{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.981211220833697,
  "eval_steps": 3000,
  "global_step": 57000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.692021942138672,
      "epoch": 0.0004369483527047103,
      "grad_norm": 13.1875,
      "learning_rate": 2e-06,
      "loss": 10.8244,
      "mean_token_accuracy": 0.0,
      "num_tokens": 10478.0,
      "step": 5
    },
    {
      "entropy": 10.691936111450195,
      "epoch": 0.0008738967054094206,
      "grad_norm": 13.5,
      "learning_rate": 4.5e-06,
      "loss": 10.755,
      "mean_token_accuracy": 0.00011547344038262964,
      "num_tokens": 21152.0,
      "step": 10
    },
    {
      "entropy": 10.69045877456665,
      "epoch": 0.0013108450581141308,
      "grad_norm": 9.875,
      "learning_rate": 7e-06,
      "loss": 10.5011,
      "mean_token_accuracy": 0.034536856785416604,
      "num_tokens": 31246.0,
      "step": 15
    },
    {
      "entropy": 10.65780954360962,
      "epoch": 0.0017477934108188413,
      "grad_norm": 5.6875,
      "learning_rate": 9.5e-06,
      "loss": 10.2113,
      "mean_token_accuracy": 0.04944819919764996,
      "num_tokens": 41934.0,
      "step": 20
    },
    {
      "entropy": 10.511194896697997,
      "epoch": 0.0021847417635235517,
      "grad_norm": 3.5,
      "learning_rate": 1.2e-05,
      "loss": 9.9808,
      "mean_token_accuracy": 0.04893734566867351,
      "num_tokens": 52830.0,
      "step": 25
    },
    {
      "entropy": 10.482695293426513,
      "epoch": 0.0026216901162282617,
      "grad_norm": 2.890625,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.7583,
      "mean_token_accuracy": 0.05473130941390991,
      "num_tokens": 62706.0,
      "step": 30
    },
    {
      "entropy": 10.45745210647583,
      "epoch": 0.003058638468932972,
      "grad_norm": 2.625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.712,
      "mean_token_accuracy": 0.051637019217014316,
      "num_tokens": 72619.0,
      "step": 35
    },
    {
      "entropy": 10.482015132904053,
      "epoch": 0.0034955868216376825,
      "grad_norm": 3.0625,
      "learning_rate": 1.95e-05,
      "loss": 9.7145,
      "mean_token_accuracy": 0.0506310198456049,
      "num_tokens": 82903.0,
      "step": 40
    },
    {
      "entropy": 10.458813381195068,
      "epoch": 0.003932535174342393,
      "grad_norm": 2.328125,
      "learning_rate": 2.2e-05,
      "loss": 9.6082,
      "mean_token_accuracy": 0.06321541368961334,
      "num_tokens": 94243.0,
      "step": 45
    },
    {
      "entropy": 10.395197868347168,
      "epoch": 0.004369483527047103,
      "grad_norm": 2.34375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.6296,
      "mean_token_accuracy": 0.054243823140859605,
      "num_tokens": 103618.0,
      "step": 50
    },
    {
      "entropy": 10.466286754608154,
      "epoch": 0.004806431879751813,
      "grad_norm": 2.203125,
      "learning_rate": 2.7e-05,
      "loss": 9.5593,
      "mean_token_accuracy": 0.06178116090595722,
      "num_tokens": 112180.0,
      "step": 55
    },
    {
      "entropy": 10.310550880432128,
      "epoch": 0.005243380232456523,
      "grad_norm": 2.359375,
      "learning_rate": 2.95e-05,
      "loss": 9.511,
      "mean_token_accuracy": 0.06281424462795257,
      "num_tokens": 122462.0,
      "step": 60
    },
    {
      "entropy": 10.273916530609132,
      "epoch": 0.005680328585161234,
      "grad_norm": 1.953125,
      "learning_rate": 3.2e-05,
      "loss": 9.4532,
      "mean_token_accuracy": 0.06505128778517247,
      "num_tokens": 132741.0,
      "step": 65
    },
    {
      "entropy": 10.26039628982544,
      "epoch": 0.006117276937865944,
      "grad_norm": 1.96875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.2917,
      "mean_token_accuracy": 0.07007538042962551,
      "num_tokens": 142973.0,
      "step": 70
    },
    {
      "entropy": 10.227308750152588,
      "epoch": 0.006554225290570655,
      "grad_norm": 1.984375,
      "learning_rate": 3.7e-05,
      "loss": 9.3612,
      "mean_token_accuracy": 0.06413876824080944,
      "num_tokens": 153199.0,
      "step": 75
    },
    {
      "entropy": 10.246389770507813,
      "epoch": 0.006991173643275365,
      "grad_norm": 2.1875,
      "learning_rate": 3.95e-05,
      "loss": 9.2351,
      "mean_token_accuracy": 0.06828652881085873,
      "num_tokens": 163079.0,
      "step": 80
    },
    {
      "entropy": 10.19476842880249,
      "epoch": 0.0074281219959800755,
      "grad_norm": 1.953125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.1397,
      "mean_token_accuracy": 0.0709759708493948,
      "num_tokens": 173308.0,
      "step": 85
    },
    {
      "entropy": 10.129861736297608,
      "epoch": 0.007865070348684786,
      "grad_norm": 1.8125,
      "learning_rate": 4.45e-05,
      "loss": 9.0684,
      "mean_token_accuracy": 0.08018519915640354,
      "num_tokens": 183856.0,
      "step": 90
    },
    {
      "entropy": 10.133734321594238,
      "epoch": 0.008302018701389496,
      "grad_norm": 1.84375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.9724,
      "mean_token_accuracy": 0.0735650360584259,
      "num_tokens": 193368.0,
      "step": 95
    },
    {
      "entropy": 10.036783123016358,
      "epoch": 0.008738967054094207,
      "grad_norm": 1.6796875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.9113,
      "mean_token_accuracy": 0.070536545291543,
      "num_tokens": 202686.0,
      "step": 100
    },
    {
      "entropy": 9.999727821350097,
      "epoch": 0.009175915406798917,
      "grad_norm": 1.7109375,
      "learning_rate": 5.2e-05,
      "loss": 8.8114,
      "mean_token_accuracy": 0.0759910672903061,
      "num_tokens": 213237.0,
      "step": 105
    },
    {
      "entropy": 9.960642242431641,
      "epoch": 0.009612863759503626,
      "grad_norm": 1.4765625,
      "learning_rate": 5.45e-05,
      "loss": 8.7434,
      "mean_token_accuracy": 0.07438907325267792,
      "num_tokens": 222406.0,
      "step": 110
    },
    {
      "entropy": 9.91720027923584,
      "epoch": 0.010049812112208336,
      "grad_norm": 1.515625,
      "learning_rate": 5.7e-05,
      "loss": 8.6931,
      "mean_token_accuracy": 0.07097940742969513,
      "num_tokens": 232139.0,
      "step": 115
    },
    {
      "entropy": 9.805039215087891,
      "epoch": 0.010486760464913047,
      "grad_norm": 1.2578125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.6117,
      "mean_token_accuracy": 0.06989408992230892,
      "num_tokens": 242438.0,
      "step": 120
    },
    {
      "entropy": 9.689086627960204,
      "epoch": 0.010923708817617757,
      "grad_norm": 1.484375,
      "learning_rate": 6.2e-05,
      "loss": 8.4953,
      "mean_token_accuracy": 0.07219259589910507,
      "num_tokens": 253082.0,
      "step": 125
    },
    {
      "entropy": 9.538719367980956,
      "epoch": 0.011360657170322468,
      "grad_norm": 1.4140625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.4276,
      "mean_token_accuracy": 0.07822968177497387,
      "num_tokens": 264067.0,
      "step": 130
    },
    {
      "entropy": 9.413427066802978,
      "epoch": 0.011797605523027178,
      "grad_norm": 1.4296875,
      "learning_rate": 6.7e-05,
      "loss": 8.3678,
      "mean_token_accuracy": 0.07066886723041535,
      "num_tokens": 274571.0,
      "step": 135
    },
    {
      "entropy": 9.260402202606201,
      "epoch": 0.012234553875731888,
      "grad_norm": 1.0546875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.2521,
      "mean_token_accuracy": 0.07624184787273407,
      "num_tokens": 284561.0,
      "step": 140
    },
    {
      "entropy": 9.093957424163818,
      "epoch": 0.012671502228436599,
      "grad_norm": 1.015625,
      "learning_rate": 7.2e-05,
      "loss": 8.1356,
      "mean_token_accuracy": 0.07842415869235993,
      "num_tokens": 295366.0,
      "step": 145
    },
    {
      "entropy": 8.952532768249512,
      "epoch": 0.01310845058114131,
      "grad_norm": 1.078125,
      "learning_rate": 7.45e-05,
      "loss": 8.2413,
      "mean_token_accuracy": 0.072519838437438,
      "num_tokens": 305423.0,
      "step": 150
    },
    {
      "entropy": 8.831371021270751,
      "epoch": 0.01354539893384602,
      "grad_norm": 1.0625,
      "learning_rate": 7.7e-05,
      "loss": 8.1022,
      "mean_token_accuracy": 0.0731633011251688,
      "num_tokens": 315182.0,
      "step": 155
    },
    {
      "entropy": 8.752772331237793,
      "epoch": 0.01398234728655073,
      "grad_norm": 1.09375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.0907,
      "mean_token_accuracy": 0.07480457425117493,
      "num_tokens": 324780.0,
      "step": 160
    },
    {
      "entropy": 8.644503688812256,
      "epoch": 0.01441929563925544,
      "grad_norm": 0.9296875,
      "learning_rate": 8.2e-05,
      "loss": 8.1188,
      "mean_token_accuracy": 0.07407689988613128,
      "num_tokens": 335104.0,
      "step": 165
    },
    {
      "entropy": 8.571752071380615,
      "epoch": 0.014856243991960151,
      "grad_norm": 1.1015625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 8.0754,
      "mean_token_accuracy": 0.07565453983843326,
      "num_tokens": 345073.0,
      "step": 170
    },
    {
      "entropy": 8.54010829925537,
      "epoch": 0.015293192344664861,
      "grad_norm": 1.0234375,
      "learning_rate": 8.7e-05,
      "loss": 8.0773,
      "mean_token_accuracy": 0.07010089345276356,
      "num_tokens": 354342.0,
      "step": 175
    },
    {
      "entropy": 8.52613639831543,
      "epoch": 0.015730140697369572,
      "grad_norm": 1.109375,
      "learning_rate": 8.95e-05,
      "loss": 7.9826,
      "mean_token_accuracy": 0.07756052985787391,
      "num_tokens": 364441.0,
      "step": 180
    },
    {
      "entropy": 8.392733097076416,
      "epoch": 0.01616708905007428,
      "grad_norm": 1.3828125,
      "learning_rate": 9.2e-05,
      "loss": 7.9982,
      "mean_token_accuracy": 0.07280723676085472,
      "num_tokens": 373918.0,
      "step": 185
    },
    {
      "entropy": 8.41452865600586,
      "epoch": 0.016604037402778993,
      "grad_norm": 0.94140625,
      "learning_rate": 9.45e-05,
      "loss": 8.0324,
      "mean_token_accuracy": 0.0724909171462059,
      "num_tokens": 383273.0,
      "step": 190
    },
    {
      "entropy": 8.292540359497071,
      "epoch": 0.0170409857554837,
      "grad_norm": 1.046875,
      "learning_rate": 9.7e-05,
      "loss": 7.8917,
      "mean_token_accuracy": 0.08331138044595718,
      "num_tokens": 393262.0,
      "step": 195
    },
    {
      "entropy": 8.423866748809814,
      "epoch": 0.017477934108188414,
      "grad_norm": 1.0234375,
      "learning_rate": 9.95e-05,
      "loss": 8.0188,
      "mean_token_accuracy": 0.0748317264020443,
      "num_tokens": 402389.0,
      "step": 200
    },
    {
      "entropy": 8.355186462402344,
      "epoch": 0.017914882460893122,
      "grad_norm": 1.203125,
      "learning_rate": 0.000102,
      "loss": 7.9521,
      "mean_token_accuracy": 0.07337616458535194,
      "num_tokens": 412997.0,
      "step": 205
    },
    {
      "entropy": 8.251720333099366,
      "epoch": 0.018351830813597834,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.9753,
      "mean_token_accuracy": 0.07370316870510578,
      "num_tokens": 422222.0,
      "step": 210
    },
    {
      "entropy": 8.27926950454712,
      "epoch": 0.018788779166302543,
      "grad_norm": 1.015625,
      "learning_rate": 0.000107,
      "loss": 7.9036,
      "mean_token_accuracy": 0.07845620810985565,
      "num_tokens": 432539.0,
      "step": 215
    },
    {
      "entropy": 8.302949237823487,
      "epoch": 0.019225727519007252,
      "grad_norm": 0.96875,
      "learning_rate": 0.0001095,
      "loss": 7.9171,
      "mean_token_accuracy": 0.07200107723474503,
      "num_tokens": 442311.0,
      "step": 220
    },
    {
      "entropy": 8.195553016662597,
      "epoch": 0.019662675871711964,
      "grad_norm": 1.046875,
      "learning_rate": 0.000112,
      "loss": 7.8904,
      "mean_token_accuracy": 0.07533022500574589,
      "num_tokens": 452727.0,
      "step": 225
    },
    {
      "entropy": 8.2304217338562,
      "epoch": 0.020099624224416673,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001145,
      "loss": 7.9264,
      "mean_token_accuracy": 0.07583003640174865,
      "num_tokens": 462461.0,
      "step": 230
    },
    {
      "entropy": 8.261456871032715,
      "epoch": 0.020536572577121385,
      "grad_norm": 1.140625,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.9027,
      "mean_token_accuracy": 0.07816779166460038,
      "num_tokens": 472362.0,
      "step": 235
    },
    {
      "entropy": 8.210902309417724,
      "epoch": 0.020973520929826094,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.8976,
      "mean_token_accuracy": 0.0767062995582819,
      "num_tokens": 482718.0,
      "step": 240
    },
    {
      "entropy": 8.280668449401855,
      "epoch": 0.021410469282530806,
      "grad_norm": 1.109375,
      "learning_rate": 0.000122,
      "loss": 7.8942,
      "mean_token_accuracy": 0.0744877964258194,
      "num_tokens": 492947.0,
      "step": 245
    },
    {
      "entropy": 8.159860420227051,
      "epoch": 0.021847417635235514,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001245,
      "loss": 7.9426,
      "mean_token_accuracy": 0.07315046004951001,
      "num_tokens": 502957.0,
      "step": 250
    },
    {
      "entropy": 8.233509922027588,
      "epoch": 0.022284365987940227,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000127,
      "loss": 7.888,
      "mean_token_accuracy": 0.07607140243053437,
      "num_tokens": 513220.0,
      "step": 255
    },
    {
      "entropy": 8.254329681396484,
      "epoch": 0.022721314340644935,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001295,
      "loss": 7.8441,
      "mean_token_accuracy": 0.07440531551837921,
      "num_tokens": 523217.0,
      "step": 260
    },
    {
      "entropy": 8.120617008209228,
      "epoch": 0.023158262693349647,
      "grad_norm": 1.078125,
      "learning_rate": 0.000132,
      "loss": 7.8656,
      "mean_token_accuracy": 0.07392945140600204,
      "num_tokens": 533789.0,
      "step": 265
    },
    {
      "entropy": 8.175964164733887,
      "epoch": 0.023595211046054356,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.8389,
      "mean_token_accuracy": 0.08100497275590897,
      "num_tokens": 544146.0,
      "step": 270
    },
    {
      "entropy": 8.17584867477417,
      "epoch": 0.02403215939875907,
      "grad_norm": 1.09375,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.8824,
      "mean_token_accuracy": 0.07800561562180519,
      "num_tokens": 554215.0,
      "step": 275
    },
    {
      "entropy": 8.092530584335327,
      "epoch": 0.024469107751463777,
      "grad_norm": 0.984375,
      "learning_rate": 0.0001395,
      "loss": 7.8448,
      "mean_token_accuracy": 0.07754362374544144,
      "num_tokens": 564587.0,
      "step": 280
    },
    {
      "entropy": 8.21709327697754,
      "epoch": 0.024906056104168486,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.8946,
      "mean_token_accuracy": 0.07792591154575348,
      "num_tokens": 574819.0,
      "step": 285
    },
    {
      "entropy": 8.139802265167237,
      "epoch": 0.025343004456873198,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0001445,
      "loss": 7.8252,
      "mean_token_accuracy": 0.07506528235971928,
      "num_tokens": 585178.0,
      "step": 290
    },
    {
      "entropy": 8.118964576721192,
      "epoch": 0.025779952809577907,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000147,
      "loss": 7.8049,
      "mean_token_accuracy": 0.07798612490296364,
      "num_tokens": 595086.0,
      "step": 295
    },
    {
      "entropy": 8.090186500549317,
      "epoch": 0.02621690116228262,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001495,
      "loss": 7.8215,
      "mean_token_accuracy": 0.08201636113226414,
      "num_tokens": 604942.0,
      "step": 300
    },
    {
      "entropy": 8.079523181915283,
      "epoch": 0.026653849514987327,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000152,
      "loss": 7.7878,
      "mean_token_accuracy": 0.07864333167672158,
      "num_tokens": 616082.0,
      "step": 305
    },
    {
      "entropy": 8.156378364562988,
      "epoch": 0.02709079786769204,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.818,
      "mean_token_accuracy": 0.07984638400375843,
      "num_tokens": 626375.0,
      "step": 310
    },
    {
      "entropy": 8.012382888793946,
      "epoch": 0.027527746220396748,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000157,
      "loss": 7.7828,
      "mean_token_accuracy": 0.07687741480767726,
      "num_tokens": 637563.0,
      "step": 315
    },
    {
      "entropy": 8.105079889297485,
      "epoch": 0.02796469457310146,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001595,
      "loss": 7.794,
      "mean_token_accuracy": 0.0784915316849947,
      "num_tokens": 647256.0,
      "step": 320
    },
    {
      "entropy": 8.144533252716064,
      "epoch": 0.02840164292580617,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000162,
      "loss": 7.8405,
      "mean_token_accuracy": 0.07891839221119881,
      "num_tokens": 657247.0,
      "step": 325
    },
    {
      "entropy": 8.161137771606445,
      "epoch": 0.02883859127851088,
      "grad_norm": 1.0625,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.7556,
      "mean_token_accuracy": 0.0753303699195385,
      "num_tokens": 667613.0,
      "step": 330
    },
    {
      "entropy": 7.930547428131104,
      "epoch": 0.02927553963121559,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.796,
      "mean_token_accuracy": 0.07889659702777863,
      "num_tokens": 678048.0,
      "step": 335
    },
    {
      "entropy": 8.04266881942749,
      "epoch": 0.029712487983920302,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.7839,
      "mean_token_accuracy": 0.07976688966155052,
      "num_tokens": 688468.0,
      "step": 340
    },
    {
      "entropy": 8.041753911972046,
      "epoch": 0.03014943633662501,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.7657,
