791360378266,
      "num_tokens": 63829515.0,
      "step": 34705
    },
    {
      "entropy": 4.945247745513916,
      "epoch": 2.8992649515536253,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004162652230321799,
      "loss": 4.6001,
      "mean_token_accuracy": 0.2318314492702484,
      "num_tokens": 63838418.0,
      "step": 34710
    },
    {
      "entropy": 5.0249334335327145,
      "epoch": 2.899682592716338,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004162420506896745,
      "loss": 4.6106,
      "mean_token_accuracy": 0.22939611226320267,
      "num_tokens": 63846753.0,
      "step": 34715
    },
    {
      "entropy": 4.9470884799957275,
      "epoch": 2.9001002338790514,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004162188758745569,
      "loss": 4.5492,
      "mean_token_accuracy": 0.24411656558513642,
      "num_tokens": 63855622.0,
      "step": 34720
    },
    {
      "entropy": 4.995726680755615,
      "epoch": 2.900517875041764,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000416195698587233,
      "loss": 4.6485,
      "mean_token_accuracy": 0.22886679619550704,
      "num_tokens": 63864797.0,
      "step": 34725
    },
    {
      "entropy": 5.023049449920654,
      "epoch": 2.900935516204477,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041617251882810843,
      "loss": 4.7011,
      "mean_token_accuracy": 0.22756680697202683,
      "num_tokens": 63874332.0,
      "step": 34730
    },
    {
      "entropy": 4.977460145950317,
      "epoch": 2.9013531573671902,
      "grad_norm": 1.046875,
      "learning_rate": 0.000416149336597589,
      "loss": 4.6089,
      "mean_token_accuracy": 0.23198046982288362,
      "num_tokens": 63883483.0,
      "step": 34735
    },
    {
      "entropy": 4.972623634338379,
      "epoch": 2.901770798529903,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004161261518960804,
      "loss": 4.5417,
      "mean_token_accuracy": 0.243876214325428,
      "num_tokens": 63892662.0,
      "step": 34740
    },
    {
      "entropy": 5.042520380020141,
      "epoch": 2.9021884396926163,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041610296472398875,
      "loss": 4.6361,
      "mean_token_accuracy": 0.22975926101207733,
      "num_tokens": 63902446.0,
      "step": 34745
    },
    {
      "entropy": 5.0409399509429935,
      "epoch": 2.902606080855329,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004160797750817199,
      "loss": 4.6842,
      "mean_token_accuracy": 0.228150837123394,
      "num_tokens": 63911187.0,
      "step": 34750
    },
    {
      "entropy": 5.036638355255127,
      "epoch": 2.903023722018042,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004160565829696797,
      "loss": 4.7272,
      "mean_token_accuracy": 0.23291797935962677,
      "num_tokens": 63920628.0,
      "step": 34755
    },
    {
      "entropy": 5.003730821609497,
      "epoch": 2.903441363180755,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041603338838827413,
      "loss": 4.659,
      "mean_token_accuracy": 0.23173725306987764,
      "num_tokens": 63929508.0,
      "step": 34760
    },
    {
      "entropy": 4.953013563156128,
      "epoch": 2.903859004343468,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041601019133790946,
      "loss": 4.5613,
      "mean_token_accuracy": 0.23840730637311935,
      "num_tokens": 63938880.0,
      "step": 34765
    },
    {
      "entropy": 5.012861108779907,
      "epoch": 2.9042766455061813,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041598699181899165,
      "loss": 4.6772,
      "mean_token_accuracy": 0.2357049837708473,
      "num_tokens": 63949137.0,
      "step": 34770
    },
    {
      "entropy": 5.024596643447876,
      "epoch": 2.904694286668894,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004159637898319268,
      "loss": 4.6741,
      "mean_token_accuracy": 0.23026847392320632,
      "num_tokens": 63958547.0,
      "step": 34775
    },
    {
      "entropy": 5.013185310363769,
      "epoch": 2.905111927831607,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00041594058537712113,
      "loss": 4.7283,
      "mean_token_accuracy": 0.22153424769639968,
      "num_tokens": 63968120.0,
      "step": 34780
    },
    {
      "entropy": 4.935317516326904,
      "epoch": 2.90552956899432,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041591737845498097,
      "loss": 4.5385,
      "mean_token_accuracy": 0.24569310247898102,
      "num_tokens": 63977503.0,
      "step": 34785
    },
    {
      "entropy": 4.999395656585693,
      "epoch": 2.905947210157033,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004158941690659123,
      "loss": 4.5846,
      "mean_token_accuracy": 0.2355215832591057,
      "num_tokens": 63986192.0,
      "step": 34790
    },
    {
      "entropy": 4.942181015014649,
      "epoch": 2.906364851319746,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041587095721032174,
      "loss": 4.5118,
      "mean_token_accuracy": 0.24505854099988938,
      "num_tokens": 63994584.0,
      "step": 34795
    },
    {
      "entropy": 4.907639026641846,
      "epoch": 2.906782492482459,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004158477428886155,
      "loss": 4.5954,
      "mean_token_accuracy": 0.2437717244029045,
      "num_tokens": 64003291.0,
      "step": 34800
    },
    {
      "entropy": 4.9269617080688475,
      "epoch": 2.907200133645172,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004158245261011999,
      "loss": 4.6045,
      "mean_token_accuracy": 0.23844176530838013,
      "num_tokens": 64013174.0,
      "step": 34805
    },
    {
      "entropy": 5.106709432601929,
      "epoch": 2.907617774807885,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004158013068484816,
      "loss": 4.759,
      "mean_token_accuracy": 0.2282482862472534,
      "num_tokens": 64023265.0,
      "step": 34810
    },
    {
      "entropy": 5.099054002761841,
      "epoch": 2.9080354159705983,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041577808513086677,
      "loss": 4.729,
      "mean_token_accuracy": 0.22577717751264573,
      "num_tokens": 64032123.0,
      "step": 34815
    },
    {
      "entropy": 4.964830923080444,
      "epoch": 2.908453057133311,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041575486094876226,
      "loss": 4.5146,
      "mean_token_accuracy": 0.23585618287324905,
      "num_tokens": 64040388.0,
      "step": 34820
    },
    {
      "entropy": 4.938343954086304,
      "epoch": 2.908870698296024,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041573163430257435,
      "loss": 4.598,
      "mean_token_accuracy": 0.24042887091636658,
      "num_tokens": 64049351.0,
      "step": 34825
    },
    {
      "entropy": 5.025505399703979,
      "epoch": 2.909288339458737,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004157084051927099,
      "loss": 4.7008,
      "mean_token_accuracy": 0.23486513942480086,
      "num_tokens": 64058124.0,
      "step": 34830
    },
    {
      "entropy": 5.0126560688018795,
      "epoch": 2.90970598062145,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004156851736195753,
      "loss": 4.6566,
      "mean_token_accuracy": 0.22981157153844833,
      "num_tokens": 64067452.0,
      "step": 34835
    },
    {
      "entropy": 4.967972993850708,
      "epoch": 2.9101236217841633,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004156619395835775,
      "loss": 4.5708,
      "mean_token_accuracy": 0.2365322858095169,
      "num_tokens": 64076470.0,
      "step": 34840
    },
    {
      "entropy": 4.93115005493164,
      "epoch": 2.910541262946876,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041563870308512313,
      "loss": 4.6267,
      "mean_token_accuracy": 0.23608130514621734,
      "num_tokens": 64085723.0,
      "step": 34845
    },
    {
      "entropy": 4.9920727729797365,
      "epoch": 2.910958904109589,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000415615464124619,
      "loss": 4.7103,
      "mean_token_accuracy": 0.2244643285870552,
      "num_tokens": 64094869.0,
      "step": 34850
    },
    {
      "entropy": 5.074561309814453,
      "epoch": 2.911376545272302,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041559222270247186,
      "loss": 4.7085,
      "mean_token_accuracy": 0.2252213567495346,
      "num_tokens": 64104252.0,
      "step": 34855
    },
    {
      "entropy": 5.015775966644287,
      "epoch": 2.911794186435015,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004155689788190886,
      "loss": 4.6412,
      "mean_token_accuracy": 0.2291973665356636,
      "num_tokens": 64115606.0,
      "step": 34860
    },
    {
      "entropy": 5.016978597640991,
      "epoch": 2.9122118275977282,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041554573247487624,
      "loss": 4.6781,
      "mean_token_accuracy": 0.2316913887858391,
      "num_tokens": 64125131.0,
      "step": 34865
    },
    {
      "entropy": 4.961655712127685,
      "epoch": 2.912629468760441,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041552248367024164,