      "mean_token_accuracy": 0.08183549717068672,
      "num_tokens": 699809.0,
      "step": 345
    },
    {
      "entropy": 7.966469430923462,
      "epoch": 0.030586384689329723,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.7193,
      "mean_token_accuracy": 0.0835344910621643,
      "num_tokens": 709227.0,
      "step": 350
    },
    {
      "entropy": 8.117505311965942,
      "epoch": 0.03102333304203443,
      "grad_norm": 0.890625,
      "learning_rate": 0.000177,
      "loss": 7.8113,
      "mean_token_accuracy": 0.0792868234217167,
      "num_tokens": 720178.0,
      "step": 355
    },
    {
      "entropy": 8.02713942527771,
      "epoch": 0.031460281394739144,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0001795,
      "loss": 7.7451,
      "mean_token_accuracy": 0.0808893047273159,
      "num_tokens": 730514.0,
      "step": 360
    },
    {
      "entropy": 7.955661630630493,
      "epoch": 0.03189722974744385,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000182,
      "loss": 7.7936,
      "mean_token_accuracy": 0.07927311882376671,
      "num_tokens": 741345.0,
      "step": 365
    },
    {
      "entropy": 8.073381757736206,
      "epoch": 0.03233417810014856,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001845,
      "loss": 7.7194,
      "mean_token_accuracy": 0.08433770835399627,
      "num_tokens": 751163.0,
      "step": 370
    },
    {
      "entropy": 7.95829062461853,
      "epoch": 0.03277112645285327,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000187,
      "loss": 7.7252,
      "mean_token_accuracy": 0.07551483511924743,
      "num_tokens": 762090.0,
      "step": 375
    },
    {
      "entropy": 8.018740129470824,
      "epoch": 0.033208074805557986,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001895,
      "loss": 7.6853,
      "mean_token_accuracy": 0.08602647930383682,
      "num_tokens": 772049.0,
      "step": 380
    },
    {
      "entropy": 7.922933340072632,
      "epoch": 0.03364502315826269,
      "grad_norm": 1.3125,
      "learning_rate": 0.000192,
      "loss": 7.6569,
      "mean_token_accuracy": 0.08615999668836594,
      "num_tokens": 781280.0,
      "step": 385
    },
    {
      "entropy": 8.000065326690674,
      "epoch": 0.0340819715109674,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001945,
      "loss": 7.7359,
      "mean_token_accuracy": 0.08202896118164063,
      "num_tokens": 791642.0,
      "step": 390
    },
    {
      "entropy": 7.94686861038208,
      "epoch": 0.034518919863672115,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.7103,
      "mean_token_accuracy": 0.07860004678368568,
      "num_tokens": 801224.0,
      "step": 395
    },
    {
      "entropy": 7.999029922485351,
      "epoch": 0.03495586821637683,
      "grad_norm": 1.0625,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.6424,
      "mean_token_accuracy": 0.0833341408520937,
      "num_tokens": 810727.0,
      "step": 400
    },
    {
      "entropy": 8.085517978668213,
      "epoch": 0.03539281656908153,
      "grad_norm": 1.140625,
      "learning_rate": 0.000202,
      "loss": 7.7379,
      "mean_token_accuracy": 0.08233671486377717,
      "num_tokens": 819943.0,
      "step": 405
    },
    {
      "entropy": 7.762646436691284,
      "epoch": 0.035829764921786245,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.6577,
      "mean_token_accuracy": 0.08188907951116561,
      "num_tokens": 829474.0,
      "step": 410
    },
    {
      "entropy": 7.970741319656372,
      "epoch": 0.03626671327449096,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000207,
      "loss": 7.6379,
      "mean_token_accuracy": 0.07744064256548881,
      "num_tokens": 839288.0,
      "step": 415
    },
    {
      "entropy": 7.942607641220093,
      "epoch": 0.03670366162719567,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002095,
      "loss": 7.6614,
      "mean_token_accuracy": 0.08381943926215171,
      "num_tokens": 850064.0,
      "step": 420
    },
    {
      "entropy": 7.798444652557373,
      "epoch": 0.037140609979900374,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000212,
      "loss": 7.6415,
      "mean_token_accuracy": 0.0831109531223774,
      "num_tokens": 860732.0,
      "step": 425
    },
    {
      "entropy": 7.909991359710693,
      "epoch": 0.037577558332605086,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002145,
      "loss": 7.6336,
      "mean_token_accuracy": 0.08323059007525443,
      "num_tokens": 870076.0,
      "step": 430
    },
    {
      "entropy": 7.853689670562744,
      "epoch": 0.0380145066853098,
      "grad_norm": 1.078125,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.5262,
      "mean_token_accuracy": 0.08503322303295135,
      "num_tokens": 880201.0,
      "step": 435
    },
    {
      "entropy": 7.877675151824951,
      "epoch": 0.038451455038014504,
      "grad_norm": 1.0,
      "learning_rate": 0.0002195,
      "loss": 7.5925,
      "mean_token_accuracy": 0.08455978333950043,
      "num_tokens": 890344.0,
      "step": 440
    },
    {
      "entropy": 7.76024980545044,
      "epoch": 0.038888403390719216,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000222,
      "loss": 7.5838,
      "mean_token_accuracy": 0.08586194440722465,
      "num_tokens": 900443.0,
      "step": 445
    },
    {
      "entropy": 7.874179077148438,
      "epoch": 0.03932535174342393,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0002245,
      "loss": 7.6019,
      "mean_token_accuracy": 0.08362102881073952,
      "num_tokens": 910284.0,
      "step": 450
    },
    {
      "entropy": 7.942639446258545,
      "epoch": 0.03976230009612864,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.692,
      "mean_token_accuracy": 0.08145863935351372,
      "num_tokens": 920986.0,
      "step": 455
    },
    {
      "entropy": 7.793898105621338,
      "epoch": 0.040199248448833345,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.5795,
      "mean_token_accuracy": 0.08823852688074112,
      "num_tokens": 930481.0,
      "step": 460
    },
    {
      "entropy": 7.896108341217041,
      "epoch": 0.04063619680153806,
      "grad_norm": 1.046875,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.615,
      "mean_token_accuracy": 0.08752772808074952,
      "num_tokens": 940554.0,
      "step": 465
    },
    {
      "entropy": 7.816930770874023,
      "epoch": 0.04107314515424277,
      "grad_norm": 1.0,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.5654,
      "mean_token_accuracy": 0.08801438808441162,
      "num_tokens": 950005.0,
      "step": 470
    },
    {
      "entropy": 7.777540969848633,
      "epoch": 0.04151009350694748,
      "grad_norm": 1.21875,
      "learning_rate": 0.000237,
      "loss": 7.5509,
      "mean_token_accuracy": 0.08131005018949508,
      "num_tokens": 959414.0,
      "step": 475
    },
    {
      "entropy": 7.822865295410156,
      "epoch": 0.04194704185965219,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002395,
      "loss": 7.5807,
      "mean_token_accuracy": 0.08017202988266944,
      "num_tokens": 969352.0,
      "step": 480
    },
    {
      "entropy": 7.85607123374939,
      "epoch": 0.0423839902123569,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000242,
      "loss": 7.6151,
      "mean_token_accuracy": 0.07956655099987983,
      "num_tokens": 980294.0,
      "step": 485
    },
    {
      "entropy": 7.760293769836426,
      "epoch": 0.04282093856506161,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002445,
      "loss": 7.4934,
      "mean_token_accuracy": 0.08754161521792411,
      "num_tokens": 990453.0,
      "step": 490
    },
    {
      "entropy": 7.754083633422852,
      "epoch": 0.04325788691776632,
      "grad_norm": 1.171875,
      "learning_rate": 0.000247,
      "loss": 7.514,
      "mean_token_accuracy": 0.09044907167553902,
      "num_tokens": 1000509.0,
      "step": 495
    },
    {
      "entropy": 7.716276216506958,
      "epoch": 0.04369483527047103,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002495,
      "loss": 7.6554,
      "mean_token_accuracy": 0.07662305198609828,
      "num_tokens": 1011747.0,
      "step": 500
    },
    {
      "entropy": 7.786904191970825,
      "epoch": 0.04413178362317574,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000252,
      "loss": 7.5966,
      "mean_token_accuracy": 0.08083347380161285,
      "num_tokens": 1021511.0,
      "step": 505
    },
    {
      "entropy": 7.863558530807495,
      "epoch": 0.04456873197588045,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002545,
      "loss": 7.6139,
      "mean_token_accuracy": 0.08253132700920104,
      "num_tokens": 1032116.0,
      "step": 510
    },
    {
      "entropy": 7.755859327316284,
      "epoch": 0.04500568032858516,
      "grad_norm": 1.03125,
      "learning_rate": 0.000257,
      "loss": 7.4897,
      "mean_token_accuracy": 0.08958897739648819,
      "num_tokens": 1041356.0,
      "step": 515
    },
    {
      "entropy": 7.731158351898193,
      "epoch": 0.04544262868128987,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002595,
      "loss": 7.5513,
      "mean_token_accuracy": 0.08268988244235516,
      "num_tokens": 1051314.0,
      "step": 520
    },
    {
      "entropy": 7.751654243469238,
      "epoch": 0.04587957703399458,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000262,
      "loss": 7.5319,
      "mean_token_accuracy": 0.08553770706057548,
      "num_tokens": 1060857.0,
      "step": 525
    },
    {
      "entropy": 7.804953765869141,
      "epoch": 0.046316525386699295,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.5423,
      "mean_token_accuracy": 0.08373178914189339,
      "num_tokens": 1071744.0,
      "step": 530
    },
    {
      "entropy": 7.729137563705445,
      "epoch": 0.046753473739404,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.5074,
      "mean_token_accuracy": 0.0874439723789692,
      "num_tokens": 1080777.0,
      "step": 535
    },
    {
      "entropy": 7.7326695919036865,
      "epoch": 0.04719042209210871,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.5089,
      "mean_token_accuracy": 0.0891585111618042,
      "num_tokens": 1090951.0,
      "step": 540
    },
    {
      "entropy": 7.670050811767578,
      "epoch": 0.047627370444813424,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.4552,
      "mean_token_accuracy": 0.08354937732219696,
      "num_tokens": 1101667.0,
      "step": 545
    },
    {
      "entropy": 7.7255943775177,
      "epoch": 0.04806431879751814,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002745,
      "loss": 7.5694,
      "mean_token_accuracy": 0.08456205204129219,
      "num_tokens": 1112652.0,
      "step": 550
    },
    {
      "entropy": 7.709576559066773,
      "epoch": 0.04850126715022284,
      "grad_norm": 1.125,
      "learning_rate": 0.000277,
      "loss": 7.529,
      "mean_token_accuracy": 0.08200532495975495,
      "num_tokens": 1122518.0,
      "step": 555
    },
    {
      "entropy": 7.815584325790406,
      "epoch": 0.048938215502927554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002795,
      "loss": 7.5337,
      "mean_token_accuracy": 0.08656449392437934,
      "num_tokens": 1132867.0,
      "step": 560
    },
    {
      "entropy": 7.730601501464844,
      "epoch": 0.049375163855632266,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.6063,
      "mean_token_accuracy": 0.08351293206214905,
      "num_tokens": 1142524.0,
      "step": 565
    },
    {
      "entropy": 7.762125158309937,
      "epoch": 0.04981211220833697,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002845,
      "loss": 7.466,
      "mean_token_accuracy": 0.09219523333013058,
      "num_tokens": 1152983.0,
      "step": 570
    },
    {
      "entropy": 7.690439462661743,
      "epoch": 0.050249060561041684,
      "grad_norm": 1.015625,
      "learning_rate": 0.000287,
      "loss": 7.5518,
      "mean_token_accuracy": 0.08294261172413826,
      "num_tokens": 1164271.0,
      "step": 575
    },
    {
      "entropy": 7.625344181060791,
      "epoch": 0.050686008913746396,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002895,
      "loss": 7.3901,
      "mean_token_accuracy": 0.08981689289212227,
      "num_tokens": 1173567.0,
      "step": 580
    },
    {
      "entropy": 7.597426557540894,
      "epoch": 0.05112295726645111,
      "grad_norm": 1.09375,
      "learning_rate": 0.000292,
      "loss": 7.4494,
      "mean_token_accuracy": 0.09101380854845047,
      "num_tokens": 1183619.0,
      "step": 585
    },
    {
      "entropy": 7.662474060058594,
      "epoch": 0.05155990561915581,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002945,
      "loss": 7.4014,
      "mean_token_accuracy": 0.0849751003086567,
      "num_tokens": 1193922.0,
      "step": 590
    },
    {
      "entropy": 7.686445474624634,
      "epoch": 0.051996853971860525,
      "grad_norm": 1.125,
      "learning_rate": 0.000297,
      "loss": 7.4905,
      "mean_token_accuracy": 0.0850920021533966,
      "num_tokens": 1203665.0,
      "step": 595
    },
    {
      "entropy": 7.688158273696899,
      "epoch": 0.05243380232456524,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002995,
      "loss": 7.4868,
      "mean_token_accuracy": 0.08693959787487984,
      "num_tokens": 1214119.0,
      "step": 600
    },
    {
      "entropy": 7.659633779525757,
      "epoch": 0.05287075067726995,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000302,
      "loss": 7.5142,
      "mean_token_accuracy": 0.08122221603989602,
      "num_tokens": 1225275.0,
      "step": 605
    },
    {
      "entropy": 7.575573873519898,
      "epoch": 0.053307699029974655,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003045,
      "loss": 7.39,
      "mean_token_accuracy": 0.08491945862770081,
      "num_tokens": 1235199.0,
      "step": 610
    },
    {
      "entropy": 7.622638702392578,
      "epoch": 0.05374464738267937,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000307,
      "loss": 7.4131,
      "mean_token_accuracy": 0.08687365427613258,
      "num_tokens": 1245648.0,
      "step": 615
    },
    {
      "entropy": 7.542647027969361,
      "epoch": 0.05418159573538408,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003095,
      "loss": 7.3606,
      "mean_token_accuracy": 0.09443019926548005,
      "num_tokens": 1254793.0,
      "step": 620
    },
    {
      "entropy": 7.674170923233032,
      "epoch": 0.05461854408808879,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000312,
      "loss": 7.4023,
      "mean_token_accuracy": 0.09142766371369362,
      "num_tokens": 1264429.0,
      "step": 625
    },
    {
      "entropy": 7.4895281314849855,
      "epoch": 0.055055492440793496,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003145,
      "loss": 7.3853,
      "mean_token_accuracy": 0.0873162291944027,
      "num_tokens": 1274449.0,
      "step": 630
    },
    {
      "entropy": 7.683523368835449,
      "epoch": 0.05549244079349821,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000317,
      "loss": 7.4964,
      "mean_token_accuracy": 0.08166366070508957,
      "num_tokens": 1284987.0,
      "step": 635
    },
    {
      "entropy": 7.648572826385498,
      "epoch": 0.05592938914620292,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003195,
      "loss": 7.4905,
      "mean_token_accuracy": 0.07985021024942399,
      "num_tokens": 1295984.0,
      "step": 640
    },
    {
      "entropy": 7.492455101013183,
      "epoch": 0.056366337498907626,
      "grad_norm": 1.0625,
      "learning_rate": 0.000322,
      "loss": 7.3794,
      "mean_token_accuracy": 0.09080474004149437,
      "num_tokens": 1305738.0,
      "step": 645
    },
    {
      "entropy": 7.597794008255005,
      "epoch": 0.05680328585161234,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.3331,
      "mean_token_accuracy": 0.0831803061068058,
      "num_tokens": 1315691.0,
      "step": 650
    },
    {
      "entropy": 7.567058038711548,
      "epoch": 0.05724023420431705,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.3392,
      "mean_token_accuracy": 0.0938805028796196,
      "num_tokens": 1326195.0,
      "step": 655
    },
    {
      "entropy": 7.406193971633911,
      "epoch": 0.05767718255702176,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.3802,
      "mean_token_accuracy": 0.09240260273218155,
      "num_tokens": 1336916.0,
      "step": 660
    },
    {
      "entropy": 7.54069561958313,
      "epoch": 0.05811413090972647,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.3984,
      "mean_token_accuracy": 0.08775042369961739,
      "num_tokens": 1347386.0,
      "step": 665
    },
    {
      "entropy": 7.649399089813232,
      "epoch": 0.05855107926243118,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.3597,
      "mean_token_accuracy": 0.08884215950965882,
      "num_tokens": 1357959.0,
      "step": 670
    },
    {
      "entropy": 7.446434020996094,
      "epoch": 0.05898802761513589,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000337,
      "loss": 7.3605,
      "mean_token_accuracy": 0.08795115426182747,
      "num_tokens": 1368135.0,
      "step": 675
    },
    {
      "entropy": 7.542385911941528,
      "epoch": 0.059424975967840604,
      "grad_norm": 1.0,
      "learning_rate": 0.0003395,
      "loss": 7.3913,
      "mean_token_accuracy": 0.08776068612933159,
      "num_tokens": 1378702.0,
      "step": 680
    },
    {
      "entropy": 7.405993509292602,
      "epoch": 0.05986192432054531,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000342,
      "loss": 7.2451,
      "mean_token_accuracy": 0.09290849268436432,
      "num_tokens": 1387902.0,
      "step": 685
    },
    {
      "entropy": 7.529328966140747,
      "epoch": 0.06029887267325002,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.2709,
      "mean_token_accuracy": 0.09545634016394615,
      "num_tokens": 1397288.0,
      "step": 690
    },
    {
      "entropy": 7.438186883926392,
      "epoch": 0.060735821025954734,
      "grad_norm": 1.09375,
      "learning_rate": 0.000347,
      "loss": 7.3202,
      "mean_token_accuracy": 0.09044579640030861,
      "num_tokens": 1407416.0,
      "step": 695
    },
    {
      "entropy": 7.5576756477355955,
      "epoch": 0.061172769378659446,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003495,
      "loss": 7.3774,
      "mean_token_accuracy": 0.08624047338962555,
      "num_tokens": 1417733.0,
      "step": 700
    },
    {
      "entropy": 7.467697048187256,
      "epoch": 0.06160971773136415,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000352,
      "loss": 7.2659,
      "mean_token_accuracy": 0.08791227973997592,
      "num_tokens": 1426723.0,
      "step": 705
    },
    {
      "entropy": 7.462012910842896,
      "epoch": 0.06204666608406886,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003545,
      "loss": 7.3369,
      "mean_token_accuracy": 0.09189029186964034,
      "num_tokens": 1437290.0,
      "step": 710
    },
    {
      "entropy": 7.403400897979736,
      "epoch": 0.062483614436773575,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000357,
      "loss": 7.2678,
      "mean_token_accuracy": 0.08849345222115516,
      "num_tokens": 1446756.0,
      "step": 715
    },
    {
      "entropy": 7.42365288734436,
      "epoch": 0.06292056278947829,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003595,
      "loss": 7.3374,
      "mean_token_accuracy": 0.08893298208713532,
      "num_tokens": 1457074.0,
      "step": 720
    },
    {
      "entropy": 7.430225610733032,
      "epoch": 0.06335751114218299,
      "grad_norm": 1.296875,
      "learning_rate": 0.000362,
      "loss": 7.3969,
      "mean_token_accuracy": 0.08372474275529385,
      "num_tokens": 1467785.0,
      "step": 725
    },
    {
      "entropy": 7.446598958969116,
      "epoch": 0.0637944594948877,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003645,
      "loss": 7.285,
      "mean_token_accuracy": 0.09203022196888924,
      "num_tokens": 1477532.0,
      "step": 730
    },
    {
      "entropy": 7.365743780136109,
      "epoch": 0.06423140784759242,
      "grad_norm": 1.0625,
      "learning_rate": 0.000367,
      "loss": 7.2159,
      "mean_token_accuracy": 0.09424635395407677,
      "num_tokens": 1487102.0,
      "step": 735
    },
    {
      "entropy": 7.39603099822998,
      "epoch": 0.06466835620029712,
      "grad_norm": 1.125,
      "learning_rate": 0.0003695,
      "loss": 7.3035,
      "mean_token_accuracy": 0.08960460349917412,
      "num_tokens": 1497459.0,
      "step": 740
    },
    {
      "entropy": 7.429128408432007,
      "epoch": 0.06510530455300184,
      "grad_norm": 0.90625,
      "learning_rate": 0.000372,
      "loss": 7.295,
      "mean_token_accuracy": 0.08902842849493027,
      "num_tokens": 1508756.0,
      "step": 745
    },
    {
      "entropy": 7.3875891208648685,
      "epoch": 0.06554225290570655,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003745,
      "loss": 7.204,
      "mean_token_accuracy": 0.09589074179530144,
      "num_tokens": 1518735.0,
      "step": 750
    },
    {
      "entropy": 7.386076498031616,
      "epoch": 0.06597920125841125,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000377,
      "loss": 7.2494,
      "mean_token_accuracy": 0.09206888377666474,
      "num_tokens": 1528632.0,
      "step": 755
    },
    {
      "entropy": 7.327648305892945,
      "epoch": 0.06641614961111597,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003795,
      "loss": 7.2499,
      "mean_token_accuracy": 0.08991473093628884,
      "num_tokens": 1538180.0,
      "step": 760
    },
    {
      "entropy": 7.369539976119995,
      "epoch": 0.06685309796382068,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000382,
      "loss": 7.2692,
      "mean_token_accuracy": 0.09155759960412979,
      "num_tokens": 1547788.0,
      "step": 765
    },
    {
      "entropy": 7.339311122894287,
      "epoch": 0.06729004631652538,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003845,
      "loss": 7.2099,
      "mean_token_accuracy": 0.09492280036211014,
      "num_tokens": 1558008.0,
      "step": 770
    },
    {
      "entropy": 7.356014108657837,
      "epoch": 0.0677269946692301,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.2706,
      "mean_token_accuracy": 0.09146112427115441,
      "num_tokens": 1567728.0,
      "step": 775
    },
    {
      "entropy": 7.376091766357422,
      "epoch": 0.0681639430219348,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.2229,
      "mean_token_accuracy": 0.0900360755622387,
      "num_tokens": 1578488.0,
      "step": 780
    },
    {
      "entropy": 7.3270101070404055,
      "epoch": 0.06860089137463951,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.1983,
      "mean_token_accuracy": 0.09384592846035958,
      "num_tokens": 1588070.0,
      "step": 785
    },
    {
      "entropy": 7.3354650974273685,
      "epoch": 0.06903783972734423,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.2717,
      "mean_token_accuracy": 0.09093666598200797,
      "num_tokens": 1598572.0,
      "step": 790
    },
    {
      "entropy": 7.418518686294556,
      "epoch": 0.06947478808004894,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.2727,
      "mean_token_accuracy": 0.08963424041867256,
      "num_tokens": 1607982.0,
      "step": 795
    },
    {
      "entropy": 7.34418568611145,
      "epoch": 0.06991173643275365,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003995,
      "loss": 7.287,
      "mean_token_accuracy": 0.09368083029985427,
      "num_tokens": 1618785.0,
      "step": 800
    },
    {
      "entropy": 7.331365442276001,
      "epoch": 0.07034868478545836,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000402,
      "loss": 7.1991,
      "mean_token_accuracy": 0.09436822384595871,
      "num_tokens": 1627988.0,
      "step": 805
    },
    {
      "entropy": 7.347531366348266,
      "epoch": 0.07078563313816306,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004045,
      "loss": 7.2996,
      "mean_token_accuracy": 0.08484280183911323,
      "num_tokens": 1638024.0,
      "step": 810
    },
    {
      "entropy": 7.368394899368286,
      "epoch": 0.07122258149086778,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.2934,
      "mean_token_accuracy": 0.08453431576490403,
      "num_tokens": 1649215.0,
      "step": 815
    },
    {
      "entropy": 7.250621223449707,
      "epoch": 0.07165952984357249,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004095,
      "loss": 7.1812,
      "mean_token_accuracy": 0.09315369427204132,
      "num_tokens": 1659198.0,
      "step": 820
    },
    {
      "entropy": 7.30466251373291,
      "epoch": 0.0720964781962772,
      "grad_norm": 1.15625,
      "learning_rate": 0.000412,
      "loss": 7.2168,
      "mean_token_accuracy": 0.08589582666754722,
      "num_tokens": 1668972.0,
      "step": 825
    },
    {
      "entropy": 7.230190420150757,
      "epoch": 0.07253342654898191,
      "grad_norm": 1.125,
      "learning_rate": 0.0004145,
      "loss": 7.2035,
      "mean_token_accuracy": 0.09499170146882534,
      "num_tokens": 1679807.0,
      "step": 830
    },
    {
      "entropy": 7.315043592453003,
      "epoch": 0.07297037490168662,
      "grad_norm": 1.171875,
      "learning_rate": 0.000417,
      "loss": 7.2514,
      "mean_token_accuracy": 0.08835556209087372,
      "num_tokens": 1690567.0,
      "step": 835
    },
    {
      "entropy": 7.1730207920074465,
      "epoch": 0.07340732325439134,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004195,
      "loss": 7.1465,
      "mean_token_accuracy": 0.09503189548850059,
      "num_tokens": 1700939.0,
      "step": 840
    },
    {
      "entropy": 7.252574443817139,
      "epoch": 0.07384427160709604,
      "grad_norm": 1.0,
      "learning_rate": 0.000422,
      "loss": 7.1955,
      "mean_token_accuracy": 0.09737142100930214,
      "num_tokens": 1711181.0,
      "step": 845
    },
    {
      "entropy": 7.33465609550476,
      "epoch": 0.07428121995980075,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004245,
      "loss": 7.2235,
      "mean_token_accuracy": 0.08807378485798836,
      "num_tokens": 1721392.0,
      "step": 850
    },
    {
      "entropy": 7.280315160751343,
      "epoch": 0.07471816831250547,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000427,
      "loss": 7.1462,
      "mean_token_accuracy": 0.09444593712687492,
      "num_tokens": 1731942.0,
      "step": 855
    },
    {
      "entropy": 7.246977376937866,
      "epoch": 0.07515511666521017,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004295,
      "loss": 7.2516,
      "mean_token_accuracy": 0.09143168479204178,
      "num_tokens": 1743409.0,
      "step": 860
    },
    {
      "entropy": 7.194182825088501,
      "epoch": 0.07559206501791488,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000432,
      "loss": 7.1375,
      "mean_token_accuracy": 0.09851979911327362,
      "num_tokens": 1753425.0,
      "step": 865
    },
    {
      "entropy": 7.228073787689209,
      "epoch": 0.0760290133706196,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004345,
      "loss": 7.1724,
      "mean_token_accuracy": 0.08897863328456879,
      "num_tokens": 1762722.0,
      "step": 870
    },
    {
      "entropy": 7.394821119308472,
      "epoch": 0.0764659617233243,
      "grad_norm": 1.078125,
      "learning_rate": 0.000437,
      "loss": 7.2404,
      "mean_token_accuracy": 0.0846147008240223,
      "num_tokens": 1771895.0,
      "step": 875
    },
    {
      "entropy": 7.15571780204773,
      "epoch": 0.07690291007602901,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004395,
      "loss": 7.1449,
      "mean_token_accuracy": 0.09327926933765411,
      "num_tokens": 1782175.0,
      "step": 880
    },
    {
      "entropy": 7.320403099060059,
      "epoch": 0.07733985842873373,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000442,
      "loss": 7.1938,
      "mean_token_accuracy": 0.09331970289349556,
      "num_tokens": 1792409.0,
      "step": 885
    },
    {
      "entropy": 7.294836235046387,
      "epoch": 0.07777680678143843,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004445,
      "loss": 7.2449,
      "mean_token_accuracy": 0.08916250318288803,
      "num_tokens": 1802721.0,
      "step": 890
    },
    {
      "entropy": 7.145002126693726,
      "epoch": 0.07821375513414315,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000447,
      "loss": 7.0706,
      "mean_token_accuracy": 0.09213897436857224,
      "num_tokens": 1812064.0,
      "step": 895
    },
    {
      "entropy": 7.291088485717774,
      "epoch": 0.07865070348684786,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.1945,
      "mean_token_accuracy": 0.08576769083738327,
      "num_tokens": 1822284.0,
      "step": 900
    },
    {
      "entropy": 7.199422168731689,
      "epoch": 0.07908765183955256,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.0905,
      "mean_token_accuracy": 0.09272913187742234,
      "num_tokens": 1832150.0,
      "step": 905
    },
    {
      "entropy": 7.295376396179199,
      "epoch": 0.07952460019225728,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.2072,
      "mean_token_accuracy": 0.08950737938284874,
      "num_tokens": 1842870.0,
      "step": 910
    },
    {
      "entropy": 7.063530540466308,
      "epoch": 0.07996154854496199,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.0461,
      "mean_token_accuracy": 0.09950183033943176,
      "num_tokens": 1852771.0,
      "step": 915
    },
    {
      "entropy": 7.172847270965576,
      "epoch": 0.08039849689766669,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.169,
      "mean_token_accuracy": 0.09458636716008187,
      "num_tokens": 1863095.0,
      "step": 920
    },
    {
      "entropy": 7.164995765686035,
      "epoch": 0.08083544525037141,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000462,
      "loss": 7.1697,
      "mean_token_accuracy": 0.09454121589660644,
      "num_tokens": 1873060.0,
      "step": 925
    },
    {
      "entropy": 7.242373037338257,
      "epoch": 0.08127239360307612,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004645,
      "loss": 7.0998,
      "mean_token_accuracy": 0.09765185564756393,
      "num_tokens": 1882574.0,
      "step": 930
    },
    {
      "entropy": 7.123794221878052,
      "epoch": 0.08170934195578082,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000467,
      "loss": 7.1642,
      "mean_token_accuracy": 0.09597432464361191,
      "num_tokens": 1892227.0,
      "step": 935
    },
    {
      "entropy": 7.194331502914428,
      "epoch": 0.08214629030848554,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004695,
      "loss": 7.1898,
      "mean_token_accuracy": 0.0918419398367405,
      "num_tokens": 1902868.0,
      "step": 940
    },
    {
      "entropy": 7.1636861801147464,
      "epoch": 0.08258323866119024,
      "grad_norm": 0.984375,
      "learning_rate": 0.000472,
      "loss": 7.1298,
      "mean_token_accuracy": 0.09083555936813355,
      "num_tokens": 1912781.0,
      "step": 945
    },
    {
      "entropy": 7.1555884838104244,
      "epoch": 0.08302018701389496,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004745,
      "loss": 7.1202,
      "mean_token_accuracy": 0.09932551309466361,
      "num_tokens": 1923638.0,
      "step": 950
    },
    {
      "entropy": 7.204519844055175,
      "epoch": 0.08345713536659967,
      "grad_norm": 1.015625,
      "learning_rate": 0.000477,
      "loss": 7.0845,
      "mean_token_accuracy": 0.0983235090970993,
      "num_tokens": 1933498.0,
      "step": 955
    },
    {
      "entropy": 7.0537872314453125,
      "epoch": 0.08389408371930437,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004795,
      "loss": 7.1033,
      "mean_token_accuracy": 0.09563426300883293,
      "num_tokens": 1943798.0,
      "step": 960
    },
    {
      "entropy": 7.188605308532715,
      "epoch": 0.0843310320720091,
      "grad_norm": 1.109375,
      "learning_rate": 0.000482,
      "loss": 7.14,
      "mean_token_accuracy": 0.09312030673027039,
      "num_tokens": 1954235.0,
      "step": 965
    },
    {
      "entropy": 7.173063611984253,
      "epoch": 0.0847679804247138,
      "grad_norm": 1.0,
      "learning_rate": 0.0004845,
      "loss": 7.1231,
      "mean_token_accuracy": 0.09110933616757393,
      "num_tokens": 1964421.0,
      "step": 970
    },
    {
      "entropy": 7.087044048309326,
      "epoch": 0.0852049287774185,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000487,
      "loss": 7.1059,
      "mean_token_accuracy": 0.09609248787164688,
      "num_tokens": 1975862.0,
      "step": 975
    },
    {
      "entropy": 7.187036323547363,
      "epoch": 0.08564187713012322,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004895,
      "loss": 7.1055,
      "mean_token_accuracy": 0.09382803216576577,
      "num_tokens": 1985758.0,
      "step": 980
    },
    {
      "entropy": 7.038282966613769,
      "epoch": 0.08607882548282793,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000492,
      "loss": 7.079,
      "mean_token_accuracy": 0.0931548923254013,
      "num_tokens": 1996904.0,
      "step": 985
    },
    {
      "entropy": 7.26365966796875,
      "epoch": 0.08651577383553263,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004945,
      "loss": 7.1611,
      "mean_token_accuracy": 0.09324508160352707,
      "num_tokens": 2006114.0,
      "step": 990
    },
    {
      "entropy": 7.074654531478882,
      "epoch": 0.08695272218823735,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000497,
      "loss": 7.0961,
      "mean_token_accuracy": 0.10224229097366333,
      "num_tokens": 2016349.0,
      "step": 995
    },
    {
      "entropy": 7.075801420211792,
      "epoch": 0.08738967054094206,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995,
      "loss": 7.0411,
      "mean_token_accuracy": 0.09501954540610313,
      "num_tokens": 2026701.0,
      "step": 1000
    },
    {
      "entropy": 7.036496257781982,
      "epoch": 0.08782661889364678,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999999986190023,
      "loss": 7.0382,
      "mean_token_accuracy": 0.09739973098039627,
      "num_tokens": 2036950.0,
      "step": 1005
    },
    {
      "entropy": 7.158604335784912,
      "epoch": 0.08826356724635148,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999999930086991,
      "loss": 7.0451,
      "mean_token_accuracy": 0.09806649461388588,
      "num_tokens": 2047655.0,
      "step": 1010
    },
    {
      "entropy": 7.032751893997192,
      "epoch": 0.08870051559905619,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499999983082778,
      "loss": 7.0286,
      "mean_token_accuracy": 0.0988488368690014,
      "num_tokens": 2056800.0,
      "step": 1015
    },
    {
      "entropy": 7.048862934112549,
      "epoch": 0.0891374639517609,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999999688412394,
      "loss": 7.0621,
      "mean_token_accuracy": 0.09844208061695099,
      "num_tokens": 2066862.0,
      "step": 1020
    },
    {
      "entropy": 7.1450450897216795,
      "epoch": 0.08957441230446561,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999999502840836,
      "loss": 7.0754,
      "mean_token_accuracy": 0.10040949657559395,
      "num_tokens": 2076585.0,
      "step": 1025
    },
    {
      "entropy": 7.057808780670166,
      "epoch": 0.09001136065717032,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999999274113107,
      "loss": 7.1469,
      "mean_token_accuracy": 0.09293891340494156,
      "num_tokens": 2087768.0,
      "step": 1030
    },
    {
      "entropy": 7.1115809917449955,
      "epoch": 0.09044830900987504,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999999002229214,
      "loss": 7.0581,
      "mean_token_accuracy": 0.09576725885272026,
      "num_tokens": 2097027.0,
      "step": 1035
    },
    {
      "entropy": 7.053522729873658,
      "epoch": 0.09088525736257974,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999998687189161,