      "loss": 4.491,
      "mean_token_accuracy": 0.24949485212564468,
      "num_tokens": 64133373.0,
      "step": 34870
    },
    {
      "entropy": 5.063514661788941,
      "epoch": 2.913047109923154,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041549923240559173,
      "loss": 4.7486,
      "mean_token_accuracy": 0.23162557780742646,
      "num_tokens": 64142812.0,
      "step": 34875
    },
    {
      "entropy": 4.894969272613525,
      "epoch": 2.913464751085867,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004154759786813338,
      "loss": 4.6671,
      "mean_token_accuracy": 0.22969848215579985,
      "num_tokens": 64152311.0,
      "step": 34880
    },
    {
      "entropy": 4.959522819519043,
      "epoch": 2.91388239224858,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004154527224978746,
      "loss": 4.6935,
      "mean_token_accuracy": 0.2288597881793976,
      "num_tokens": 64161253.0,
      "step": 34885
    },
    {
      "entropy": 5.046966600418091,
      "epoch": 2.914300033411293,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041542946385562154,
      "loss": 4.6045,
      "mean_token_accuracy": 0.23583932220935822,
      "num_tokens": 64170415.0,
      "step": 34890
    },
    {
      "entropy": 4.9924415111541744,
      "epoch": 2.914717674574006,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004154062027549817,
      "loss": 4.5618,
      "mean_token_accuracy": 0.2347815752029419,
      "num_tokens": 64179680.0,
      "step": 34895
    },
    {
      "entropy": 4.86518292427063,
      "epoch": 2.915135315736719,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041538293919636223,
      "loss": 4.612,
      "mean_token_accuracy": 0.24075186997652054,
      "num_tokens": 64188594.0,
      "step": 34900
    },
    {
      "entropy": 5.044591093063355,
      "epoch": 2.915552956899432,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041535967318017037,
      "loss": 4.7545,
      "mean_token_accuracy": 0.22686173468828202,
      "num_tokens": 64199357.0,
      "step": 34905
    },
    {
      "entropy": 5.03189902305603,
      "epoch": 2.915970598062145,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041533640470681356,
      "loss": 4.6218,
      "mean_token_accuracy": 0.23188094943761825,
      "num_tokens": 64208889.0,
      "step": 34910
    },
    {
      "entropy": 5.0614031791687015,
      "epoch": 2.916388239224858,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004153131337766991,
      "loss": 4.8302,
      "mean_token_accuracy": 0.2242926239967346,
      "num_tokens": 64219432.0,
      "step": 34915
    },
    {
      "entropy": 5.021493244171142,
      "epoch": 2.916805880387571,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041528986039023416,
      "loss": 4.6605,
      "mean_token_accuracy": 0.22614866644144058,
      "num_tokens": 64228143.0,
      "step": 34920
    },
    {
      "entropy": 5.126728057861328,
      "epoch": 2.9172235215502837,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004152665845478265,
      "loss": 4.7277,
      "mean_token_accuracy": 0.22919271886348724,
      "num_tokens": 64237706.0,
      "step": 34925
    },
    {
      "entropy": 4.987454986572265,
      "epoch": 2.917641162712997,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004152433062498834,
      "loss": 4.5923,
      "mean_token_accuracy": 0.23914608359336853,
      "num_tokens": 64246639.0,
      "step": 34930
    },
    {
      "entropy": 4.896990871429443,
      "epoch": 2.9180588038757103,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041522002549681245,
      "loss": 4.5686,
      "mean_token_accuracy": 0.23086388856172563,
      "num_tokens": 64255688.0,
      "step": 34935
    },
    {
      "entropy": 4.85283260345459,
      "epoch": 2.918476445038423,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004151967422890211,
      "loss": 4.5076,
      "mean_token_accuracy": 0.24475740939378737,
      "num_tokens": 64264635.0,
      "step": 34940
    },
    {
      "entropy": 5.0234997272491455,
      "epoch": 2.918894086201136,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041517345662691707,
      "loss": 4.6769,
      "mean_token_accuracy": 0.22886662781238556,
      "num_tokens": 64272686.0,
      "step": 34945
    },
    {
      "entropy": 4.953027009963989,
      "epoch": 2.919311727363849,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004151501685109079,
      "loss": 4.5571,
      "mean_token_accuracy": 0.24017488658428193,
      "num_tokens": 64281891.0,
      "step": 34950
    },
    {
      "entropy": 5.0004072189331055,
      "epoch": 2.919729368526562,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004151268779414014,
      "loss": 4.6189,
      "mean_token_accuracy": 0.23460728526115418,
      "num_tokens": 64290633.0,
      "step": 34955
    },
    {
      "entropy": 5.023769569396973,
      "epoch": 2.920147009689275,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004151035849188053,
      "loss": 4.6239,
      "mean_token_accuracy": 0.23307288736104964,
      "num_tokens": 64300667.0,
      "step": 34960
    },
    {
      "entropy": 5.002420425415039,
      "epoch": 2.920564650851988,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004150802894435271,
      "loss": 4.6275,
      "mean_token_accuracy": 0.23833222240209578,
      "num_tokens": 64309701.0,
      "step": 34965
    },
    {
      "entropy": 4.961587285995483,
      "epoch": 2.920982292014701,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000415056991515975,
      "loss": 4.5871,
      "mean_token_accuracy": 0.2392573907971382,
      "num_tokens": 64318728.0,
      "step": 34970
    },
    {
      "entropy": 4.979658937454223,
      "epoch": 2.921399933177414,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041503369113655655,
      "loss": 4.708,
      "mean_token_accuracy": 0.23013217598199845,
      "num_tokens": 64328510.0,
      "step": 34975
    },
    {
      "entropy": 5.056866216659546,
      "epoch": 2.921817574340127,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004150103883056799,
      "loss": 4.7112,
      "mean_token_accuracy": 0.2296220362186432,
      "num_tokens": 64337566.0,
      "step": 34980
    },
    {
      "entropy": 4.999590730667114,
      "epoch": 2.92223521550284,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041498708302375277,
      "loss": 4.6616,
      "mean_token_accuracy": 0.22976861596107484,
      "num_tokens": 64347009.0,
      "step": 34985
    },
    {
      "entropy": 5.096301889419555,
      "epoch": 2.922652856665553,
      "grad_norm": 0.9375,
      "learning_rate": 0.00041496377529118327,
      "loss": 4.778,
      "mean_token_accuracy": 0.22534715235233307,
      "num_tokens": 64357926.0,
      "step": 34990
    },
    {
      "entropy": 5.032536172866822,
      "epoch": 2.923070497828266,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004149404651083794,
      "loss": 4.6565,
      "mean_token_accuracy": 0.23029040694236755,
      "num_tokens": 64367619.0,
      "step": 34995
    },
    {
      "entropy": 5.1061779975891115,
      "epoch": 2.923488138990979,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004149171524757493,
      "loss": 4.6783,
      "mean_token_accuracy": 0.23096412271261216,
      "num_tokens": 64375954.0,
      "step": 35000
    },
    {
      "entropy": 4.916999578475952,
      "epoch": 2.923905780153692,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041489383739370093,
      "loss": 4.5078,
      "mean_token_accuracy": 0.23693153858184815,
      "num_tokens": 64384794.0,
      "step": 35005
    },
    {
      "entropy": 4.8421995639801025,
      "epoch": 2.924323421316405,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041487051986264264,
      "loss": 4.5548,
      "mean_token_accuracy": 0.2405933067202568,
      "num_tokens": 64393847.0,
      "step": 35010
    },
    {
      "entropy": 4.986077356338501,
      "epoch": 2.924741062479118,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041484719988298254,
      "loss": 4.6073,
      "mean_token_accuracy": 0.23009549379348754,
      "num_tokens": 64403439.0,
      "step": 35015
    },
    {
      "entropy": 5.009205675125122,
      "epoch": 2.9251587036418307,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004148238774551288,
      "loss": 4.6931,
      "mean_token_accuracy": 0.22786321341991425,
      "num_tokens": 64412899.0,
      "step": 35020
    },
    {
      "entropy": 5.007264852523804,
      "epoch": 2.925576344804544,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041480055257948976,
      "loss": 4.6923,
      "mean_token_accuracy": 0.23297519385814666,
      "num_tokens": 64421431.0,
      "step": 35025
    },
    {
      "entropy": 5.052433061599731,
      "epoch": 2.925993985967257,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004147772252564739,
      "loss": 4.804,
      "mean_token_accuracy": 0.22286815643310548,
      "num_tokens": 64430942.0,
      "step": 35030
    },
    {
      "entropy": 5.00485668182373,