      "loss": 6.9899,
      "mean_token_accuracy": 0.0992590568959713,
      "num_tokens": 2107055.0,
      "step": 1040
    },
    {
      "entropy": 7.008532524108887,
      "epoch": 0.09132220571528446,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999998328992954,
      "loss": 6.9253,
      "mean_token_accuracy": 0.1000043123960495,
      "num_tokens": 2116771.0,
      "step": 1045
    },
    {
      "entropy": 7.04312653541565,
      "epoch": 0.09175915406798917,
      "grad_norm": 1.140625,
      "learning_rate": 0.00049999979276406,
      "loss": 7.055,
      "mean_token_accuracy": 0.10065520852804184,
      "num_tokens": 2126881.0,
      "step": 1050
    },
    {
      "entropy": 7.017210865020752,
      "epoch": 0.09219610242069387,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999997483132108,
      "loss": 7.0334,
      "mean_token_accuracy": 0.0969104826450348,
      "num_tokens": 2136964.0,
      "step": 1055
    },
    {
      "entropy": 7.135955476760865,
      "epoch": 0.09263305077339859,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999996995467483,
      "loss": 7.077,
      "mean_token_accuracy": 0.09480597153306007,
      "num_tokens": 2147263.0,
      "step": 1060
    },
    {
      "entropy": 6.98509464263916,
      "epoch": 0.0930699991261033,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999996464646738,
      "loss": 7.0904,
      "mean_token_accuracy": 0.09320821538567543,
      "num_tokens": 2157474.0,
      "step": 1065
    },
    {
      "entropy": 7.075286531448365,
      "epoch": 0.093506947478808,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999995890669882,
      "loss": 7.0241,
      "mean_token_accuracy": 0.09614943489432334,
      "num_tokens": 2167701.0,
      "step": 1070
    },
    {
      "entropy": 6.915413427352905,
      "epoch": 0.09394389583151272,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999995273536925,
      "loss": 6.998,
      "mean_token_accuracy": 0.1028121143579483,
      "num_tokens": 2178581.0,
      "step": 1075
    },
    {
      "entropy": 7.083644104003906,
      "epoch": 0.09438084418421742,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499999461324788,
      "loss": 6.988,
      "mean_token_accuracy": 0.09867269322276115,
      "num_tokens": 2189529.0,
      "step": 1080
    },
    {
      "entropy": 6.977810955047607,
      "epoch": 0.09481779253692213,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999993909802759,
      "loss": 7.0636,
      "mean_token_accuracy": 0.09306210651993752,
      "num_tokens": 2199805.0,
      "step": 1085
    },
    {
      "entropy": 7.00365309715271,
      "epoch": 0.09525474088962685,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999993163201577,
      "loss": 7.0044,
      "mean_token_accuracy": 0.09724057838320732,
      "num_tokens": 2210108.0,
      "step": 1090
    },
    {
      "entropy": 7.164579772949219,
      "epoch": 0.09569168924233155,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999992373444347,
      "loss": 7.0016,
      "mean_token_accuracy": 0.09872443079948426,
      "num_tokens": 2220325.0,
      "step": 1095
    },
    {
      "entropy": 6.893038415908814,
      "epoch": 0.09612863759503627,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999991540531084,
      "loss": 6.9957,
      "mean_token_accuracy": 0.0961542434990406,
      "num_tokens": 2230521.0,
      "step": 1100
    },
    {
      "entropy": 7.113245630264283,
      "epoch": 0.09656558594774098,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999990664461804,
      "loss": 7.0346,
      "mean_token_accuracy": 0.0963917538523674,
      "num_tokens": 2240372.0,
      "step": 1105
    },
    {
      "entropy": 7.000665855407715,
      "epoch": 0.09700253430044568,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999989745236524,
      "loss": 6.9949,
      "mean_token_accuracy": 0.10437084212899209,
      "num_tokens": 2250522.0,
      "step": 1110
    },
    {
      "entropy": 6.946602296829224,
      "epoch": 0.0974394826531504,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999988782855261,
      "loss": 6.9321,
      "mean_token_accuracy": 0.09969532936811447,
      "num_tokens": 2260142.0,
      "step": 1115
    },
    {
      "entropy": 7.026501417160034,
      "epoch": 0.09787643100585511,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999987777318035,
      "loss": 6.9769,
      "mean_token_accuracy": 0.09623479098081589,
      "num_tokens": 2270372.0,
      "step": 1120
    },
    {
      "entropy": 6.853050279617309,
      "epoch": 0.09831337935855981,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999986728624865,
      "loss": 6.8943,
      "mean_token_accuracy": 0.10797081291675567,
      "num_tokens": 2280517.0,
      "step": 1125
    },
    {
      "entropy": 7.0577069282531735,
      "epoch": 0.09875032771126453,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499998563677577,
      "loss": 7.001,
      "mean_token_accuracy": 0.09663023054599762,
      "num_tokens": 2290778.0,
      "step": 1130
    },
    {
      "entropy": 7.018427753448487,
      "epoch": 0.09918727606396924,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999984501770771,
      "loss": 7.0469,
      "mean_token_accuracy": 0.09329727813601493,
      "num_tokens": 2301692.0,
      "step": 1135
    },
    {
      "entropy": 6.9677423477172855,
      "epoch": 0.09962422441667394,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999983323609891,
      "loss": 6.9154,
      "mean_token_accuracy": 0.10333282649517059,
      "num_tokens": 2312759.0,
      "step": 1140
    },
    {
      "entropy": 6.9618792057037355,
      "epoch": 0.10006117276937866,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999982102293153,
      "loss": 6.902,
      "mean_token_accuracy": 0.10509304627776146,
      "num_tokens": 2322016.0,
      "step": 1145
    },
    {
      "entropy": 6.956529331207276,
      "epoch": 0.10049812112208337,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999980837820578,
      "loss": 6.9316,
      "mean_token_accuracy": 0.10547636151313781,
      "num_tokens": 2332476.0,
      "step": 1150
    },
    {
      "entropy": 6.9808799743652346,
      "epoch": 0.10093506947478809,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999979530192192,
      "loss": 6.9422,
      "mean_token_accuracy": 0.09846181347966194,
      "num_tokens": 2342914.0,
      "step": 1155
    },
    {
      "entropy": 6.8698526382446286,
      "epoch": 0.10137201782749279,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999978179408019,
      "loss": 7.0014,
      "mean_token_accuracy": 0.09670989662408828,
      "num_tokens": 2353381.0,
      "step": 1160
    },
    {
      "entropy": 7.047989082336426,
      "epoch": 0.1018089661801975,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999976785468085,
      "loss": 6.9778,
      "mean_token_accuracy": 0.10065887346863747,
      "num_tokens": 2364207.0,
      "step": 1165
    },
    {
      "entropy": 6.905608797073365,
      "epoch": 0.10224591453290222,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999975348372418,
      "loss": 6.9048,
      "mean_token_accuracy": 0.10400475561618805,
      "num_tokens": 2374985.0,
      "step": 1170
    },
    {
      "entropy": 6.932749795913696,
      "epoch": 0.10268286288560692,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999973868121044,
      "loss": 6.9768,
      "mean_token_accuracy": 0.10295218899846077,
      "num_tokens": 2385684.0,
      "step": 1175
    },
    {
      "entropy": 7.058686876296997,
      "epoch": 0.10311981123831163,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999972344713994,
      "loss": 6.9109,
      "mean_token_accuracy": 0.10049357637763023,
      "num_tokens": 2395177.0,
      "step": 1180
    },
    {
      "entropy": 6.822031545639038,
      "epoch": 0.10355675959101635,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999970778151295,
      "loss": 6.9316,
      "mean_token_accuracy": 0.10832231491804123,
      "num_tokens": 2405154.0,
      "step": 1185
    },
    {
      "entropy": 6.980653572082519,
      "epoch": 0.10399370794372105,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999969168432976,
      "loss": 6.9107,
      "mean_token_accuracy": 0.1029654860496521,
      "num_tokens": 2415818.0,
      "step": 1190
    },
    {
      "entropy": 6.871922588348388,
      "epoch": 0.10443065629642576,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999967515559071,
      "loss": 6.8791,
      "mean_token_accuracy": 0.1029189296066761,
      "num_tokens": 2425072.0,
      "step": 1195
    },
    {
      "entropy": 6.889013576507568,
      "epoch": 0.10486760464913047,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499996581952961,
      "loss": 6.9411,
      "mean_token_accuracy": 0.10594153180718421,
      "num_tokens": 2435055.0,
      "step": 1200
    },
    {
      "entropy": 6.952979373931885,
      "epoch": 0.10530455300183518,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999964080344625,
      "loss": 6.8394,
      "mean_token_accuracy": 0.10662424638867378,
      "num_tokens": 2444434.0,
      "step": 1205
    },
    {
      "entropy": 6.836973428726196,
      "epoch": 0.1057415013545399,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999962298004151,
      "loss": 6.8662,
      "mean_token_accuracy": 0.10213011056184769,
      "num_tokens": 2453860.0,
      "step": 1210
    },
    {
      "entropy": 6.965761756896972,
      "epoch": 0.1061784497072446,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999960472508219,
      "loss": 6.985,
      "mean_token_accuracy": 0.10095644146203994,
      "num_tokens": 2464235.0,
      "step": 1215
    },
    {
      "entropy": 6.952071714401245,
      "epoch": 0.10661539805994931,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999958603856869,
      "loss": 6.8871,
      "mean_token_accuracy": 0.10538947805762292,
      "num_tokens": 2474561.0,
      "step": 1220
    },
    {
      "entropy": 6.8396659851074215,
      "epoch": 0.10705234641265403,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999956692050132,
      "loss": 6.8777,
      "mean_token_accuracy": 0.10440976917743683,
      "num_tokens": 2484574.0,
      "step": 1225
    },
    {
      "entropy": 6.999291944503784,
      "epoch": 0.10748929476535873,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999954737088048,
      "loss": 6.8676,
      "mean_token_accuracy": 0.110504861921072,
      "num_tokens": 2494146.0,
      "step": 1230
    },
    {
      "entropy": 6.806334829330444,
      "epoch": 0.10792624311806344,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999952738970652,
      "loss": 6.9511,
      "mean_token_accuracy": 0.10179314985871316,
      "num_tokens": 2504157.0,
      "step": 1235
    },
    {
      "entropy": 6.896573305130005,
      "epoch": 0.10836319147076816,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999950697697983,
      "loss": 6.8821,
      "mean_token_accuracy": 0.10277172699570655,
      "num_tokens": 2512979.0,
      "step": 1240
    },
    {
      "entropy": 6.921594572067261,
      "epoch": 0.10880013982347286,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999948613270081,
      "loss": 6.9518,
      "mean_token_accuracy": 0.09981353655457496,
      "num_tokens": 2523202.0,
      "step": 1245
    },
    {
      "entropy": 6.904537105560303,
      "epoch": 0.10923708817617758,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999946485686987,
      "loss": 6.8928,
      "mean_token_accuracy": 0.1013514257967472,
      "num_tokens": 2532790.0,
      "step": 1250
    },
    {
      "entropy": 6.879450464248658,
      "epoch": 0.10967403652888229,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499994431494874,
      "loss": 6.8721,
      "mean_token_accuracy": 0.10644448697566986,
      "num_tokens": 2542337.0,
      "step": 1255
    },
    {
      "entropy": 6.82066798210144,
      "epoch": 0.11011098488158699,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999942101055381,
      "loss": 6.8112,
      "mean_token_accuracy": 0.11042425706982613,
      "num_tokens": 2552762.0,
      "step": 1260
    },
    {
      "entropy": 6.91101107597351,
      "epoch": 0.11054793323429171,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999939844006955,
      "loss": 6.8508,
      "mean_token_accuracy": 0.10850365459918976,
      "num_tokens": 2563261.0,
      "step": 1265
    },
    {
      "entropy": 6.881007385253906,
      "epoch": 0.11098488158699642,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999937543803503,
      "loss": 6.8283,
      "mean_token_accuracy": 0.11036632806062699,
      "num_tokens": 2573089.0,
      "step": 1270
    },
    {
      "entropy": 6.852328968048096,
      "epoch": 0.11142182993970112,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499993520044507,
      "loss": 6.9193,
      "mean_token_accuracy": 0.09846171364188194,
      "num_tokens": 2584358.0,
      "step": 1275
    },
    {
      "entropy": 6.966355800628662,
      "epoch": 0.11185877829240584,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999932813931702,
      "loss": 6.8869,
      "mean_token_accuracy": 0.10662171319127083,
      "num_tokens": 2594981.0,
      "step": 1280
    },
    {
      "entropy": 6.8475768089294435,
      "epoch": 0.11229572664511055,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999930384263441,
      "loss": 6.8897,
      "mean_token_accuracy": 0.10364025831222534,
      "num_tokens": 2603508.0,
      "step": 1285
    },
    {
      "entropy": 6.908575344085693,
      "epoch": 0.11273267499781525,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499992791144034,
      "loss": 6.8442,
      "mean_token_accuracy": 0.10457368195056915,
      "num_tokens": 2613666.0,
      "step": 1290
    },
    {
      "entropy": 6.943562364578247,
      "epoch": 0.11316962335051997,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999925395462439,
      "loss": 6.9156,
      "mean_token_accuracy": 0.10008556246757508,
      "num_tokens": 2624441.0,
      "step": 1295
    },
    {
      "entropy": 6.885890769958496,
      "epoch": 0.11360657170322468,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999922836329793,
      "loss": 6.8662,
      "mean_token_accuracy": 0.10351174920797349,
      "num_tokens": 2635261.0,
      "step": 1300
    },
    {
      "entropy": 6.785940265655517,
      "epoch": 0.1140435200559294,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999920234042446,
      "loss": 6.8386,
      "mean_token_accuracy": 0.10835616141557694,
      "num_tokens": 2645493.0,
      "step": 1305
    },
    {
      "entropy": 6.885877275466919,
      "epoch": 0.1144804684086341,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999917588600451,
      "loss": 6.789,
      "mean_token_accuracy": 0.10901109129190445,
      "num_tokens": 2655430.0,
      "step": 1310
    },
    {
      "entropy": 6.838509273529053,
      "epoch": 0.1149174167613388,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999914900003857,
      "loss": 6.8596,
      "mean_token_accuracy": 0.10467283949255943,
      "num_tokens": 2665675.0,
      "step": 1315
    },
    {
      "entropy": 6.920448350906372,
      "epoch": 0.11535436511404352,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999912168252717,
      "loss": 6.8163,
      "mean_token_accuracy": 0.10676529109477997,
      "num_tokens": 2675962.0,
      "step": 1320
    },
    {
      "entropy": 6.824034214019775,
      "epoch": 0.11579131346674823,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999909393347083,
      "loss": 6.8647,
      "mean_token_accuracy": 0.10482599660754204,
      "num_tokens": 2687015.0,
      "step": 1325
    },
    {
      "entropy": 6.845885992050171,
      "epoch": 0.11622826181945294,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999906575287006,
      "loss": 6.9197,
      "mean_token_accuracy": 0.0967514120042324,
      "num_tokens": 2698065.0,
      "step": 1330
    },
    {
      "entropy": 6.878113698959351,
      "epoch": 0.11666521017215765,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999903714072544,
      "loss": 6.7906,
      "mean_token_accuracy": 0.11809995025396347,
      "num_tokens": 2709046.0,
      "step": 1335
    },
    {
      "entropy": 6.882908725738526,
      "epoch": 0.11710215852486236,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999900809703749,
      "loss": 6.883,
      "mean_token_accuracy": 0.10379704162478447,
      "num_tokens": 2719705.0,
      "step": 1340
    },
    {
      "entropy": 6.807536888122558,
      "epoch": 0.11753910687756706,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004999897862180678,
      "loss": 6.7552,
      "mean_token_accuracy": 0.11392875537276267,
      "num_tokens": 2729556.0,
      "step": 1345
    },
    {
      "entropy": 6.787845277786255,
      "epoch": 0.11797605523027178,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999894871503386,
      "loss": 6.8665,
      "mean_token_accuracy": 0.10998239666223526,
      "num_tokens": 2741144.0,
      "step": 1350
    },
    {
      "entropy": 6.9081967830657955,
      "epoch": 0.11841300358297649,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999891837671932,
      "loss": 6.7691,
      "mean_token_accuracy": 0.10925441533327103,
      "num_tokens": 2751384.0,
      "step": 1355
    },
    {
      "entropy": 6.789922571182251,
      "epoch": 0.11884995193568121,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999888760686375,
      "loss": 6.8099,
      "mean_token_accuracy": 0.10559153854846955,
      "num_tokens": 2761044.0,
      "step": 1360
    },
    {
      "entropy": 6.742934608459473,
      "epoch": 0.11928690028838591,
      "grad_norm": 0.890625,
      "learning_rate": 0.000499988564054677,
      "loss": 6.7547,
      "mean_token_accuracy": 0.10845592692494392,
      "num_tokens": 2770946.0,
      "step": 1365
    },
    {
      "entropy": 6.923913621902466,
      "epoch": 0.11972384864109062,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999882477253182,
      "loss": 6.8218,
      "mean_token_accuracy": 0.10306366384029389,
      "num_tokens": 2781247.0,
      "step": 1370
    },
    {
      "entropy": 6.7561554431915285,
      "epoch": 0.12016079699379534,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999879270805667,
      "loss": 6.6709,
      "mean_token_accuracy": 0.11467958092689515,
      "num_tokens": 2792078.0,
      "step": 1375
    },
    {
      "entropy": 6.782488059997559,
      "epoch": 0.12059774534650004,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999876021204291,
      "loss": 6.7598,
      "mean_token_accuracy": 0.11112385168671608,
      "num_tokens": 2802458.0,
      "step": 1380
    },
    {
      "entropy": 6.8228302001953125,
      "epoch": 0.12103469369920475,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999872728449112,
      "loss": 6.909,
      "mean_token_accuracy": 0.10087578147649764,
      "num_tokens": 2813352.0,
      "step": 1385
    },
    {
      "entropy": 6.8556358337402346,
      "epoch": 0.12147164205190947,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999869392540195,
      "loss": 6.7914,
      "mean_token_accuracy": 0.10912273600697517,
      "num_tokens": 2822954.0,
      "step": 1390
    },
    {
      "entropy": 6.758626174926758,
      "epoch": 0.12190859040461417,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999866013477605,
      "loss": 6.811,
      "mean_token_accuracy": 0.10802412927150726,
      "num_tokens": 2833375.0,
      "step": 1395
    },
    {
      "entropy": 6.806664562225341,
      "epoch": 0.12234553875731889,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999862591261407,
      "loss": 6.7713,
      "mean_token_accuracy": 0.10752915516495705,
      "num_tokens": 2843103.0,
      "step": 1400
    },
    {
      "entropy": 6.839266633987426,
      "epoch": 0.1227824871100236,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999859125891664,
      "loss": 6.8285,
      "mean_token_accuracy": 0.10644137039780617,
      "num_tokens": 2853801.0,
      "step": 1405
    },
    {
      "entropy": 6.855780410766601,
      "epoch": 0.1232194354627283,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999855617368445,
      "loss": 6.8144,
      "mean_token_accuracy": 0.10442313924431801,
      "num_tokens": 2863851.0,
      "step": 1410
    },
    {
      "entropy": 6.725363159179688,
      "epoch": 0.12365638381543302,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999852065691816,
      "loss": 6.7627,
      "mean_token_accuracy": 0.10936231687664985,
      "num_tokens": 2875073.0,
      "step": 1415
    },
    {
      "entropy": 6.893968534469605,
      "epoch": 0.12409333216813773,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999848470861846,
      "loss": 6.7854,
      "mean_token_accuracy": 0.10944370701909065,
      "num_tokens": 2885676.0,
      "step": 1420
    },
    {
      "entropy": 6.8066630363464355,
      "epoch": 0.12453028052084243,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999844832878602,
      "loss": 6.7683,
      "mean_token_accuracy": 0.11356849074363709,
      "num_tokens": 2896663.0,
      "step": 1425
    },
    {
      "entropy": 6.7106787204742435,
      "epoch": 0.12496722887354715,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999841151742156,
      "loss": 6.7929,
      "mean_token_accuracy": 0.10626860707998276,
      "num_tokens": 2907555.0,
      "step": 1430
    },
    {
      "entropy": 6.88362135887146,
      "epoch": 0.12540417722625186,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999837427452578,
      "loss": 6.7858,
      "mean_token_accuracy": 0.10847901776432992,
      "num_tokens": 2917173.0,
      "step": 1435
    },
    {
      "entropy": 6.720801258087159,
      "epoch": 0.12584112557895658,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499983366000994,
      "loss": 6.7811,
      "mean_token_accuracy": 0.1066952884197235,
      "num_tokens": 2928268.0,
      "step": 1440
    },
    {
      "entropy": 6.832601022720337,
      "epoch": 0.12627807393166127,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999829849414312,
      "loss": 6.725,
      "mean_token_accuracy": 0.11428027451038361,
      "num_tokens": 2938203.0,
      "step": 1445
    },
    {
      "entropy": 6.6934897899627686,
      "epoch": 0.12671502228436599,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499982599566577,
      "loss": 6.764,
      "mean_token_accuracy": 0.11221247315406799,
      "num_tokens": 2948094.0,
      "step": 1450
    },
    {
      "entropy": 6.872226905822754,
      "epoch": 0.1271519706370707,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999822098764385,
      "loss": 6.7674,
      "mean_token_accuracy": 0.11034838110208511,
      "num_tokens": 2957865.0,
      "step": 1455
    },
    {
      "entropy": 6.716366529464722,
      "epoch": 0.1275889189897754,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999818158710234,
      "loss": 6.6692,
      "mean_token_accuracy": 0.11622118279337883,
      "num_tokens": 2967880.0,
      "step": 1460
    },
    {
      "entropy": 6.719323587417603,
      "epoch": 0.12802586734248012,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999814175503392,
      "loss": 6.7614,
      "mean_token_accuracy": 0.11026630774140359,
      "num_tokens": 2979537.0,
      "step": 1465
    },
    {
      "entropy": 6.794031620025635,
      "epoch": 0.12846281569518483,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999810149143935,
      "loss": 6.7243,
      "mean_token_accuracy": 0.10677018910646438,
      "num_tokens": 2989700.0,
      "step": 1470
    },
    {
      "entropy": 6.648648405075074,
      "epoch": 0.12889976404788953,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499980607963194,
      "loss": 6.7292,
      "mean_token_accuracy": 0.11203910410404205,
      "num_tokens": 2999141.0,
      "step": 1475
    },
    {
      "entropy": 6.8095193862915036,
      "epoch": 0.12933671240059424,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999801966967486,
      "loss": 6.6825,
      "mean_token_accuracy": 0.11426105126738548,
      "num_tokens": 3008259.0,
      "step": 1480
    },
    {
      "entropy": 6.771524477005005,
      "epoch": 0.12977366075329896,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999797811150652,
      "loss": 6.7414,
      "mean_token_accuracy": 0.10887274071574211,
      "num_tokens": 3018281.0,
      "step": 1485
    },
    {
      "entropy": 6.8037293434143065,
      "epoch": 0.13021060910600368,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999793612181516,
      "loss": 6.7327,
      "mean_token_accuracy": 0.1142962783575058,
      "num_tokens": 3029062.0,
      "step": 1490
    },
    {
      "entropy": 6.670297622680664,
      "epoch": 0.13064755745870837,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999789370060161,
      "loss": 6.7104,
      "mean_token_accuracy": 0.1160611629486084,
      "num_tokens": 3038142.0,
      "step": 1495
    },
    {
      "entropy": 6.867388343811035,
      "epoch": 0.1310845058114131,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999785084786667,
      "loss": 6.8046,
      "mean_token_accuracy": 0.10011466220021248,
      "num_tokens": 3049123.0,
      "step": 1500
    },
    {
      "entropy": 6.710518217086792,
      "epoch": 0.1315214541641178,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999780756361114,
      "loss": 6.7168,
      "mean_token_accuracy": 0.10940457209944725,
      "num_tokens": 3059031.0,
      "step": 1505
    },
    {
      "entropy": 6.76721339225769,
      "epoch": 0.1319584025168225,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999776384783589,
      "loss": 6.7418,
      "mean_token_accuracy": 0.11014442816376686,
      "num_tokens": 3068830.0,
      "step": 1510
    },
    {
      "entropy": 6.657623338699341,
      "epoch": 0.13239535086952722,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999771970054174,
      "loss": 6.7262,
      "mean_token_accuracy": 0.11154688596725464,
      "num_tokens": 3080124.0,
      "step": 1515
    },
    {
      "entropy": 6.7857428073883055,
      "epoch": 0.13283229922223194,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999767512172953,
      "loss": 6.7318,
      "mean_token_accuracy": 0.10622003003954887,
      "num_tokens": 3089607.0,
      "step": 1520
    },
    {
      "entropy": 6.769694852828979,
      "epoch": 0.13326924757493663,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999763011140013,
      "loss": 6.8108,
      "mean_token_accuracy": 0.10131187736988068,
      "num_tokens": 3099363.0,
      "step": 1525
    },
    {
      "entropy": 6.8420090675354,
      "epoch": 0.13370619592764135,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999758466955439,
      "loss": 6.7243,
      "mean_token_accuracy": 0.11234695836901665,
      "num_tokens": 3109552.0,
      "step": 1530
    },
    {
      "entropy": 6.67326865196228,
      "epoch": 0.13414314428034607,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999753879619319,
      "loss": 6.6964,
      "mean_token_accuracy": 0.11057108193635941,
      "num_tokens": 3119536.0,
      "step": 1535
    },
    {
      "entropy": 6.774283361434937,
      "epoch": 0.13458009263305076,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999749249131741,
      "loss": 6.7401,
      "mean_token_accuracy": 0.11017183065414429,
      "num_tokens": 3130552.0,
      "step": 1540
    },
    {
      "entropy": 6.77594485282898,
      "epoch": 0.13501704098575548,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999744575492793,
      "loss": 6.6427,
      "mean_token_accuracy": 0.11230673864483834,
      "num_tokens": 3140723.0,
      "step": 1545
    },
    {
      "entropy": 6.734894800186157,
      "epoch": 0.1354539893384602,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999739858702567,
      "loss": 6.6867,
      "mean_token_accuracy": 0.11417788490653039,
      "num_tokens": 3151031.0,
      "step": 1550
    },
    {
      "entropy": 6.681126499176026,
      "epoch": 0.1358909376911649,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499973509876115,
      "loss": 6.6992,
      "mean_token_accuracy": 0.11300377771258355,
      "num_tokens": 3161373.0,
      "step": 1555
    },
    {
      "entropy": 6.732465410232544,
      "epoch": 0.1363278860438696,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999730295668634,
      "loss": 6.7103,
      "mean_token_accuracy": 0.11075188368558883,
      "num_tokens": 3171094.0,
      "step": 1560
    },
    {
      "entropy": 6.730536413192749,
      "epoch": 0.13676483439657433,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999725449425114,
      "loss": 6.6527,
      "mean_token_accuracy": 0.1153961479663849,
      "num_tokens": 3181646.0,
      "step": 1565
    },
    {
      "entropy": 6.669804573059082,
      "epoch": 0.13720178274927902,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499972056003068,
      "loss": 6.6392,
      "mean_token_accuracy": 0.12061795890331269,
      "num_tokens": 3190727.0,
      "step": 1570
    },
    {
      "entropy": 6.760992622375488,
      "epoch": 0.13763873110198374,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999715627485426,
      "loss": 6.7383,
      "mean_token_accuracy": 0.10512830913066865,
      "num_tokens": 3200316.0,
      "step": 1575
    },
    {
      "entropy": 6.72571120262146,
      "epoch": 0.13807567945468846,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499971065178945,
      "loss": 6.6925,
      "mean_token_accuracy": 0.10780871957540512,
      "num_tokens": 3211830.0,
      "step": 1580
    },
    {
      "entropy": 6.666191148757934,
      "epoch": 0.13851262780739318,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999705632942843,
      "loss": 6.7023,
      "mean_token_accuracy": 0.10752889290452003,
      "num_tokens": 3222711.0,
      "step": 1585
    },
    {
      "entropy": 6.766510057449341,
      "epoch": 0.13894957616009787,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999700570945704,
      "loss": 6.6191,
      "mean_token_accuracy": 0.11982477754354477,
      "num_tokens": 3232810.0,
      "step": 1590
    },
    {
      "entropy": 6.643466806411743,
      "epoch": 0.1393865245128026,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999695465798128,
      "loss": 6.709,
      "mean_token_accuracy": 0.10902300477027893,
      "num_tokens": 3243366.0,
      "step": 1595
    },
    {
      "entropy": 6.7340904712677006,
      "epoch": 0.1398234728655073,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999690317500215,
      "loss": 6.6817,
      "mean_token_accuracy": 0.11203390434384346,
      "num_tokens": 3254000.0,
      "step": 1600
    },
    {
      "entropy": 6.770872020721436,
      "epoch": 0.140260421218212,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999685126052063,
      "loss": 6.7182,
      "mean_token_accuracy": 0.11055498123168946,
      "num_tokens": 3263601.0,
      "step": 1605
    },
    {
      "entropy": 6.6786277294158936,
      "epoch": 0.14069736957091672,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999679891453772,
      "loss": 6.7518,
      "mean_token_accuracy": 0.10704593062400818,
      "num_tokens": 3273243.0,
      "step": 1610
    },
    {
      "entropy": 6.725984144210815,
      "epoch": 0.14113431792362144,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000499967461370544,
      "loss": 6.6946,
      "mean_token_accuracy": 0.11100869923830033,
      "num_tokens": 3284208.0,
      "step": 1615
    },
    {
      "entropy": 6.7146117210388185,
      "epoch": 0.14157126627632613,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999669292807172,
      "loss": 6.6585,
      "mean_token_accuracy": 0.11279672384262085,
      "num_tokens": 3294216.0,
      "step": 1620
    },
    {
      "entropy": 6.724279832839966,
      "epoch": 0.14200821462903085,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999663928759066,
      "loss": 6.7318,
      "mean_token_accuracy": 0.10651321336627007,
      "num_tokens": 3304576.0,
      "step": 1625
    },
    {
      "entropy": 6.742155408859253,
      "epoch": 0.14244516298173557,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999658521561229,
      "loss": 6.6799,
      "mean_token_accuracy": 0.10923416763544083,
      "num_tokens": 3314214.0,
      "step": 1630
    },
    {
      "entropy": 6.732270574569702,
      "epoch": 0.14288211133444026,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999653071213763,
      "loss": 6.6919,
      "mean_token_accuracy": 0.10857749730348587,
      "num_tokens": 3325527.0,
      "step": 1635
    },
    {
      "entropy": 6.634456062316895,
      "epoch": 0.14331905968714498,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999647577716771,
      "loss": 6.6186,
      "mean_token_accuracy": 0.11368991807103157,
      "num_tokens": 3334511.0,
      "step": 1640
    },
    {
      "entropy": 6.681470584869385,
      "epoch": 0.1437560080398497,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999642041070361,
      "loss": 6.6049,
      "mean_token_accuracy": 0.123313669860363,
      "num_tokens": 3343922.0,
      "step": 1645
    },
    {
      "entropy": 6.711009740829468,
      "epoch": 0.1441929563925544,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999636461274637,
      "loss": 6.6236,
      "mean_token_accuracy": 0.10806566402316094,
      "num_tokens": 3353895.0,
      "step": 1650
    },
    {
      "entropy": 6.664011001586914,
      "epoch": 0.1446299047452591,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999630838329706,
      "loss": 6.6989,
      "mean_token_accuracy": 0.10959775745868683,
      "num_tokens": 3364637.0,
      "step": 1655
    },
    {
      "entropy": 6.716980361938477,
      "epoch": 0.14506685309796383,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999625172235679,
      "loss": 6.6585,
      "mean_token_accuracy": 0.1092151403427124,
      "num_tokens": 3374308.0,
      "step": 1660
    },
    {
      "entropy": 6.650211811065674,
      "epoch": 0.14550380145066852,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499961946299266,
      "loss": 6.6329,
      "mean_token_accuracy": 0.11122519969940185,
      "num_tokens": 3383637.0,
      "step": 1665
    },
    {
      "entropy": 6.707240295410156,
      "epoch": 0.14594074980337324,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999613710600763,
      "loss": 6.7082,
      "mean_token_accuracy": 0.11246861442923546,
      "num_tokens": 3394150.0,
      "step": 1670
    },
    {
      "entropy": 6.7496161460876465,
      "epoch": 0.14637769815607796,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999607915060094,
      "loss": 6.6754,
      "mean_token_accuracy": 0.11157950460910797,
      "num_tokens": 3404445.0,
      "step": 1675
    },
    {
      "entropy": 6.595765924453735,
      "epoch": 0.14681464650878268,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999602076370768,
      "loss": 6.5737,
      "mean_token_accuracy": 0.12239945828914642,
      "num_tokens": 3415183.0,
      "step": 1680
    },
    {
      "entropy": 6.603966903686524,
      "epoch": 0.14725159486148737,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999596194532895,
      "loss": 6.5654,
      "mean_token_accuracy": 0.12099364176392555,
      "num_tokens": 3424824.0,
      "step": 1685
    },
    {
      "entropy": 6.689273071289063,
      "epoch": 0.1476885432141921,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999590269546588,
      "loss": 6.678,
      "mean_token_accuracy": 0.10761870443820953,
      "num_tokens": 3434789.0,
      "step": 1690
    },
    {
      "entropy": 6.553783273696899,
      "epoch": 0.1481254915668968,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999584301411961,
      "loss": 6.5636,
      "mean_token_accuracy": 0.1355303257703781,
      "num_tokens": 3445810.0,
      "step": 1695
    },
    {
      "entropy": 6.779795694351196,
      "epoch": 0.1485624399196015,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999578290129129,
      "loss": 6.6832,