      "epoch": 2.92641162712997,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041475389548648934,
      "loss": 4.6543,
      "mean_token_accuracy": 0.22449611127376556,
      "num_tokens": 64440199.0,
      "step": 35035
    },
    {
      "entropy": 5.0656133651733395,
      "epoch": 2.926829268292683,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004147305632699447,
      "loss": 4.6461,
      "mean_token_accuracy": 0.22707951068878174,
      "num_tokens": 64449567.0,
      "step": 35040
    },
    {
      "entropy": 4.959896373748779,
      "epoch": 2.9272469094553957,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004147072286072483,
      "loss": 4.6199,
      "mean_token_accuracy": 0.23580022007226945,
      "num_tokens": 64458840.0,
      "step": 35045
    },
    {
      "entropy": 4.957993078231811,
      "epoch": 2.927664550618109,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004146838914988087,
      "loss": 4.5495,
      "mean_token_accuracy": 0.2353445291519165,
      "num_tokens": 64468029.0,
      "step": 35050
    },
    {
      "entropy": 5.003376436233521,
      "epoch": 2.928082191780822,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004146605519450345,
      "loss": 4.681,
      "mean_token_accuracy": 0.23314826041460038,
      "num_tokens": 64478074.0,
      "step": 35055
    },
    {
      "entropy": 4.876706504821778,
      "epoch": 2.928499832943535,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041463720994633417,
      "loss": 4.5673,
      "mean_token_accuracy": 0.2401459187269211,
      "num_tokens": 64487194.0,
      "step": 35060
    },
    {
      "entropy": 5.008473539352417,
      "epoch": 2.928917474106248,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004146138655031165,
      "loss": 4.7499,
      "mean_token_accuracy": 0.22410617917776107,
      "num_tokens": 64497276.0,
      "step": 35065
    },
    {
      "entropy": 5.029555988311768,
      "epoch": 2.929335115268961,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041459051861578996,
      "loss": 4.6391,
      "mean_token_accuracy": 0.23103071749210358,
      "num_tokens": 64506699.0,
      "step": 35070
    },
    {
      "entropy": 5.004623937606811,
      "epoch": 2.929752756431674,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041456716928476336,
      "loss": 4.6572,
      "mean_token_accuracy": 0.22494951337575914,
      "num_tokens": 64516751.0,
      "step": 35075
    },
    {
      "entropy": 5.02100977897644,
      "epoch": 2.930170397594387,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004145438175104455,
      "loss": 4.6575,
      "mean_token_accuracy": 0.2364555299282074,
      "num_tokens": 64525150.0,
      "step": 35080
    },
    {
      "entropy": 4.962375783920288,
      "epoch": 2.9305880387571,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004145204632932451,
      "loss": 4.5985,
      "mean_token_accuracy": 0.2311079755425453,
      "num_tokens": 64533947.0,
      "step": 35085
    },
    {
      "entropy": 4.934594917297363,
      "epoch": 2.9310056799198128,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004144971066335711,
      "loss": 4.6468,
      "mean_token_accuracy": 0.23274013251066208,
      "num_tokens": 64542349.0,
      "step": 35090
    },
    {
      "entropy": 4.993484258651733,
      "epoch": 2.931423321082526,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041447374753183233,
      "loss": 4.6389,
      "mean_token_accuracy": 0.23117692917585372,
      "num_tokens": 64551681.0,
      "step": 35095
    },
    {
      "entropy": 5.043640375137329,
      "epoch": 2.931840962245239,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004144503859884377,
      "loss": 4.763,
      "mean_token_accuracy": 0.23426744788885118,
      "num_tokens": 64562252.0,
      "step": 35100
    },
    {
      "entropy": 5.027439546585083,
      "epoch": 2.932258603407952,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041442702200379623,
      "loss": 4.6222,
      "mean_token_accuracy": 0.23207056224346162,
      "num_tokens": 64570697.0,
      "step": 35105
    },
    {
      "entropy": 5.02152681350708,
      "epoch": 2.932676244570665,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041440365557831693,
      "loss": 4.6743,
      "mean_token_accuracy": 0.22736351639032365,
      "num_tokens": 64580022.0,
      "step": 35110
    },
    {
      "entropy": 4.943802404403686,
      "epoch": 2.9330938857333777,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041438028671240875,
      "loss": 4.657,
      "mean_token_accuracy": 0.22749731838703155,
      "num_tokens": 64588862.0,
      "step": 35115
    },
    {
      "entropy": 4.968376493453979,
      "epoch": 2.933511526896091,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041435691540648103,
      "loss": 4.6291,
      "mean_token_accuracy": 0.2335723116993904,
      "num_tokens": 64598380.0,
      "step": 35120
    },
    {
      "entropy": 4.996688890457153,
      "epoch": 2.9339291680588038,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004143335416609426,
      "loss": 4.6215,
      "mean_token_accuracy": 0.23534803390502929,
      "num_tokens": 64608159.0,
      "step": 35125
    },
    {
      "entropy": 5.093958806991577,
      "epoch": 2.934346809221517,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004143101654762029,
      "loss": 4.7263,
      "mean_token_accuracy": 0.22414579838514329,
      "num_tokens": 64617940.0,
      "step": 35130
    },
    {
      "entropy": 5.050606679916382,
      "epoch": 2.93476445038423,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041428678685267106,
      "loss": 4.6297,
      "mean_token_accuracy": 0.23532142192125322,
      "num_tokens": 64627399.0,
      "step": 35135
    },
    {
      "entropy": 5.009492588043213,
      "epoch": 2.9351820915469427,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00041426340579075636,
      "loss": 4.5915,
      "mean_token_accuracy": 0.23369524031877517,
      "num_tokens": 64637443.0,
      "step": 35140
    },
    {
      "entropy": 4.959989547729492,
      "epoch": 2.935599732709656,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004142400222908681,
      "loss": 4.7034,
      "mean_token_accuracy": 0.22379156202077866,
      "num_tokens": 64646381.0,
      "step": 35145
    },
    {
      "entropy": 5.009006452560425,
      "epoch": 2.9360173738723687,
      "grad_norm": 1.125,
      "learning_rate": 0.0004142166363534156,
      "loss": 4.6452,
      "mean_token_accuracy": 0.23403552919626236,
      "num_tokens": 64655241.0,
      "step": 35150
    },
    {
      "entropy": 5.006038618087769,
      "epoch": 2.936435015035082,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004141932479788084,
      "loss": 4.6367,
      "mean_token_accuracy": 0.23881327956914902,
      "num_tokens": 64663796.0,
      "step": 35155
    },
    {
      "entropy": 5.0206883430480955,
      "epoch": 2.936852656197795,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041416985716745576,
      "loss": 4.5828,
      "mean_token_accuracy": 0.23457230180501937,
      "num_tokens": 64672620.0,
      "step": 35160
    },
    {
      "entropy": 4.963537311553955,
      "epoch": 2.9372702973605076,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004141464639197673,
      "loss": 4.5771,
      "mean_token_accuracy": 0.23751726746559143,
      "num_tokens": 64681553.0,
      "step": 35165
    },
    {
      "entropy": 5.041569232940674,
      "epoch": 2.937687938523221,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004141230682361525,
      "loss": 4.7122,
      "mean_token_accuracy": 0.2185621902346611,
      "num_tokens": 64690941.0,
      "step": 35170
    },
    {
      "entropy": 4.961095857620239,
      "epoch": 2.938105579685934,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041409967011702087,
      "loss": 4.6178,
      "mean_token_accuracy": 0.23593253940343856,
      "num_tokens": 64699535.0,
      "step": 35175
    },
    {
      "entropy": 4.919082355499268,
      "epoch": 2.938523220848647,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004140762695627822,
      "loss": 4.6482,
      "mean_token_accuracy": 0.2322545051574707,
      "num_tokens": 64708034.0,
      "step": 35180
    },
    {
      "entropy": 5.010694694519043,
      "epoch": 2.9389408620113597,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041405286657384587,
      "loss": 4.6722,
      "mean_token_accuracy": 0.22749723494052887,
      "num_tokens": 64716868.0,
      "step": 35185
    },
    {
      "entropy": 5.087616968154907,
      "epoch": 2.939358503174073,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041402946115062187,
      "loss": 4.6586,
      "mean_token_accuracy": 0.233573254942894,
      "num_tokens": 64726064.0,
      "step": 35190
    },
    {
      "entropy": 5.024508714675903,
      "epoch": 2.939776144336786,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004140060532935197,
      "loss": 4.6024,
      "mean_token_accuracy": 0.2459116756916046,