      "mean_token_accuracy": 0.1057976596057415,
      "num_tokens": 3455510.0,
      "step": 1700
    },
    {
      "entropy": 6.636298894882202,
      "epoch": 0.14899938827230622,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999572235698206,
      "loss": 6.6615,
      "mean_token_accuracy": 0.1132010854780674,
      "num_tokens": 3465785.0,
      "step": 1705
    },
    {
      "entropy": 6.681069374084473,
      "epoch": 0.14943633662501093,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999566138119308,
      "loss": 6.7201,
      "mean_token_accuracy": 0.1101304367184639,
      "num_tokens": 3477568.0,
      "step": 1710
    },
    {
      "entropy": 6.6410560131073,
      "epoch": 0.14987328497771563,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999559997392554,
      "loss": 6.5654,
      "mean_token_accuracy": 0.12847595363855363,
      "num_tokens": 3487703.0,
      "step": 1715
    },
    {
      "entropy": 6.643354415893555,
      "epoch": 0.15031023333042035,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999553813518061,
      "loss": 6.6404,
      "mean_token_accuracy": 0.113068488240242,
      "num_tokens": 3497757.0,
      "step": 1720
    },
    {
      "entropy": 6.62676682472229,
      "epoch": 0.15074718168312506,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999547586495946,
      "loss": 6.5846,
      "mean_token_accuracy": 0.11577531546354294,
      "num_tokens": 3507594.0,
      "step": 1725
    },
    {
      "entropy": 6.71900110244751,
      "epoch": 0.15118413003582976,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999541316326329,
      "loss": 6.6673,
      "mean_token_accuracy": 0.11186658293008804,
      "num_tokens": 3517889.0,
      "step": 1730
    },
    {
      "entropy": 6.641316223144531,
      "epoch": 0.15162107838853447,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999535003009332,
      "loss": 6.654,
      "mean_token_accuracy": 0.11276111975312234,
      "num_tokens": 3528158.0,
      "step": 1735
    },
    {
      "entropy": 6.716168689727783,
      "epoch": 0.1520580267412392,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999528646545074,
      "loss": 6.6543,
      "mean_token_accuracy": 0.11286873891949653,
      "num_tokens": 3539099.0,
      "step": 1740
    },
    {
      "entropy": 6.686854839324951,
      "epoch": 0.15249497509394389,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999522246933678,
      "loss": 6.6539,
      "mean_token_accuracy": 0.11401691287755966,
      "num_tokens": 3549819.0,
      "step": 1745
    },
    {
      "entropy": 6.659570646286011,
      "epoch": 0.1529319234466486,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999515804175268,
      "loss": 6.6727,
      "mean_token_accuracy": 0.11552643105387687,
      "num_tokens": 3560022.0,
      "step": 1750
    },
    {
      "entropy": 6.690343952178955,
      "epoch": 0.15336887179935332,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999509318269964,
      "loss": 6.6105,
      "mean_token_accuracy": 0.1143421933054924,
      "num_tokens": 3569681.0,
      "step": 1755
    },
    {
      "entropy": 6.675909900665284,
      "epoch": 0.15380582015205801,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999502789217894,
      "loss": 6.6883,
      "mean_token_accuracy": 0.11230143532156944,
      "num_tokens": 3580731.0,
      "step": 1760
    },
    {
      "entropy": 6.674851751327514,
      "epoch": 0.15424276850476273,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999496217019181,
      "loss": 6.6569,
      "mean_token_accuracy": 0.11778324693441392,
      "num_tokens": 3590482.0,
      "step": 1765
    },
    {
      "entropy": 6.62106499671936,
      "epoch": 0.15467971685746745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999489601673952,
      "loss": 6.6372,
      "mean_token_accuracy": 0.10963449701666832,
      "num_tokens": 3600988.0,
      "step": 1770
    },
    {
      "entropy": 6.827673149108887,
      "epoch": 0.15511666521017214,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999482943182334,
      "loss": 6.7154,
      "mean_token_accuracy": 0.10713721886277199,
      "num_tokens": 3611226.0,
      "step": 1775
    },
    {
      "entropy": 6.550780200958252,
      "epoch": 0.15555361356287686,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999476241544454,
      "loss": 6.631,
      "mean_token_accuracy": 0.11362414360046387,
      "num_tokens": 3621324.0,
      "step": 1780
    },
    {
      "entropy": 6.73470458984375,
      "epoch": 0.15599056191558158,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999469496760441,
      "loss": 6.5608,
      "mean_token_accuracy": 0.10904188230633735,
      "num_tokens": 3631650.0,
      "step": 1785
    },
    {
      "entropy": 6.592597961425781,
      "epoch": 0.1564275102682863,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999462708830425,
      "loss": 6.579,
      "mean_token_accuracy": 0.1228872999548912,
      "num_tokens": 3641173.0,
      "step": 1790
    },
    {
      "entropy": 6.629960441589356,
      "epoch": 0.156864458620991,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999455877754535,
      "loss": 6.5856,
      "mean_token_accuracy": 0.11571357101202011,
      "num_tokens": 3651846.0,
      "step": 1795
    },
    {
      "entropy": 6.641428709030151,
      "epoch": 0.1573014069736957,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999449003532903,
      "loss": 6.5976,
      "mean_token_accuracy": 0.11091143637895584,
      "num_tokens": 3661174.0,
      "step": 1800
    },
    {
      "entropy": 6.59941554069519,
      "epoch": 0.15773835532640043,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999442086165661,
      "loss": 6.5618,
      "mean_token_accuracy": 0.11522511392831802,
      "num_tokens": 3671042.0,
      "step": 1805
    },
    {
      "entropy": 6.636082983016967,
      "epoch": 0.15817530367910512,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499943512565294,
      "loss": 6.5383,
      "mean_token_accuracy": 0.11486825719475746,
      "num_tokens": 3681206.0,
      "step": 1810
    },
    {
      "entropy": 6.601233339309692,
      "epoch": 0.15861225203180984,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999428121994874,
      "loss": 6.6437,
      "mean_token_accuracy": 0.11468076482415199,
      "num_tokens": 3691748.0,
      "step": 1815
    },
    {
      "entropy": 6.592959594726563,
      "epoch": 0.15904920038451456,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999421075191599,
      "loss": 6.5286,
      "mean_token_accuracy": 0.11978306695818901,
      "num_tokens": 3701730.0,
      "step": 1820
    },
    {
      "entropy": 6.626725864410401,
      "epoch": 0.15948614873721925,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999413985243249,
      "loss": 6.6348,
      "mean_token_accuracy": 0.11135040819644929,
      "num_tokens": 3712525.0,
      "step": 1825
    },
    {
      "entropy": 6.75208683013916,
      "epoch": 0.15992309708992397,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499940685214996,
      "loss": 6.6015,
      "mean_token_accuracy": 0.11548954620957375,
      "num_tokens": 3722016.0,
      "step": 1830
    },
    {
      "entropy": 6.588428592681884,
      "epoch": 0.1603600454426287,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999399675911869,
      "loss": 6.5126,
      "mean_token_accuracy": 0.11435215696692466,
      "num_tokens": 3731299.0,
      "step": 1835
    },
    {
      "entropy": 6.589688014984131,
      "epoch": 0.16079699379533338,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999392456529113,
      "loss": 6.587,
      "mean_token_accuracy": 0.11893209591507911,
      "num_tokens": 3740922.0,
      "step": 1840
    },
    {
      "entropy": 6.591811609268189,
      "epoch": 0.1612339421480381,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999385194001831,
      "loss": 6.5647,
      "mean_token_accuracy": 0.11818592995405197,
      "num_tokens": 3750339.0,
      "step": 1845
    },
    {
      "entropy": 6.626106214523316,
      "epoch": 0.16167089050074282,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999377888330164,
      "loss": 6.5444,
      "mean_token_accuracy": 0.11974458321928978,
      "num_tokens": 3761379.0,
      "step": 1850
    },
    {
      "entropy": 6.565863084793091,
      "epoch": 0.1621078388534475,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999370539514249,
      "loss": 6.6178,
      "mean_token_accuracy": 0.11443219929933549,
      "num_tokens": 3771649.0,
      "step": 1855
    },
    {
      "entropy": 6.616359758377075,
      "epoch": 0.16254478720615223,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999363147554229,
      "loss": 6.5317,
      "mean_token_accuracy": 0.11977923512458802,
      "num_tokens": 3782393.0,
      "step": 1860
    },
    {
      "entropy": 6.595934057235718,
      "epoch": 0.16298173555885695,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999355712450244,
      "loss": 6.5811,
      "mean_token_accuracy": 0.11747404783964158,
      "num_tokens": 3792849.0,
      "step": 1865
    },
    {
      "entropy": 6.613617277145385,
      "epoch": 0.16341868391156164,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999348234202438,
      "loss": 6.5427,
      "mean_token_accuracy": 0.12124217823147773,
      "num_tokens": 3803848.0,
      "step": 1870
    },
    {
      "entropy": 6.488130903244018,
      "epoch": 0.16385563226426636,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999340712810956,
      "loss": 6.4728,
      "mean_token_accuracy": 0.13200217932462693,
      "num_tokens": 3814704.0,
      "step": 1875
    },
    {
      "entropy": 6.589626598358154,
      "epoch": 0.16429258061697108,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999333148275939,
      "loss": 6.6042,
      "mean_token_accuracy": 0.11308980882167816,
      "num_tokens": 3824955.0,
      "step": 1880
    },
    {
      "entropy": 6.65059003829956,
      "epoch": 0.1647295289696758,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999325540597535,
      "loss": 6.5923,
      "mean_token_accuracy": 0.11620239540934563,
      "num_tokens": 3836540.0,
      "step": 1885
    },
    {
      "entropy": 6.508061504364013,
      "epoch": 0.1651664773223805,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999317889775887,
      "loss": 6.472,
      "mean_token_accuracy": 0.127332104742527,
      "num_tokens": 3846472.0,
      "step": 1890
    },
    {
      "entropy": 6.606565713882446,
      "epoch": 0.1656034256750852,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999310195811143,
      "loss": 6.5876,
      "mean_token_accuracy": 0.11355203092098236,
      "num_tokens": 3857187.0,
      "step": 1895
    },
    {
      "entropy": 6.635375547409057,
      "epoch": 0.16604037402778993,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999302458703451,
      "loss": 6.5539,
      "mean_token_accuracy": 0.12006403431296349,
      "num_tokens": 3866259.0,
      "step": 1900
    },
    {
      "entropy": 6.508198642730713,
      "epoch": 0.16647732238049462,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499929467845296,
      "loss": 6.6008,
      "mean_token_accuracy": 0.11160141155123711,
      "num_tokens": 3876869.0,
      "step": 1905
    },
    {
      "entropy": 6.69843864440918,
      "epoch": 0.16691427073319934,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999286855059818,
      "loss": 6.5887,
      "mean_token_accuracy": 0.11653141602873802,
      "num_tokens": 3888278.0,
      "step": 1910
    },
    {
      "entropy": 6.593745183944702,
      "epoch": 0.16735121908590406,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999278988524177,
      "loss": 6.6379,
      "mean_token_accuracy": 0.11633663773536682,
      "num_tokens": 3898648.0,
      "step": 1915
    },
    {
      "entropy": 6.625266742706299,
      "epoch": 0.16778816743860875,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999271078846184,
      "loss": 6.5605,
      "mean_token_accuracy": 0.11227926462888718,
      "num_tokens": 3908282.0,
      "step": 1920
    },
    {
      "entropy": 6.506586313247681,
      "epoch": 0.16822511579131347,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999263126025994,
      "loss": 6.5511,
      "mean_token_accuracy": 0.11529682129621506,
      "num_tokens": 3918333.0,
      "step": 1925
    },
    {
      "entropy": 6.6185715675354,
      "epoch": 0.1686620641440182,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999255130063759,
      "loss": 6.551,
      "mean_token_accuracy": 0.11844497621059417,
      "num_tokens": 3927403.0,
      "step": 1930
    },
    {
      "entropy": 6.623962688446045,
      "epoch": 0.16909901249672288,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999247090959633,
      "loss": 6.5766,
      "mean_token_accuracy": 0.11571719124913216,
      "num_tokens": 3937352.0,
      "step": 1935
    },
    {
      "entropy": 6.5212033748626705,
      "epoch": 0.1695359608494276,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999239008713767,
      "loss": 6.4001,
      "mean_token_accuracy": 0.12067319750785828,
      "num_tokens": 3947164.0,
      "step": 1940
    },
    {
      "entropy": 6.612851142883301,
      "epoch": 0.16997290920213232,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999230883326319,
      "loss": 6.5711,
      "mean_token_accuracy": 0.1215322732925415,
      "num_tokens": 3956887.0,
      "step": 1945
    },
    {
      "entropy": 6.603779935836792,
      "epoch": 0.170409857554837,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999222714797445,
      "loss": 6.5448,
      "mean_token_accuracy": 0.11428779810667038,
      "num_tokens": 3966542.0,
      "step": 1950
    },
    {
      "entropy": 6.621708679199219,
      "epoch": 0.17084680590754173,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049992145031273,
      "loss": 6.5031,
      "mean_token_accuracy": 0.12730235159397124,
      "num_tokens": 3976127.0,
      "step": 1955
    },
    {
      "entropy": 6.489267015457154,
      "epoch": 0.17128375426024645,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999206248316043,
      "loss": 6.4794,
      "mean_token_accuracy": 0.12245322167873382,
      "num_tokens": 3987371.0,
      "step": 1960
    },
    {
      "entropy": 6.467897510528564,
      "epoch": 0.17172070261295114,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499919795036383,
      "loss": 6.4856,
      "mean_token_accuracy": 0.12608123421669007,
      "num_tokens": 3997924.0,
      "step": 1965
    },
    {
      "entropy": 6.592717838287354,
      "epoch": 0.17215765096565586,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999189609270823,
      "loss": 6.5879,
      "mean_token_accuracy": 0.11945871189236641,
      "num_tokens": 4009150.0,
      "step": 1970
    },
    {
      "entropy": 6.613911104202271,
      "epoch": 0.17259459931836058,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499918122503718,
      "loss": 6.5901,
      "mean_token_accuracy": 0.11914976313710213,
      "num_tokens": 4018899.0,
      "step": 1975
    },
    {
      "entropy": 6.609164476394653,
      "epoch": 0.17303154767106527,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999172797663064,
      "loss": 6.5304,
      "mean_token_accuracy": 0.11551554650068283,
      "num_tokens": 4028652.0,
      "step": 1980
    },
    {
      "entropy": 6.543050098419189,
      "epoch": 0.17346849602376999,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999164327148634,
      "loss": 6.5897,
      "mean_token_accuracy": 0.11552718579769135,
      "num_tokens": 4039819.0,
      "step": 1985
    },
    {
      "entropy": 6.634558057785034,
      "epoch": 0.1739054443764747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999155813494053,
      "loss": 6.4774,
      "mean_token_accuracy": 0.1238390825688839,
      "num_tokens": 4050029.0,
      "step": 1990
    },
    {
      "entropy": 6.476582670211792,
      "epoch": 0.17434239272917942,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999147256699487,
      "loss": 6.5391,
      "mean_token_accuracy": 0.11595378592610359,
      "num_tokens": 4060497.0,
      "step": 1995
    },
    {
      "entropy": 6.684065818786621,
      "epoch": 0.17477934108188412,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999138656765096,
      "loss": 6.4877,
      "mean_token_accuracy": 0.12127815932035446,
      "num_tokens": 4071477.0,
      "step": 2000
    },
    {
      "entropy": 6.5196648120880125,
      "epoch": 0.17521628943458883,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999130013691049,
      "loss": 6.6098,
      "mean_token_accuracy": 0.11518096625804901,
      "num_tokens": 4081869.0,
      "step": 2005
    },
    {
      "entropy": 6.655038213729858,
      "epoch": 0.17565323778729355,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999121327477509,
      "loss": 6.5682,
      "mean_token_accuracy": 0.12180981114506721,
      "num_tokens": 4092797.0,
      "step": 2010
    },
    {
      "entropy": 6.532944250106811,
      "epoch": 0.17609018613999824,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999112598124642,
      "loss": 6.5063,
      "mean_token_accuracy": 0.1199454739689827,
      "num_tokens": 4103688.0,
      "step": 2015
    },
    {
      "entropy": 6.571459817886352,
      "epoch": 0.17652713449270296,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999103825632618,
      "loss": 6.4782,
      "mean_token_accuracy": 0.11821270734071732,
      "num_tokens": 4114037.0,
      "step": 2020
    },
    {
      "entropy": 6.5301916122436525,
      "epoch": 0.17696408284540768,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999095010001605,
      "loss": 6.4682,
      "mean_token_accuracy": 0.1155319221317768,
      "num_tokens": 4124164.0,
      "step": 2025
    },
    {
      "entropy": 6.594848585128784,
      "epoch": 0.17740103119811237,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499908615123177,
      "loss": 6.5857,
      "mean_token_accuracy": 0.11203168854117393,
      "num_tokens": 4134225.0,