      "num_tokens": 64734986.0,
      "step": 35195
    },
    {
      "entropy": 4.971071481704712,
      "epoch": 2.940193785499499,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004139826430029493,
      "loss": 4.5816,
      "mean_token_accuracy": 0.24056950807571412,
      "num_tokens": 64743586.0,
      "step": 35200
    },
    {
      "entropy": 4.965557956695557,
      "epoch": 2.940611426662212,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041395923027932045,
      "loss": 4.6913,
      "mean_token_accuracy": 0.22724137753248214,
      "num_tokens": 64752815.0,
      "step": 35205
    },
    {
      "entropy": 5.164203929901123,
      "epoch": 2.9410290678249247,
      "grad_norm": 1.171875,
      "learning_rate": 0.000413935815123043,
      "loss": 4.813,
      "mean_token_accuracy": 0.22795238792896272,
      "num_tokens": 64762364.0,
      "step": 35210
    },
    {
      "entropy": 5.060838317871093,
      "epoch": 2.941446708987638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041391239753452673,
      "loss": 4.7985,
      "mean_token_accuracy": 0.2283138871192932,
      "num_tokens": 64772025.0,
      "step": 35215
    },
    {
      "entropy": 4.95279860496521,
      "epoch": 2.9418643501503507,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004138889775141819,
      "loss": 4.6075,
      "mean_token_accuracy": 0.24099801182746888,
      "num_tokens": 64780251.0,
      "step": 35220
    },
    {
      "entropy": 5.029263257980347,
      "epoch": 2.942281991313064,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004138655550624183,
      "loss": 4.6363,
      "mean_token_accuracy": 0.23622937351465226,
      "num_tokens": 64789227.0,
      "step": 35225
    },
    {
      "entropy": 4.948154783248901,
      "epoch": 2.942699632475777,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004138421301796461,
      "loss": 4.5562,
      "mean_token_accuracy": 0.2377259165048599,
      "num_tokens": 64798010.0,
      "step": 35230
    },
    {
      "entropy": 4.923325157165527,
      "epoch": 2.9431172736384896,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041381870286627515,
      "loss": 4.5884,
      "mean_token_accuracy": 0.23916563689708709,
      "num_tokens": 64807606.0,
      "step": 35235
    },
    {
      "entropy": 4.991160678863525,
      "epoch": 2.943534914801203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041379527312271575,
      "loss": 4.6247,
      "mean_token_accuracy": 0.23552255928516388,
      "num_tokens": 64816516.0,
      "step": 35240
    },
    {
      "entropy": 4.9166907787323,
      "epoch": 2.9439525559639157,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041377184094937796,
      "loss": 4.5732,
      "mean_token_accuracy": 0.24270747005939483,
      "num_tokens": 64825186.0,
      "step": 35245
    },
    {
      "entropy": 5.094076538085938,
      "epoch": 2.944370197126629,
      "grad_norm": 1.125,
      "learning_rate": 0.0004137484063466722,
      "loss": 4.7479,
      "mean_token_accuracy": 0.23116208016872405,
      "num_tokens": 64835232.0,
      "step": 35250
    },
    {
      "entropy": 5.0980737686157225,
      "epoch": 2.9447878382893418,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041372496931500855,
      "loss": 4.6661,
      "mean_token_accuracy": 0.22431330978870392,
      "num_tokens": 64845017.0,
      "step": 35255
    },
    {
      "entropy": 5.0325054168701175,
      "epoch": 2.9452054794520546,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041370152985479723,
      "loss": 4.7363,
      "mean_token_accuracy": 0.22447236478328705,
      "num_tokens": 64853957.0,
      "step": 35260
    },
    {
      "entropy": 4.964989376068115,
      "epoch": 2.945623120614768,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004136780879664487,
      "loss": 4.6853,
      "mean_token_accuracy": 0.23404948115348817,
      "num_tokens": 64862644.0,
      "step": 35265
    },
    {
      "entropy": 5.056398820877075,
      "epoch": 2.9460407617774806,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004136546436503733,
      "loss": 4.797,
      "mean_token_accuracy": 0.2226780101656914,
      "num_tokens": 64872138.0,
      "step": 35270
    },
    {
      "entropy": 4.986986255645752,
      "epoch": 2.946458402940194,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004136311969069815,
      "loss": 4.6796,
      "mean_token_accuracy": 0.229396191239357,
      "num_tokens": 64881287.0,
      "step": 35275
    },
    {
      "entropy": 5.041085910797119,
      "epoch": 2.9468760441029067,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00041360774773668374,
      "loss": 4.7203,
      "mean_token_accuracy": 0.22221737056970597,
      "num_tokens": 64891793.0,
      "step": 35280
    },
    {
      "entropy": 5.154180335998535,
      "epoch": 2.9472936852656195,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041358429613989054,
      "loss": 4.7914,
      "mean_token_accuracy": 0.22040532231330873,
      "num_tokens": 64902181.0,
      "step": 35285
    },
    {
      "entropy": 5.030657148361206,
      "epoch": 2.947711326428333,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004135608421170123,
      "loss": 4.6835,
      "mean_token_accuracy": 0.2268950968980789,
      "num_tokens": 64911160.0,
      "step": 35290
    },
    {
      "entropy": 4.9008386611938475,
      "epoch": 2.948128967591046,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004135373856684599,
      "loss": 4.624,
      "mean_token_accuracy": 0.235726398229599,
      "num_tokens": 64920438.0,
      "step": 35295
    },
    {
      "entropy": 4.957119369506836,
      "epoch": 2.948546608753759,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041351392679464364,
      "loss": 4.6316,
      "mean_token_accuracy": 0.23366450667381286,
      "num_tokens": 64929371.0,
      "step": 35300
    },
    {
      "entropy": 4.979986953735351,
      "epoch": 2.9489642499164717,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004134904654959745,
      "loss": 4.642,
      "mean_token_accuracy": 0.23982950448989868,
      "num_tokens": 64939171.0,
      "step": 35305
    },
    {
      "entropy": 4.95352258682251,
      "epoch": 2.949381891079185,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000413467001772863,
      "loss": 4.5626,
      "mean_token_accuracy": 0.23709273636341094,
      "num_tokens": 64948374.0,
      "step": 35310
    },
    {
      "entropy": 4.958831167221069,
      "epoch": 2.9497995322418977,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041344353562572,
      "loss": 4.6677,
      "mean_token_accuracy": 0.23018596768379213,
      "num_tokens": 64956845.0,
      "step": 35315
    },
    {
      "entropy": 5.01873025894165,
      "epoch": 2.950217173404611,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041342006705495616,
      "loss": 4.699,
      "mean_token_accuracy": 0.22695979326963425,
      "num_tokens": 64965571.0,
      "step": 35320
    },
    {
      "entropy": 4.988738298416138,
      "epoch": 2.950634814567324,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004133965960609825,
      "loss": 4.6318,
      "mean_token_accuracy": 0.22873208224773406,
      "num_tokens": 64974326.0,
      "step": 35325
    },
    {
      "entropy": 5.0295148372650145,
      "epoch": 2.9510524557300366,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004133731226442099,
      "loss": 4.6635,
      "mean_token_accuracy": 0.23387012779712676,
      "num_tokens": 64984404.0,
      "step": 35330
    },
    {
      "entropy": 5.041138315200806,
      "epoch": 2.95147009689275,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004133496468050491,
      "loss": 4.6945,
      "mean_token_accuracy": 0.23368597626686097,
      "num_tokens": 64993899.0,
      "step": 35335
    },
    {
      "entropy": 5.075259780883789,
      "epoch": 2.9518877380554627,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004133261685439113,
      "loss": 4.7526,
      "mean_token_accuracy": 0.22737547010183334,
      "num_tokens": 65003928.0,
      "step": 35340
    },
    {
      "entropy": 4.8919378280639645,
      "epoch": 2.952305379218176,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004133026878612074,
      "loss": 4.5235,
      "mean_token_accuracy": 0.24398411214351653,
      "num_tokens": 65012079.0,
      "step": 35345
    },
    {
      "entropy": 5.013081216812134,
      "epoch": 2.9527230203808887,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004132792047573484,
      "loss": 4.6814,
      "mean_token_accuracy": 0.2249648630619049,
      "num_tokens": 65021122.0,
      "step": 35350
    },
    {
      "entropy": 5.033694887161255,
      "epoch": 2.9531406615436016,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041325571923274564,
      "loss": 4.6583,
      "mean_token_accuracy": 0.2293359011411667,
      "num_tokens": 65030345.0,
      "step": 35355
    },
    {
      "entropy": 5.0734619140625,
      "epoch": 2.953558302706315,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004132322312878099,
      "loss": 4.7041,
      "mean_token_accuracy": 0.22781761586666108,
      "num_tokens": 65038671.0,
      "step": 35360
    },
    {
      "entropy": 4.977784490585327,
      "epoch": 2.9539759438690276,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004132087409229527,
      "loss": 4.6271,
      "mean_token_accuracy": 0.23328956514596938,
      "num_tokens": 65047773.0,
      "step": 35365
    },
    {
      "entropy": 4.9635289192199705,
      "epoch": 2.954393585031741,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041318524813858503,
      "loss": 4.7118,
      "mean_token_accuracy": 0.23145900219678878,
      "num_tokens": 65057543.0,
      "step": 35370
    },
    {
      "entropy": 5.004591417312622,
      "epoch": 2.9548112261944537,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041316175293511833,
      "loss": 4.6364,
      "mean_token_accuracy": 0.22598893642425538,
      "num_tokens": 65066602.0,
      "step": 35375
    },
    {
      "entropy": 5.024883174896241,
      "epoch": 2.9552288673571665,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041313825531296376,
      "loss": 4.7117,
      "mean_token_accuracy": 0.22943028509616853,
      "num_tokens": 65076399.0,
      "step": 35380
    },
    {
      "entropy": 4.955183219909668,
      "epoch": 2.9556465085198798,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041311475527253284,
      "loss": 4.6483,
      "mean_token_accuracy": 0.2329040914773941,
      "num_tokens": 65085222.0,
      "step": 35385
    },
    {
      "entropy": 4.983193874359131,
      "epoch": 2.9560641496825926,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004130912528142368,
      "loss": 4.5983,
      "mean_token_accuracy": 0.2414560064673424,
      "num_tokens": 65094932.0,
      "step": 35390
    },
    {
      "entropy": 5.015091180801392,
      "epoch": 2.956481790845306,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004130677479384873,
      "loss": 4.6876,
      "mean_token_accuracy": 0.22912085950374603,
      "num_tokens": 65104926.0,
      "step": 35395
    },
    {
      "entropy": 4.962862968444824,
      "epoch": 2.9568994320080186,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004130442406456956,
      "loss": 4.6296,
      "mean_token_accuracy": 0.22742156833410263,
      "num_tokens": 65113672.0,
      "step": 35400
    },
    {
      "entropy": 5.019912958145142,
      "epoch": 2.9573170731707314,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041302073093627326,
      "loss": 4.7344,
      "mean_token_accuracy": 0.22050172835588455,
      "num_tokens": 65123465.0,
      "step": 35405
    },
    {
      "entropy": 5.109918260574341,
      "epoch": 2.9577347143334447,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041299721881063196,
      "loss": 4.8149,
      "mean_token_accuracy": 0.2167931392788887,
      "num_tokens": 65133834.0,
      "step": 35410
    },
    {
      "entropy": 5.141035079956055,
      "epoch": 2.958152355496158,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004129737042691832,
      "loss": 4.6901,
      "mean_token_accuracy": 0.22656739503145218,
      "num_tokens": 65143260.0,
      "step": 35415
    },
    {
      "entropy": 4.956868505477905,
      "epoch": 2.9585699966588708,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041295018731233864,
      "loss": 4.5275,
      "mean_token_accuracy": 0.2449065163731575,
      "num_tokens": 65151842.0,
      "step": 35420
    },
    {
      "entropy": 5.0008077144622805,
      "epoch": 2.9589876378215836,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041292666794051,
      "loss": 4.7761,
      "mean_token_accuracy": 0.22514447271823884,
      "num_tokens": 65161517.0,
      "step": 35425
    },
    {
      "entropy": 4.987197542190552,
      "epoch": 2.959405278984297,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041290314615410905,
      "loss": 4.5826,
      "mean_token_accuracy": 0.23759932368993758,
      "num_tokens": 65171147.0,
      "step": 35430
    },
    {
      "entropy": 5.053900527954101,
      "epoch": 2.9598229201470097,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041287962195354753,
      "loss": 4.7485,
      "mean_token_accuracy": 0.2204094260931015,
      "num_tokens": 65180558.0,
      "step": 35435
    },
    {
      "entropy": 5.020114374160767,
      "epoch": 2.960240561309723,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041285609533923723,
      "loss": 4.6959,
      "mean_token_accuracy": 0.22791697680950165,
      "num_tokens": 65189556.0,
      "step": 35440
    },
    {
      "entropy": 4.982840871810913,
      "epoch": 2.9606582024724357,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004128325663115901,
      "loss": 4.569,
      "mean_token_accuracy": 0.239373180270195,
      "num_tokens": 65198489.0,
      "step": 35445
    },
    {
      "entropy": 5.025400638580322,
      "epoch": 2.9610758436351485,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041280903487101795,
      "loss": 4.6586,
      "mean_token_accuracy": 0.2334505572915077,
      "num_tokens": 65207787.0,
      "step": 35450
    },
    {
      "entropy": 4.965394926071167,
      "epoch": 2.961493484797862,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041278550101793275,
      "loss": 4.6633,
      "mean_token_accuracy": 0.23434125930070876,
      "num_tokens": 65217151.0,
      "step": 35455
    },
    {
      "entropy": 5.088230466842651,
      "epoch": 2.9619111259605746,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004127619647527466,
      "loss": 4.6769,
      "mean_token_accuracy": 0.23272960484027863,
      "num_tokens": 65225991.0,
      "step": 35460
    },
    {
      "entropy": 4.980584001541137,
      "epoch": 2.962328767123288,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041273842607587137,
      "loss": 4.6058,
      "mean_token_accuracy": 0.2358881950378418,
      "num_tokens": 65235042.0,
      "step": 35465
    },
    {
      "entropy": 5.016364288330078,
      "epoch": 2.9627464082860007,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041271488498771913,
      "loss": 4.7405,
      "mean_token_accuracy": 0.22655658721923827,
      "num_tokens": 65244603.0,
      "step": 35470
    },
    {
      "entropy": 4.9932090759277346,
      "epoch": 2.9631640494487135,
      "grad_norm": 1.125,
      "learning_rate": 0.00041269134148870214,
      "loss": 4.5884,
      "mean_token_accuracy": 0.2377037763595581,
      "num_tokens": 65253738.0,
      "step": 35475
    },
    {
      "entropy": 5.085699033737183,
      "epoch": 2.9635816906114267,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004126677955792324,
      "loss": 4.7891,
      "mean_token_accuracy": 0.2275228261947632,
      "num_tokens": 65263067.0,
      "step": 35480
    },
    {
      "entropy": 5.102451372146606,
      "epoch": 2.9639993317741395,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004126442472597223,
      "loss": 4.7142,
      "mean_token_accuracy": 0.23367758542299272,
      "num_tokens": 65271685.0,
      "step": 35485
    },
    {
      "entropy": 5.0682610988616945,
      "epoch": 2.964416972936853,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041262069653058385,
      "loss": 4.759,
      "mean_token_accuracy": 0.22687118649482726,
      "num_tokens": 65280969.0,
      "step": 35490
    },
    {
      "entropy": 5.034529066085815,
      "epoch": 2.9648346140995656,
      "grad_norm": 1.125,
      "learning_rate": 0.0004125971433922295,
      "loss": 4.6668,
      "mean_token_accuracy": 0.2308736637234688,
      "num_tokens": 65289753.0,
      "step": 35495
    },
    {
      "entropy": 5.0133050918579105,
      "epoch": 2.9652522552622784,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041257358784507157,
      "loss": 4.6488,
      "mean_token_accuracy": 0.22126165479421617,
      "num_tokens": 65298829.0,
      "step": 35500
    },
    {
      "entropy": 5.0169602394104,
      "epoch": 2.9656698964249917,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004125500298895224,
      "loss": 4.6771,
      "mean_token_accuracy": 0.23218038231134414,
      "num_tokens": 65308193.0,
      "step": 35505
    },
    {
      "entropy": 5.0775469779968265,
      "epoch": 2.9660875375877045,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004125264695259943,
      "loss": 4.7012,
      "mean_token_accuracy": 0.22674076855182648,
      "num_tokens": 65317665.0,
      "step": 35510
    },
    {
      "entropy": 5.0527294158935545,
      "epoch": 2.9665051787504177,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004125029067548999,
      "loss": 4.7228,
      "mean_token_accuracy": 0.23538990765810014,
      "num_tokens": 65328820.0,
      "step": 35515
    },
    {
      "entropy": 5.012055730819702,
      "epoch": 2.9669228199131306,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041247934157665156,
      "loss": 4.6418,
      "mean_token_accuracy": 0.2265583321452141,
      "num_tokens": 65338576.0,
      "step": 35520
    },
    {
      "entropy": 5.006043434143066,