      "step": 2030
    },
    {
      "entropy": 6.591685009002686,
      "epoch": 0.1778379795508171,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999077249323284,
      "loss": 6.6335,
      "mean_token_accuracy": 0.11418751999735832,
      "num_tokens": 4145962.0,
      "step": 2035
    },
    {
      "entropy": 6.635752868652344,
      "epoch": 0.1782749279035218,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999068304276319,
      "loss": 6.5198,
      "mean_token_accuracy": 0.11557328552007676,
      "num_tokens": 4156572.0,
      "step": 2040
    },
    {
      "entropy": 6.526685047149658,
      "epoch": 0.1787118762562265,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999059316091045,
      "loss": 6.5968,
      "mean_token_accuracy": 0.11482413336634636,
      "num_tokens": 4166960.0,
      "step": 2045
    },
    {
      "entropy": 6.634673261642456,
      "epoch": 0.17914882460893122,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999050284767634,
      "loss": 6.5518,
      "mean_token_accuracy": 0.11969516798853874,
      "num_tokens": 4177833.0,
      "step": 2050
    },
    {
      "entropy": 6.4771088600158695,
      "epoch": 0.17958577296163594,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999041210306261,
      "loss": 6.476,
      "mean_token_accuracy": 0.12267953082919121,
      "num_tokens": 4187858.0,
      "step": 2055
    },
    {
      "entropy": 6.624364042282105,
      "epoch": 0.18002272131434063,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999032092707098,
      "loss": 6.597,
      "mean_token_accuracy": 0.11280876621603966,
      "num_tokens": 4198207.0,
      "step": 2060
    },
    {
      "entropy": 6.574216651916504,
      "epoch": 0.18045966966704535,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999022931970323,
      "loss": 6.4885,
      "mean_token_accuracy": 0.12329766750335694,
      "num_tokens": 4207594.0,
      "step": 2065
    },
    {
      "entropy": 6.514097881317139,
      "epoch": 0.18089661801975007,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999013728096107,
      "loss": 6.4908,
      "mean_token_accuracy": 0.12111081257462501,
      "num_tokens": 4217334.0,
      "step": 2070
    },
    {
      "entropy": 6.534528684616089,
      "epoch": 0.18133356637245476,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499900448108463,
      "loss": 6.462,
      "mean_token_accuracy": 0.12237232476472855,
      "num_tokens": 4226407.0,
      "step": 2075
    },
    {
      "entropy": 6.516916418075562,
      "epoch": 0.18177051472515948,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499899519093607,
      "loss": 6.4228,
      "mean_token_accuracy": 0.1265766814351082,
      "num_tokens": 4237818.0,
      "step": 2080
    },
    {
      "entropy": 6.613302946090698,
      "epoch": 0.1822074630778642,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004998985857650603,
      "loss": 6.5469,
      "mean_token_accuracy": 0.11210652962327003,
      "num_tokens": 4247783.0,
      "step": 2085
    },
    {
      "entropy": 6.468383121490478,
      "epoch": 0.18264441143056892,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998976481228409,
      "loss": 6.4822,
      "mean_token_accuracy": 0.12208920419216156,
      "num_tokens": 4258049.0,
      "step": 2090
    },
    {
      "entropy": 6.525755500793457,
      "epoch": 0.1830813597832736,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998967061669668,
      "loss": 6.4925,
      "mean_token_accuracy": 0.11374819204211235,
      "num_tokens": 4269357.0,
      "step": 2095
    },
    {
      "entropy": 6.536129570007324,
      "epoch": 0.18351830813597833,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998957598974559,
      "loss": 6.4138,
      "mean_token_accuracy": 0.12686009854078292,
      "num_tokens": 4279576.0,
      "step": 2100
    },
    {
      "entropy": 6.499485492706299,
      "epoch": 0.18395525648868305,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998948093143266,
      "loss": 6.4758,
      "mean_token_accuracy": 0.11781678795814514,
      "num_tokens": 4289761.0,
      "step": 2105
    },
    {
      "entropy": 6.524201726913452,
      "epoch": 0.18439220484138774,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499893854417597,
      "loss": 6.3922,
      "mean_token_accuracy": 0.12270399779081345,
      "num_tokens": 4299231.0,
      "step": 2110
    },
    {
      "entropy": 6.541791009902954,
      "epoch": 0.18482915319409246,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998928952072854,
      "loss": 6.5593,
      "mean_token_accuracy": 0.11242108196020126,
      "num_tokens": 4308687.0,
      "step": 2115
    },
    {
      "entropy": 6.591986656188965,
      "epoch": 0.18526610154679718,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998919316834102,
      "loss": 6.5182,
      "mean_token_accuracy": 0.11508347168564796,
      "num_tokens": 4318183.0,
      "step": 2120
    },
    {
      "entropy": 6.558227300643921,
      "epoch": 0.18570304989950187,
      "grad_norm": 1.125,
      "learning_rate": 0.00049989096384599,
      "loss": 6.5573,
      "mean_token_accuracy": 0.11798967346549034,
      "num_tokens": 4328482.0,
      "step": 2125
    },
    {
      "entropy": 6.584009790420533,
      "epoch": 0.1861399982522066,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004998899916950431,
      "loss": 6.4366,
      "mean_token_accuracy": 0.1197984792292118,
      "num_tokens": 4339313.0,
      "step": 2130
    },
    {
      "entropy": 6.453553342819214,
      "epoch": 0.1865769466049113,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998890152305885,
      "loss": 6.5547,
      "mean_token_accuracy": 0.11854925975203515,
      "num_tokens": 4350299.0,
      "step": 2135
    },
    {
      "entropy": 6.635061883926392,
      "epoch": 0.187013894957616,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004998880344526447,
      "loss": 6.4894,
      "mean_token_accuracy": 0.12394139692187309,
      "num_tokens": 4361424.0,
      "step": 2140
    },
    {
      "entropy": 6.466842317581177,
      "epoch": 0.18745084331032072,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998870493612306,
      "loss": 6.4538,
      "mean_token_accuracy": 0.12028718739748001,
      "num_tokens": 4371480.0,
      "step": 2145
    },
    {
      "entropy": 6.470487928390503,
      "epoch": 0.18788779166302544,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998860599563649,
      "loss": 6.4967,
      "mean_token_accuracy": 0.12331848666071892,
      "num_tokens": 4381951.0,
      "step": 2150
    },
    {
      "entropy": 6.558203125,
      "epoch": 0.18832474001573013,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998850662380668,
      "loss": 6.5616,
      "mean_token_accuracy": 0.11244588792324066,
      "num_tokens": 4392290.0,
      "step": 2155
    },
    {
      "entropy": 6.525679016113282,
      "epoch": 0.18876168836843485,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998840682063553,
      "loss": 6.2939,
      "mean_token_accuracy": 0.13558808639645575,
      "num_tokens": 4403350.0,
      "step": 2160
    },
    {
      "entropy": 6.414237642288208,
      "epoch": 0.18919863672113957,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998830658612496,
      "loss": 6.4999,
      "mean_token_accuracy": 0.11366144940257072,
      "num_tokens": 4413171.0,
      "step": 2165
    },
    {
      "entropy": 6.537461614608764,
      "epoch": 0.18963558507384426,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998820592027688,
      "loss": 6.4523,
      "mean_token_accuracy": 0.12490663677453995,
      "num_tokens": 4424371.0,
      "step": 2170
    },
    {
      "entropy": 6.459207391738891,
      "epoch": 0.19007253342654898,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998810482309322,
      "loss": 6.415,
      "mean_token_accuracy": 0.11684294268488885,
      "num_tokens": 4434040.0,
      "step": 2175
    },
    {
      "entropy": 6.612886381149292,
      "epoch": 0.1905094817792537,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998800329457595,
      "loss": 6.5237,
      "mean_token_accuracy": 0.11428215354681015,
      "num_tokens": 4444663.0,
      "step": 2180
    },
    {
      "entropy": 6.530249738693238,
      "epoch": 0.1909464301319584,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998790133472697,
      "loss": 6.4732,
      "mean_token_accuracy": 0.1187725804746151,
      "num_tokens": 4454733.0,
      "step": 2185
    },
    {
      "entropy": 6.478503894805908,
      "epoch": 0.1913833784846631,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998779894354827,
      "loss": 6.4561,
      "mean_token_accuracy": 0.12121548429131508,
      "num_tokens": 4464286.0,
      "step": 2190
    },
    {
      "entropy": 6.554786252975464,
      "epoch": 0.19182032683736783,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499876961210418,
      "loss": 6.4435,
      "mean_token_accuracy": 0.12957765012979508,
      "num_tokens": 4475005.0,
      "step": 2195
    },
    {
      "entropy": 6.456162691116333,
      "epoch": 0.19225727519007255,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998759286720953,
      "loss": 6.4038,
      "mean_token_accuracy": 0.12335258722305298,
      "num_tokens": 4484606.0,
      "step": 2200
    },
    {
      "entropy": 6.4117199897766115,
      "epoch": 0.19269422354277724,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998748918205345,
      "loss": 6.5107,
      "mean_token_accuracy": 0.115721445530653,
      "num_tokens": 4494930.0,
      "step": 2205
    },
    {
      "entropy": 6.56764326095581,
      "epoch": 0.19313117189548196,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998738506557554,
      "loss": 6.5316,
      "mean_token_accuracy": 0.11446557492017746,
      "num_tokens": 4505304.0,
      "step": 2210
    },
    {
      "entropy": 6.6263429641723635,
      "epoch": 0.19356812024818668,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499872805177778,
      "loss": 6.4505,
      "mean_token_accuracy": 0.1278298854827881,
      "num_tokens": 4514333.0,
      "step": 2215
    },
    {
      "entropy": 6.442032718658448,
      "epoch": 0.19400506860089137,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998717553866225,
      "loss": 6.5229,
      "mean_token_accuracy": 0.12303071394562722,
      "num_tokens": 4525175.0,
      "step": 2220
    },
    {
      "entropy": 6.5615222454071045,
      "epoch": 0.1944420169535961,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998707012823087,
      "loss": 6.3765,
      "mean_token_accuracy": 0.12679129466414452,
      "num_tokens": 4534212.0,
      "step": 2225
    },
    {
      "entropy": 6.452094221115113,
      "epoch": 0.1948789653063008,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998696428648573,
      "loss": 6.5735,
      "mean_token_accuracy": 0.11512410044670104,
      "num_tokens": 4545141.0,
      "step": 2230
    },
    {
      "entropy": 6.557396984100341,
      "epoch": 0.1953159136590055,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998685801342881,
      "loss": 6.4302,
      "mean_token_accuracy": 0.12757914289832115,
      "num_tokens": 4554410.0,
      "step": 2235
    },
    {
      "entropy": 6.421574354171753,
      "epoch": 0.19575286201171022,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998675130906217,
      "loss": 6.4167,
      "mean_token_accuracy": 0.12503303810954094,
      "num_tokens": 4564397.0,
      "step": 2240
    },
    {
      "entropy": 6.4371904850006105,
      "epoch": 0.19618981036441493,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998664417338787,
      "loss": 6.4093,
      "mean_token_accuracy": 0.13016692399978638,
      "num_tokens": 4573986.0,
      "step": 2245
    },
    {
      "entropy": 6.574130344390869,
      "epoch": 0.19662675871711963,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004998653660640794,
      "loss": 6.4194,
      "mean_token_accuracy": 0.1290357455611229,
      "num_tokens": 4583449.0,
      "step": 2250
    },
    {
      "entropy": 6.453009510040284,
      "epoch": 0.19706370706982435,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998642860812447,
      "loss": 6.4579,
      "mean_token_accuracy": 0.12105687931180001,
      "num_tokens": 4593637.0,
      "step": 2255
    },
    {
      "entropy": 6.440600728988647,
      "epoch": 0.19750065542252906,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004998632017853951,
      "loss": 6.439,
      "mean_token_accuracy": 0.12805744558572768,
      "num_tokens": 4603865.0,
      "step": 2260
    },
    {
      "entropy": 6.638007259368896,
      "epoch": 0.19793760377523376,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998621131765514,
      "loss": 6.4676,
      "mean_token_accuracy": 0.12696166336536407,
      "num_tokens": 4613502.0,
      "step": 2265
    },
    {
      "entropy": 6.4463948726654055,
      "epoch": 0.19837455212793847,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998610202547345,
      "loss": 6.4225,
      "mean_token_accuracy": 0.12365630269050598,
      "num_tokens": 4623163.0,
      "step": 2270
    },
    {
      "entropy": 6.512554073333741,
      "epoch": 0.1988115004806432,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998599230199654,
      "loss": 6.4507,
      "mean_token_accuracy": 0.12848633974790574,
      "num_tokens": 4632850.0,
      "step": 2275
    },
    {
      "entropy": 6.491110992431641,
      "epoch": 0.19924844883334789,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998588214722653,
      "loss": 6.4608,
      "mean_token_accuracy": 0.12021621614694596,
      "num_tokens": 4642858.0,
      "step": 2280
    },
    {
      "entropy": 6.567204236984253,
      "epoch": 0.1996853971860526,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998577156116551,
      "loss": 6.4891,
      "mean_token_accuracy": 0.11761298626661301,
      "num_tokens": 4653083.0,
      "step": 2285
    },
    {
      "entropy": 6.529744482040405,
      "epoch": 0.20012234553875732,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998566054381561,
      "loss": 6.4926,
      "mean_token_accuracy": 0.11979148462414742,
      "num_tokens": 4663353.0,
      "step": 2290
    },
    {
      "entropy": 6.469255495071411,
      "epoch": 0.20055929389146204,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004998554909517895,
      "loss": 6.3575,
      "mean_token_accuracy": 0.12421375289559364,
      "num_tokens": 4673878.0,
      "step": 2295
    },
    {
      "entropy": 6.480479335784912,
      "epoch": 0.20099624224416673,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998543721525769,
      "loss": 6.4255,
      "mean_token_accuracy": 0.12483477517962456,
      "num_tokens": 4683661.0,
      "step": 2300
    },
    {
      "entropy": 6.437173128128052,
      "epoch": 0.20143319059687145,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004998532490405397,
      "loss": 6.4822,
      "mean_token_accuracy": 0.12061880454421044,
      "num_tokens": 4693817.0,
      "step": 2305
    },
    {
      "entropy": 6.519210433959961,
      "epoch": 0.20187013894957617,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998521216156991,
      "loss": 6.4082,
      "mean_token_accuracy": 0.12354693114757538,
      "num_tokens": 4703550.0,
      "step": 2310
    },
    {
      "entropy": 6.495586681365967,
      "epoch": 0.20230708730228086,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998509898780771,
      "loss": 6.5167,
      "mean_token_accuracy": 0.12095422595739365,
      "num_tokens": 4714872.0,
      "step": 2315
    },
    {
      "entropy": 6.496729040145874,
      "epoch": 0.20274403565498558,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998498538276954,
      "loss": 6.4195,
      "mean_token_accuracy": 0.1262986145913601,
      "num_tokens": 4725745.0,
      "step": 2320
    },
    {
      "entropy": 6.488230037689209,
      "epoch": 0.2031809840076903,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998487134645755,
      "loss": 6.5061,
      "mean_token_accuracy": 0.11861919611692429,
      "num_tokens": 4735560.0,
      "step": 2325
    },
    {
      "entropy": 6.54272050857544,
      "epoch": 0.203617932360395,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998475687887396,
      "loss": 6.4698,
      "mean_token_accuracy": 0.11695239022374153,
      "num_tokens": 4746304.0,
      "step": 2330
    },
    {
      "entropy": 6.514755916595459,
      "epoch": 0.2040548807130997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998464198002095,
      "loss": 6.4377,
      "mean_token_accuracy": 0.1227838471531868,
      "num_tokens": 4756200.0,
      "step": 2335
    },
    {
      "entropy": 6.447860813140869,
      "epoch": 0.20449182906580443,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998452664990072,
      "loss": 6.4622,
      "mean_token_accuracy": 0.11562853530049325,
      "num_tokens": 4766249.0,
      "step": 2340
    },
    {
      "entropy": 6.608063697814941,
      "epoch": 0.20492877741850912,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998441088851548,
      "loss": 6.4832,
      "mean_token_accuracy": 0.11795579493045807,
      "num_tokens": 4776510.0,
      "step": 2345
    },
    {
      "entropy": 6.442056512832641,
      "epoch": 0.20536572577121384,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998429469586748,
      "loss": 6.4677,
      "mean_token_accuracy": 0.12453974261879922,
      "num_tokens": 4786798.0,
      "step": 2350
    },
    {
      "entropy": 6.467907285690307,
      "epoch": 0.20580267412391856,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499841780719589,
      "loss": 6.3917,
      "mean_token_accuracy": 0.12231776416301728,
      "num_tokens": 4796001.0,
      "step": 2355
    },
    {
      "entropy": 6.50298957824707,
      "epoch": 0.20623962247662325,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004998406101679202,
      "loss": 6.3698,
      "mean_token_accuracy": 0.13163737431168557,
      "num_tokens": 4807634.0,