      "epoch": 2.9673404610758434,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004124557739916618,
      "loss": 4.7728,
      "mean_token_accuracy": 0.22184758335351945,
      "num_tokens": 65348095.0,
      "step": 35525
    },
    {
      "entropy": 5.058735799789429,
      "epoch": 2.9677581022385566,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004124322040003434,
      "loss": 4.6162,
      "mean_token_accuracy": 0.23431245237588882,
      "num_tokens": 65357856.0,
      "step": 35530
    },
    {
      "entropy": 4.984920263290405,
      "epoch": 2.96817574340127,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004124086316031087,
      "loss": 4.5517,
      "mean_token_accuracy": 0.23905995190143586,
      "num_tokens": 65366414.0,
      "step": 35535
    },
    {
      "entropy": 5.049533843994141,
      "epoch": 2.9685933845639827,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041238505680037064,
      "loss": 4.7312,
      "mean_token_accuracy": 0.22630893886089326,
      "num_tokens": 65375498.0,
      "step": 35540
    },
    {
      "entropy": 4.974251794815063,
      "epoch": 2.9690110257266955,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041236147959254166,
      "loss": 4.558,
      "mean_token_accuracy": 0.2438766300678253,
      "num_tokens": 65384283.0,
      "step": 35545
    },
    {
      "entropy": 4.903799629211425,
      "epoch": 2.9694286668894088,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041233789998003476,
      "loss": 4.5791,
      "mean_token_accuracy": 0.2324470967054367,
      "num_tokens": 65392777.0,
      "step": 35550
    },
    {
      "entropy": 4.945676565170288,
      "epoch": 2.9698463080521216,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041231431796326254,
      "loss": 4.6008,
      "mean_token_accuracy": 0.23710873126983642,
      "num_tokens": 65402690.0,
      "step": 35555
    },
    {
      "entropy": 5.041498136520386,
      "epoch": 2.970263949214835,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041229073354263787,
      "loss": 4.7617,
      "mean_token_accuracy": 0.22268722355365753,
      "num_tokens": 65413257.0,
      "step": 35560
    },
    {
      "entropy": 5.043329286575317,
      "epoch": 2.9706815903775476,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041226714671857367,
      "loss": 4.6166,
      "mean_token_accuracy": 0.23001018166542053,
      "num_tokens": 65422450.0,
      "step": 35565
    },
    {
      "entropy": 5.052088260650635,
      "epoch": 2.9710992315402605,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004122435574914829,
      "loss": 4.7011,
      "mean_token_accuracy": 0.2268374517560005,
      "num_tokens": 65431513.0,
      "step": 35570
    },
    {
      "entropy": 4.939896917343139,
      "epoch": 2.9715168727029737,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041221996586177843,
      "loss": 4.661,
      "mean_token_accuracy": 0.22732344567775725,
      "num_tokens": 65440424.0,
      "step": 35575
    },
    {
      "entropy": 4.93636794090271,
      "epoch": 2.9719345138656865,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041219637182987324,
      "loss": 4.5211,
      "mean_token_accuracy": 0.240147003531456,
      "num_tokens": 65449024.0,
      "step": 35580
    },
    {
      "entropy": 5.073016214370727,
      "epoch": 2.9723521550284,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00041217277539618046,
      "loss": 4.768,
      "mean_token_accuracy": 0.22103366106748581,
      "num_tokens": 65459269.0,
      "step": 35585
    },
    {
      "entropy": 4.954869413375855,
      "epoch": 2.9727697961911126,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004121491765611131,
      "loss": 4.6458,
      "mean_token_accuracy": 0.23398257046937943,
      "num_tokens": 65469133.0,
      "step": 35590
    },
    {
      "entropy": 5.010003662109375,
      "epoch": 2.9731874373538254,
      "grad_norm": 1.125,
      "learning_rate": 0.00041212557532508434,
      "loss": 4.608,
      "mean_token_accuracy": 0.23394520580768585,
      "num_tokens": 65478814.0,
      "step": 35595
    },
    {
      "entropy": 4.946983289718628,
      "epoch": 2.9736050785165387,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004121019716885074,
      "loss": 4.588,
      "mean_token_accuracy": 0.23611663281917572,
      "num_tokens": 65487960.0,
      "step": 35600
    },
    {
      "entropy": 4.96616005897522,
      "epoch": 2.9740227196792515,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004120783656517953,
      "loss": 4.6486,
      "mean_token_accuracy": 0.2334630697965622,
      "num_tokens": 65497306.0,
      "step": 35605
    },
    {
      "entropy": 4.994467115402221,
      "epoch": 2.9744403608419647,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004120547572153614,
      "loss": 4.6398,
      "mean_token_accuracy": 0.23102869987487792,
      "num_tokens": 65507584.0,
      "step": 35610
    },
    {
      "entropy": 5.077969264984131,
      "epoch": 2.9748580020046775,
      "grad_norm": 1.0,
      "learning_rate": 0.0004120311463796191,
      "loss": 4.7354,
      "mean_token_accuracy": 0.23167214393615723,
      "num_tokens": 65517040.0,
      "step": 35615
    },
    {
      "entropy": 5.04672908782959,
      "epoch": 2.9752756431673903,
      "grad_norm": 1.125,
      "learning_rate": 0.00041200753314498155,
      "loss": 4.6224,
      "mean_token_accuracy": 0.23019203245639802,
      "num_tokens": 65526267.0,
      "step": 35620
    },
    {
      "entropy": 4.977765035629273,
      "epoch": 2.9756932843301036,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041198391751186227,
      "loss": 4.5863,
      "mean_token_accuracy": 0.2336301773786545,
      "num_tokens": 65534871.0,
      "step": 35625
    },
    {
      "entropy": 4.946200847625732,
      "epoch": 2.9761109254928164,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041196029948067464,
      "loss": 4.6421,
      "mean_token_accuracy": 0.22847547233104706,
      "num_tokens": 65543780.0,
      "step": 35630
    },
    {
      "entropy": 4.981973266601562,
      "epoch": 2.9765285666555297,
      "grad_norm": 1.109375,
      "learning_rate": 0.000411936679051832,
      "loss": 4.6156,
      "mean_token_accuracy": 0.2283690720796585,
      "num_tokens": 65552537.0,
      "step": 35635
    },
    {
      "entropy": 4.975505828857422,
      "epoch": 2.9769462078182425,
      "grad_norm": 1.125,
      "learning_rate": 0.00041191305622574815,
      "loss": 4.664,
      "mean_token_accuracy": 0.23865541070699692,
      "num_tokens": 65562059.0,
      "step": 35640
    },
    {
      "entropy": 4.9849388122558596,
      "epoch": 2.9773638489809553,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004118894310028362,
      "loss": 4.6446,
      "mean_token_accuracy": 0.22967188209295272,
      "num_tokens": 65572655.0,
      "step": 35645
    },
    {
      "entropy": 5.063277244567871,
      "epoch": 2.9777814901436686,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041186580338351014,
      "loss": 4.7067,
      "mean_token_accuracy": 0.22717323154211044,
      "num_tokens": 65582204.0,
      "step": 35650
    },
    {
      "entropy": 5.028395366668701,
      "epoch": 2.978199131306382,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004118421733681835,
      "loss": 4.6764,
      "mean_token_accuracy": 0.2299712046980858,
      "num_tokens": 65590926.0,
      "step": 35655
    },
    {
      "entropy": 5.13959903717041,
      "epoch": 2.9786167724690946,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041181854095726986,
      "loss": 4.7677,
      "mean_token_accuracy": 0.22504076808691026,
      "num_tokens": 65600466.0,
      "step": 35660
    },
    {
      "entropy": 4.983289194107056,
      "epoch": 2.9790344136318074,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004117949061511829,
      "loss": 4.6581,
      "mean_token_accuracy": 0.23422613590955735,
      "num_tokens": 65609710.0,
      "step": 35665
    },
    {
      "entropy": 5.057341432571411,
      "epoch": 2.9794520547945207,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041177126895033646,
      "loss": 4.7713,
      "mean_token_accuracy": 0.22718438059091567,
      "num_tokens": 65620439.0,
      "step": 35670
    },
    {
      "entropy": 4.947841739654541,
      "epoch": 2.9798696959572335,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041174762935514435,
      "loss": 4.4911,
      "mean_token_accuracy": 0.2460385739803314,
      "num_tokens": 65629886.0,
      "step": 35675
    },
    {
      "entropy": 5.085532140731812,
      "epoch": 2.9802873371199468,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004117239873660204,
      "loss": 4.7627,
      "mean_token_accuracy": 0.2215793177485466,
      "num_tokens": 65639434.0,
      "step": 35680
    },
    {
      "entropy": 4.950967025756836,
      "epoch": 2.9807049782826596,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004117003429833785,
      "loss": 4.5547,
      "mean_token_accuracy": 0.23725772202014922,
      "num_tokens": 65648221.0,
      "step": 35685
    },
    {
      "entropy": 5.025537776947021,
      "epoch": 2.9811226194453724,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004116766962076325,
      "loss": 4.6964,
      "mean_token_accuracy": 0.23192944824695588,
      "num_tokens": 65656613.0,
      "step": 35690
    },
    {
      "entropy": 4.8823669910430905,
      "epoch": 2.9815402606080856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041165304703919643,
      "loss": 4.5264,
      "mean_token_accuracy": 0.23727023750543594,
      "num_tokens": 65665822.0,
      "step": 35695
    },
    {
      "entropy": 4.998573970794678,
      "epoch": 2.9819579017707984,
      "grad_norm": 1.125,
      "learning_rate": 0.00041162939547848427,
      "loss": 4.64,
      "mean_token_accuracy": 0.23680972158908845,
      "num_tokens": 65675473.0,
      "step": 35700
    },
    {
      "entropy": 5.063964462280273,
      "epoch": 2.9823755429335117,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004116057415259101,
      "loss": 4.7144,
      "mean_token_accuracy": 0.23160064220428467,
      "num_tokens": 65684539.0,
      "step": 35705
    },
    {
      "entropy": 5.045689249038697,
      "epoch": 2.9827931840962245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000411582085181888,
      "loss": 4.6799,
      "mean_token_accuracy": 0.22964898943901063,
      "num_tokens": 65693187.0,
      "step": 35710
    },
    {
      "entropy": 5.008457279205322,
      "epoch": 2.9832108252589373,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00041155842644683213,
      "loss": 4.6657,
      "mean_token_accuracy": 0.23525939881801605,
      "num_tokens": 65703108.0,
      "step": 35715
    },
    {
      "entropy": 4.969210290908814,
      "epoch": 2.9836284664216506,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004115347653211565,
      "loss": 4.6223,
      "mean_token_accuracy": 0.23690847903490067,
      "num_tokens": 65711390.0,
      "step": 35720
    },
    {
      "entropy": 4.937403678894043,
      "epoch": 2.9840461075843634,
      "grad_norm": 1.125,
      "learning_rate": 0.00041151110180527557,
      "loss": 4.5823,
      "mean_token_accuracy": 0.23518804609775543,
      "num_tokens": 65720283.0,
      "step": 35725
    },
    {
      "entropy": 5.049863815307617,
      "epoch": 2.9844637487470767,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004114874358996034,
      "loss": 4.694,
      "mean_token_accuracy": 0.22736530750989914,
      "num_tokens": 65729611.0,
      "step": 35730
    },
    {
      "entropy": 5.002207517623901,
      "epoch": 2.9848813899097895,
      "grad_norm": 1.125,
      "learning_rate": 0.0004114637676045545,
      "loss": 4.6192,
      "mean_token_accuracy": 0.23766732960939407,
      "num_tokens": 65739435.0,
      "step": 35735
    },
    {
      "entropy": 4.986007785797119,
      "epoch": 2.9852990310725023,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041144009692054294,
      "loss": 4.6437,
      "mean_token_accuracy": 0.2388564795255661,
      "num_tokens": 65748740.0,
      "step": 35740
    },
    {
      "entropy": 5.097720575332642,
      "epoch": 2.9857166722352155,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004114164238479834,
      "loss": 4.7656,
      "mean_token_accuracy": 0.22732373774051667,
      "num_tokens": 65757735.0,
      "step": 35745
    },
    {
      "entropy": 4.982567405700683,
      "epoch": 2.9861343133979283,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004113927483872901,
      "loss": 4.5747,
      "mean_token_accuracy": 0.24085482358932495,
      "num_tokens": 65766694.0,
      "step": 35750
    },
    {
      "entropy": 4.98926100730896,
      "epoch": 2.9865519545606416,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004113690705388775,
      "loss": 4.5964,
      "mean_token_accuracy": 0.23387946039438248,
      "num_tokens": 65776571.0,
      "step": 35755
    },
    {
      "entropy": 5.0153909683227536,
      "epoch": 2.9869695957233544,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041134539030316026,
      "loss": 4.6701,
      "mean_token_accuracy": 0.22750888764858246,
      "num_tokens": 65786498.0,
      "step": 35760
    },
    {
      "entropy": 4.978699445724487,
      "epoch": 2.987387236886067,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041132170768055275,
      "loss": 4.6172,
      "mean_token_accuracy": 0.2405083417892456,
      "num_tokens": 65794547.0,
      "step": 35765
    },
    {
      "entropy": 4.900330829620361,
      "epoch": 2.9878048780487805,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004112980226714698,
      "loss": 4.5756,
      "mean_token_accuracy": 0.23649940490722657,
      "num_tokens": 65803374.0,
      "step": 35770
    },
    {
      "entropy": 5.014865827560425,
      "epoch": 2.9882225192114937,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041127433527632574,
      "loss": 4.6008,
      "mean_token_accuracy": 0.23047195076942445,
      "num_tokens": 65812278.0,
      "step": 35775
    },
    {
      "entropy": 4.988469123840332,
      "epoch": 2.9886401603742065,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041125064549553544,
      "loss": 4.6769,
      "mean_token_accuracy": 0.2328192785382271,
      "num_tokens": 65821369.0,
      "step": 35780
    },
    {
      "entropy": 5.025727367401123,
      "epoch": 2.9890578015369194,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041122695332951363,
      "loss": 4.6452,
      "mean_token_accuracy": 0.22902542501688003,
      "num_tokens": 65830616.0,
      "step": 35785
    },
    {
      "entropy": 5.037309408187866,
      "epoch": 2.9894754426996326,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041120325877867504,
      "loss": 4.5711,
      "mean_token_accuracy": 0.2411714091897011,
      "num_tokens": 65839707.0,
      "step": 35790
    },
    {
      "entropy": 4.9681963443756105,
      "epoch": 2.9898930838623454,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004111795618434344,
      "loss": 4.6604,
      "mean_token_accuracy": 0.23333467096090316,
      "num_tokens": 65849688.0,
      "step": 35795
    },
    {
      "entropy": 4.967243719100952,
      "epoch": 2.9903107250250587,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004111558625242067,
      "loss": 4.636,
      "mean_token_accuracy": 0.23449434489011764,
      "num_tokens": 65859829.0,
      "step": 35800
    },
    {
      "entropy": 5.023140907287598,
      "epoch": 2.9907283661877715,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004111321608214067,
      "loss": 4.6764,
      "mean_token_accuracy": 0.2277153715491295,
      "num_tokens": 65869987.0,
      "step": 35805
    },
    {
      "entropy": 4.99981164932251,
      "epoch": 2.9911460073504843,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004111084567354493,
      "loss": 4.5985,
      "mean_token_accuracy": 0.24259741306304933,
      "num_tokens": 65878752.0,
      "step": 35810
    },
    {
      "entropy": 4.972296857833863,
      "epoch": 2.9915636485131976,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041108475026674964,
      "loss": 4.6204,
      "mean_token_accuracy": 0.23677788376808168,
      "num_tokens": 65888276.0,
      "step": 35815
    },
    {
      "entropy": 4.9416248321533205,
      "epoch": 2.9919812896759104,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041106104141572247,
      "loss": 4.6049,
      "mean_token_accuracy": 0.22507477253675462,
      "num_tokens": 65897660.0,
      "step": 35820
    },
    {
      "entropy": 5.015506601333618,
      "epoch": 2.9923989308386236,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004110373301827832,
      "loss": 4.652,
      "mean_token_accuracy": 0.2345539703965187,
      "num_tokens": 65907353.0,
      "step": 35825
    },
    {
      "entropy": 5.056511497497558,
      "epoch": 2.9928165720013364,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004110136165683465,
      "loss": 4.6199,
      "mean_token_accuracy": 0.23965059071779252,
      "num_tokens": 65916424.0,
      "step": 35830
    },
    {
      "entropy": 5.0946797847747805,
      "epoch": 2.9932342131640493,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004109899005728278,
      "loss": 4.6942,
      "mean_token_accuracy": 0.2253382220864296,
      "num_tokens": 65925864.0,
      "step": 35835
    },
    {
      "entropy": 4.947732019424438,
      "epoch": 2.9936518543267625,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041096618219664224,
      "loss": 4.5552,
      "mean_token_accuracy": 0.23427940756082535,
      "num_tokens": 65934072.0,
      "step": 35840
    },
    {
      "entropy": 4.8381105899810795,
      "epoch": 2.9940694954894753,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041094246144020497,
      "loss": 4.5616,
      "mean_token_accuracy": 0.2437281548976898,
      "num_tokens": 65942870.0,
      "step": 35845
    },
    {
      "entropy": 4.9291339874267575,
      "epoch": 2.9944871366521886,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004109187383039313,
      "loss": 4.5647,
      "mean_token_accuracy": 0.23463579416275024,
      "num_tokens": 65950757.0,
      "step": 35850
    },
    {
      "entropy": 4.9765111923217775,
      "epoch": 2.9949047778149014,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004108950127882365,
      "loss": 4.6056,
      "mean_token_accuracy": 0.22989096194505693,
      "num_tokens": 65960109.0,
      "step": 35855
    },
    {
      "entropy": 5.007131481170655,
      "epoch": 2.995322418977614,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041087128489353594,
      "loss": 4.5882,
      "mean_token_accuracy": 0.23819988518953322,
      "num_tokens": 65968814.0,
      "step": 35860
    },
    {
      "entropy": 4.9834636688232425,
      "epoch": 2.9957400601403275,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004108475546202449,
      "loss": 4.6437,
      "mean_token_accuracy": 0.23942963182926177,
      "num_tokens": 65978498.0,
      "step": 35865
    },
    {
      "entropy": 4.9229332447052006,
      "epoch": 2.9961577013030403,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000410823821968779,
      "loss": 4.5601,
      "mean_token_accuracy": 0.23155440390110016,
      "num_tokens": 65987458.0,
      "step": 35870
    },
    {
      "entropy": 4.9916579246521,
      "epoch": 2.9965753424657535,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041080008693955355,
      "loss": 4.6001,
      "mean_token_accuracy": 0.24031443595886232,
      "num_tokens": 65996293.0,
      "step": 35875
    },
    {
      "entropy": 4.985259628295898,
      "epoch": 2.9969929836284663,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041077634953298417,
      "loss": 4.6204,
      "mean_token_accuracy": 0.2283273696899414,
      "num_tokens": 66005357.0,
      "step": 35880
    },
    {
      "entropy": 5.034653520584106,
      "epoch": 2.997410624791179,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004107526097494864,
      "loss": 4.7118,
      "mean_token_accuracy": 0.22305493354797362,
      "num_tokens": 66014230.0,
      "step": 35885
    },
    {
      "entropy": 5.004455852508545,
      "epoch": 2.9978282659538924,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041072886758947565,
      "loss": 4.5794,
      "mean_token_accuracy": 0.24636793434619902,
      "num_tokens": 66022940.0,
      "step": 35890
    },
    {
      "entropy": 5.155687713623047,
      "epoch": 2.9982459071166057,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041070512305336777,
      "loss": 4.8149,
      "mean_token_accuracy": 0.2263651818037033,
      "num_tokens": 66034096.0,
      "step": 35895
    },
    {
      "entropy": 4.908500146865845,
      "epoch": 2.9986635482793185,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004106813761415783,
      "loss": 4.5778,
      "mean_token_accuracy": 0.2387808606028557,
      "num_tokens": 66042972.0,
      "step": 35900
    },
    {
      "entropy": 5.015705204010009,
      "epoch": 2.9990811894420313,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041065762685452313,
      "loss": 4.6684,
      "mean_token_accuracy": 0.23571996092796327,
      "num_tokens": 66051617.0,
      "step": 35905
    },
    {
      "entropy": 4.982551240921021,
      "epoch": 2.9994988306047445,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00041063387519261785,
      "loss": 4.7018,
      "mean_token_accuracy": 0.23263768404722213,
      "num_tokens": 66061713.0,
      "step": 35910
    },
    {
      "entropy": 5.041855573654175,
      "epoch": 2.9999164717674573,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004106101211562783,
      "loss": 4.6898,
      "mean_token_accuracy": 0.23368869572877884,
      "num_tokens": 66070741.0,
      "step": 35915
    },
    {
      "entropy": 5.0081963539123535,
      "epoch": 3.0003341129301706,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041058636474592037,
      "loss": 4.5837,
      "mean_token_accuracy": 0.2464366674423218,
      "num_tokens": 66080183.0,
      "step": 35920
    },
    {
      "entropy": 5.066572427749634,
      "epoch": 3.0007517540928834,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004105626059619599,
      "loss": 4.6215,
      "mean_token_accuracy": 0.23620574176311493,
      "num_tokens": 66088764.0,
      "step": 35925
    },
    {
      "entropy": 4.906340169906616,
      "epoch": 3.0011693952555962,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041053884480481286,
      "loss": 4.4052,
      "mean_token_accuracy": 0.2598342910408974,
      "num_tokens": 66096689.0,
      "step": 35930
    },
    {
      "entropy": 4.940366172790528,
      "epoch": 3.0015870364183095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041051508127489526,
      "loss": 4.6236,
      "mean_token_accuracy": 0.22944509983062744,
      "num_tokens": 66105987.0,
      "step": 35935
    },
    {
      "entropy": 4.991948986053467,
      "epoch": 3.0020046775810223,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041049131537262295,
      "loss": 4.5225,
      "mean_token_accuracy": 0.24454313963651658,
      "num_tokens": 66115397.0,
      "step": 35940
    },
    {
      "entropy": 4.99781231880188,
      "epoch": 3.0024223187437356,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004104675470984121,
      "loss": 4.5281,
      "mean_token_accuracy": 0.24604572653770446,
      "num_tokens": 66123936.0,
      "step": 35945
    },
    {
      "entropy": 4.944944190979004,
      "epoch": 3.0028399599064484,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041044377645267876,
      "loss": 4.518,
      "mean_token_accuracy": 0.23983910530805588,
      "num_tokens": 66133156.0,
      "step": 35950
    },
    {
      "entropy": 4.961456060409546,
      "epoch": 3.003257601069161,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041042000343583906,
      "loss": 4.4965,
      "mean_token_accuracy": 0.24498020857572556,
      "num_tokens": 66141329.0,
      "step": 35955
    },
    {
      "entropy": 4.99146671295166,
      "epoch": 3.0036752422318744,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004103962280483093,
      "loss": 4.5777,
      "mean_token_accuracy": 0.23418734073638917,
      "num_tokens": 66151870.0,
      "step": 35960
    },
    {
      "entropy": 4.965459632873535,
      "epoch": 3.0040928833945872,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004103724502905055,
      "loss": 4.5576,
      "mean_token_accuracy": 0.24200000166893004,
      "num_tokens": 66161230.0,
      "step": 35965
    },
    {
      "entropy": 5.007116603851318,
      "epoch": 3.0045105245573005,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041034867016284405,
      "loss": 4.5835,
      "mean_token_accuracy": 0.23997550308704377,
      "num_tokens": 66169395.0,
      "step": 35970
    },
    {
      "entropy": 4.933009672164917,
      "epoch": 3.0049281657200133,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00041032488766574115,
      "loss": 4.5582,
      "mean_token_accuracy": 0.23522154241800308,
      "num_tokens": 66179315.0,
      "step": 35975
    },
    {
      "entropy": 4.9714240550994875,
      "epoch": 3.0053458068827266,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00041030110279961324,
      "loss": 4.6207,
      "mean_token_accuracy": 0.23216463774442672,
      "num_tokens": 66188183.0,
      "step": 35980
    },
    {
      "entropy": 5.023123931884766,
      "epoch": 3.0057634480454394,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004102773155648767,
      "loss": 4.5509,
      "mean_token_accuracy": 0.2369758203625679,
      "num_tokens": 66196905.0,
      "step": 35985
    },
    {
      "entropy": 4.9001366138458256,
      "epoch": 3.006181089208152,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004102535259619479,
      "loss": 4.519,
      "mean_token_accuracy": 0.24864081889390946,
      "num_tokens": 66205525.0,
      "step": 35990
    },
    {
      "entropy": 4.93825478553772,
      "epoch": 3.0065987303708654,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041022973399124324,
      "loss": 4.5886,
      "mean_token_accuracy": 0.2405540317296982,
      "num_tokens": 66214722.0,
      "step": 35995
    },
    {
      "entropy": 5.0326807498931885,
      "epoch": 3.0070163715335783,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041020593965317945,
      "loss": 4.591,
      "mean_token_accuracy": 0.2370741292834282,
      "num_tokens": 66223826.0,
      "step": 36000
    },
    {
      "epoch": 3.0070163715335783,
      "eval_entropy": 4.820708036523733,
      "eval_loss": 4.676023483276367,
      "eval_mean_token_accuracy": 0.24216299704628155,
      "eval_num_tokens": 66223826.0,
      "eval_runtime": 27.6737,
      "eval_samples_per_second": 1364.58,
      "eval_steps_per_second": 170.595,
      "step": 36000
    }
  ],
  "logging_steps": 5,
  "max_steps": 119720,
  "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": 9.8128149129216e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}