      "step": 2360
    },
    {
      "entropy": 6.424554872512817,
      "epoch": 0.20667657082932797,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998394353036906,
      "loss": 6.3798,
      "mean_token_accuracy": 0.12814454063773156,
      "num_tokens": 4817840.0,
      "step": 2365
    },
    {
      "entropy": 6.408347225189209,
      "epoch": 0.2071135191820327,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998382561269229,
      "loss": 6.361,
      "mean_token_accuracy": 0.12966743409633635,
      "num_tokens": 4827186.0,
      "step": 2370
    },
    {
      "entropy": 6.389866161346435,
      "epoch": 0.20755046753473738,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998370726376395,
      "loss": 6.4079,
      "mean_token_accuracy": 0.122555311024189,
      "num_tokens": 4837140.0,
      "step": 2375
    },
    {
      "entropy": 6.533910131454467,
      "epoch": 0.2079874158874421,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998358848358634,
      "loss": 6.431,
      "mean_token_accuracy": 0.12573982998728753,
      "num_tokens": 4846907.0,
      "step": 2380
    },
    {
      "entropy": 6.3821924209594725,
      "epoch": 0.20842436424014682,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499834692721617,
      "loss": 6.46,
      "mean_token_accuracy": 0.12661178037524223,
      "num_tokens": 4856165.0,
      "step": 2385
    },
    {
      "entropy": 6.529715347290039,
      "epoch": 0.2088613125928515,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998334962949237,
      "loss": 6.3174,
      "mean_token_accuracy": 0.13617229014635085,
      "num_tokens": 4865308.0,
      "step": 2390
    },
    {
      "entropy": 6.401957368850708,
      "epoch": 0.20929826094555623,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004998322955558059,
      "loss": 6.3776,
      "mean_token_accuracy": 0.13593824356794357,
      "num_tokens": 4876286.0,
      "step": 2395
    },
    {
      "entropy": 6.444190835952758,
      "epoch": 0.20973520929826095,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998310905042869,
      "loss": 6.4117,
      "mean_token_accuracy": 0.1278536468744278,
      "num_tokens": 4887227.0,
      "step": 2400
    },
    {
      "entropy": 6.500918197631836,
      "epoch": 0.21017215765096567,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998298811403898,
      "loss": 6.4362,
      "mean_token_accuracy": 0.12585031986236572,
      "num_tokens": 4896994.0,
      "step": 2405
    },
    {
      "entropy": 6.4714161396026615,
      "epoch": 0.21060910600367036,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998286674641378,
      "loss": 6.4233,
      "mean_token_accuracy": 0.12609584704041482,
      "num_tokens": 4907623.0,
      "step": 2410
    },
    {
      "entropy": 6.38978328704834,
      "epoch": 0.21104605435637508,
      "grad_norm": 1.125,
      "learning_rate": 0.000499827449475554,
      "loss": 6.3345,
      "mean_token_accuracy": 0.1257795587182045,
      "num_tokens": 4917040.0,
      "step": 2415
    },
    {
      "entropy": 6.572758913040161,
      "epoch": 0.2114830027090798,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499826227174662,
      "loss": 6.4813,
      "mean_token_accuracy": 0.12101781442761421,
      "num_tokens": 4928390.0,
      "step": 2420
    },
    {
      "entropy": 6.456791114807129,
      "epoch": 0.2119199510617845,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998250005614852,
      "loss": 6.4405,
      "mean_token_accuracy": 0.12130651250481606,
      "num_tokens": 4939040.0,
      "step": 2425
    },
    {
      "entropy": 6.4252156734466555,
      "epoch": 0.2123568994144892,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000499823769636047,
      "loss": 6.4799,
      "mean_token_accuracy": 0.1219657190144062,
      "num_tokens": 4949915.0,
      "step": 2430
    },
    {
      "entropy": 6.486149072647095,
      "epoch": 0.21279384776719393,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998225343983712,
      "loss": 6.3746,
      "mean_token_accuracy": 0.1280059665441513,
      "num_tokens": 4960127.0,
      "step": 2435
    },
    {
      "entropy": 6.488802576065064,
      "epoch": 0.21323079611989862,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998212948484814,
      "loss": 6.4092,
      "mean_token_accuracy": 0.12064960524439812,
      "num_tokens": 4970152.0,
      "step": 2440
    },
    {
      "entropy": 6.453671264648437,
      "epoch": 0.21366774447260334,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998200509864012,
      "loss": 6.3906,
      "mean_token_accuracy": 0.12469479888677597,
      "num_tokens": 4979745.0,
      "step": 2445
    },
    {
      "entropy": 6.503656196594238,
      "epoch": 0.21410469282530806,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998188028121547,
      "loss": 6.4243,
      "mean_token_accuracy": 0.13096271976828575,
      "num_tokens": 4989924.0,
      "step": 2450
    },
    {
      "entropy": 6.437544631958008,
      "epoch": 0.21454164117801275,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998175503257658,
      "loss": 6.3653,
      "mean_token_accuracy": 0.12972407415509224,
      "num_tokens": 4999698.0,
      "step": 2455
    },
    {
      "entropy": 6.409564542770386,
      "epoch": 0.21497858953071747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998162935272585,
      "loss": 6.4533,
      "mean_token_accuracy": 0.1265200860798359,
      "num_tokens": 5010625.0,
      "step": 2460
    },
    {
      "entropy": 6.484053325653076,
      "epoch": 0.2154155378834222,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998150324166568,
      "loss": 6.3428,
      "mean_token_accuracy": 0.13328861966729164,
      "num_tokens": 5019740.0,
      "step": 2465
    },
    {
      "entropy": 6.527134990692138,
      "epoch": 0.21585248623612688,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000499813766993985,
      "loss": 6.4651,
      "mean_token_accuracy": 0.11876360327005386,
      "num_tokens": 5030601.0,
      "step": 2470
    },
    {
      "entropy": 6.523973608016968,
      "epoch": 0.2162894345888316,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004998124972592674,
      "loss": 6.4179,
      "mean_token_accuracy": 0.12937648892402648,
      "num_tokens": 5040567.0,
      "step": 2475
    },
    {
      "entropy": 6.355432462692261,
      "epoch": 0.21672638294153632,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004998112232125283,
      "loss": 6.3016,
      "mean_token_accuracy": 0.12884993553161622,
      "num_tokens": 5049512.0,
      "step": 2480
    },
    {
      "entropy": 6.446121454238892,
      "epoch": 0.217163331294241,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499809944853792,
      "loss": 6.4913,
      "mean_token_accuracy": 0.11906690523028374,
      "num_tokens": 5060986.0,
      "step": 2485
    },
    {
      "entropy": 6.456017398834229,
      "epoch": 0.21760027964694573,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998086621830833,
      "loss": 6.3963,
      "mean_token_accuracy": 0.12860883325338363,
      "num_tokens": 5071492.0,
      "step": 2490
    },
    {
      "entropy": 6.459888029098511,
      "epoch": 0.21803722799965045,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998073752004267,
      "loss": 6.4277,
      "mean_token_accuracy": 0.12244511172175407,
      "num_tokens": 5081722.0,
      "step": 2495
    },
    {
      "entropy": 6.4814647197723385,
      "epoch": 0.21847417635235516,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998060839058467,
      "loss": 6.4034,
      "mean_token_accuracy": 0.1271081253886223,
      "num_tokens": 5091906.0,
      "step": 2500
    },
    {
      "entropy": 6.439739561080932,
      "epoch": 0.21891112470505986,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998047882993683,
      "loss": 6.3715,
      "mean_token_accuracy": 0.12823385149240493,
      "num_tokens": 5102025.0,
      "step": 2505
    },
    {
      "entropy": 6.395084476470947,
      "epoch": 0.21934807305776458,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998034883810162,
      "loss": 6.3832,
      "mean_token_accuracy": 0.12374164834618569,
      "num_tokens": 5112354.0,
      "step": 2510
    },
    {
      "entropy": 6.446513080596924,
      "epoch": 0.2197850214104693,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998021841508155,
      "loss": 6.3168,
      "mean_token_accuracy": 0.13202272802591325,
      "num_tokens": 5121695.0,
      "step": 2515
    },
    {
      "entropy": 6.439951753616333,
      "epoch": 0.22022196976317399,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998008756087911,
      "loss": 6.4343,
      "mean_token_accuracy": 0.1225411631166935,
      "num_tokens": 5132561.0,
      "step": 2520
    },
    {
      "entropy": 6.497098779678344,
      "epoch": 0.2206589181158787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499799562754968,
      "loss": 6.4641,
      "mean_token_accuracy": 0.12008634880185128,
      "num_tokens": 5141981.0,
      "step": 2525
    },
    {
      "entropy": 6.426127338409424,
      "epoch": 0.22109586646858342,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004997982455893716,
      "loss": 6.3423,
      "mean_token_accuracy": 0.13012854084372522,
      "num_tokens": 5151550.0,
      "step": 2530
    },
    {
      "entropy": 6.481066370010376,
      "epoch": 0.22153281482128812,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004997969241120271,
      "loss": 6.4272,
      "mean_token_accuracy": 0.1182328850030899,
      "num_tokens": 5161679.0,
      "step": 2535
    },
    {
      "entropy": 6.384076070785523,
      "epoch": 0.22196976317399283,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997955983229597,
      "loss": 6.4474,
      "mean_token_accuracy": 0.12612560242414475,
      "num_tokens": 5171575.0,
      "step": 2540
    },
    {
      "entropy": 6.43102331161499,
      "epoch": 0.22240671152669755,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499794268222195,
      "loss": 6.3124,
      "mean_token_accuracy": 0.1305751197040081,
      "num_tokens": 5180152.0,
      "step": 2545
    },
    {
      "entropy": 6.3732438564300535,
      "epoch": 0.22284365987940224,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997929338097583,
      "loss": 6.4073,
      "mean_token_accuracy": 0.12662707194685935,
      "num_tokens": 5190518.0,
      "step": 2550
    },
    {
      "entropy": 6.544715642929077,
      "epoch": 0.22328060823210696,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997915950856755,
      "loss": 6.4306,
      "mean_token_accuracy": 0.11940659433603287,
      "num_tokens": 5201729.0,
      "step": 2555
    },
    {
      "entropy": 6.457584857940674,
      "epoch": 0.22371755658481168,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997902520499721,
      "loss": 6.4058,
      "mean_token_accuracy": 0.12618774101138114,
      "num_tokens": 5211940.0,
      "step": 2560
    },
    {
      "entropy": 6.353214883804322,
      "epoch": 0.22415450493751637,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997889047026738,
      "loss": 6.337,
      "mean_token_accuracy": 0.12396835684776306,
      "num_tokens": 5221028.0,
      "step": 2565
    },
    {
      "entropy": 6.432541036605835,
      "epoch": 0.2245914532902211,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997875530438066,
      "loss": 6.3487,
      "mean_token_accuracy": 0.12881741002202035,
      "num_tokens": 5230498.0,
      "step": 2570
    },
    {
      "entropy": 6.4462562084198,
      "epoch": 0.2250284016429258,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997861970733963,
      "loss": 6.3848,
      "mean_token_accuracy": 0.12332283109426498,
      "num_tokens": 5240112.0,
      "step": 2575
    },
    {
      "entropy": 6.438518524169922,
      "epoch": 0.2254653499956305,
      "grad_norm": 1.125,
      "learning_rate": 0.000499784836791469,
      "loss": 6.3426,
      "mean_token_accuracy": 0.12764344960451127,
      "num_tokens": 5249201.0,
      "step": 2580
    },
    {
      "entropy": 6.401770162582397,
      "epoch": 0.22590229834833522,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997834721980507,
      "loss": 6.3552,
      "mean_token_accuracy": 0.1318911150097847,
      "num_tokens": 5258541.0,
      "step": 2585
    },
    {
      "entropy": 6.547457313537597,
      "epoch": 0.22633924670103994,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004997821032931676,
      "loss": 6.4627,
      "mean_token_accuracy": 0.12480694651603699,
      "num_tokens": 5268524.0,
      "step": 2590
    },
    {
      "entropy": 6.396187162399292,
      "epoch": 0.22677619505374463,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499780730076846,
      "loss": 6.3779,
      "mean_token_accuracy": 0.1255694232881069,
      "num_tokens": 5280702.0,
      "step": 2595
    },
    {
      "entropy": 6.43830943107605,
      "epoch": 0.22721314340644935,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997793525491122,
      "loss": 6.3554,
      "mean_token_accuracy": 0.1204657144844532,
      "num_tokens": 5291214.0,
      "step": 2600
    },
    {
      "entropy": 6.344881296157837,
      "epoch": 0.22765009175915407,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997779707099926,
      "loss": 6.3195,
      "mean_token_accuracy": 0.13391299471259116,
      "num_tokens": 5300729.0,
      "step": 2605
    },
    {
      "entropy": 6.431374835968017,
      "epoch": 0.2280870401118588,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004997765845595138,
      "loss": 6.4578,
      "mean_token_accuracy": 0.12322496548295021,
      "num_tokens": 5310551.0,
      "step": 2610
    },
    {
      "entropy": 6.486410856246948,
      "epoch": 0.22852398846456348,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997751940977024,
      "loss": 6.3379,
      "mean_token_accuracy": 0.1314255066215992,
      "num_tokens": 5320038.0,
      "step": 2615
    },
    {
      "entropy": 6.34283185005188,
      "epoch": 0.2289609368172682,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997737993245848,
      "loss": 6.3722,
      "mean_token_accuracy": 0.1269502080976963,
      "num_tokens": 5329753.0,
      "step": 2620
    },
    {
      "entropy": 6.414773797988891,
      "epoch": 0.22939788516997292,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997724002401881,
      "loss": 6.3641,
      "mean_token_accuracy": 0.1279063567519188,
      "num_tokens": 5340546.0,
      "step": 2625
    },
    {
      "entropy": 6.377361679077149,
      "epoch": 0.2298348335226776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997709968445391,
      "loss": 6.4354,
      "mean_token_accuracy": 0.12187184765934944,
      "num_tokens": 5352097.0,
      "step": 2630
    },
    {
      "entropy": 6.543209218978882,
      "epoch": 0.23027178187538233,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004997695891376644,
      "loss": 6.3752,
      "mean_token_accuracy": 0.12566474825143814,
      "num_tokens": 5361646.0,
      "step": 2635
    },
    {
      "entropy": 6.391681861877442,
      "epoch": 0.23070873022808705,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004997681771195912,
      "loss": 6.3519,
      "mean_token_accuracy": 0.1263906978070736,
      "num_tokens": 5371695.0,
      "step": 2640
    },
    {
      "entropy": 6.422570037841797,
      "epoch": 0.23114567858079174,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997667607903466,
      "loss": 6.4234,
      "mean_token_accuracy": 0.12023279443383217,
      "num_tokens": 5381978.0,
      "step": 2645
    },
    {
      "entropy": 6.589051151275635,
      "epoch": 0.23158262693349646,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997653401499578,
      "loss": 6.4247,
      "mean_token_accuracy": 0.12453131377696991,
      "num_tokens": 5391891.0,
      "step": 2650
    },
    {
      "entropy": 6.452299404144287,
      "epoch": 0.23201957528620118,
      "grad_norm": 1.125,
      "learning_rate": 0.000499763915198452,
      "loss": 6.3723,
      "mean_token_accuracy": 0.1255788743495941,
      "num_tokens": 5403172.0,
      "step": 2655
    },
    {
      "entropy": 6.394309711456299,
      "epoch": 0.23245652363890587,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997624859358565,
      "loss": 6.3823,
      "mean_token_accuracy": 0.12750936523079873,
      "num_tokens": 5413130.0,
      "step": 2660
    },
    {
      "entropy": 6.447233057022094,
      "epoch": 0.2328934719916106,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997610523621988,
      "loss": 6.4223,
      "mean_token_accuracy": 0.12197845950722694,
      "num_tokens": 5423510.0,
      "step": 2665
    },
    {
      "entropy": 6.41208930015564,
      "epoch": 0.2333304203443153,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997596144775063,
      "loss": 6.2494,
      "mean_token_accuracy": 0.1339675985276699,
      "num_tokens": 5433707.0,
      "step": 2670
    },
    {
      "entropy": 6.259990262985229,
      "epoch": 0.23376736869702,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997581722818067,
      "loss": 6.3522,
      "mean_token_accuracy": 0.12899682372808458,
      "num_tokens": 5444381.0,
      "step": 2675
    },
    {
      "entropy": 6.577976036071777,
      "epoch": 0.23420431704972472,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997567257751275,
      "loss": 6.3112,
      "mean_token_accuracy": 0.12926476150751115,
      "num_tokens": 5453905.0,
      "step": 2680
    },
    {
      "entropy": 6.344999027252197,
      "epoch": 0.23464126540242944,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997552749574965,
      "loss": 6.3311,
      "mean_token_accuracy": 0.1283065065741539,
      "num_tokens": 5464839.0,
      "step": 2685
    },
    {
      "entropy": 6.396442413330078,
      "epoch": 0.23507821375513413,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997538198289416,
      "loss": 6.3734,
      "mean_token_accuracy": 0.12314258888363838,
      "num_tokens": 5475459.0,
      "step": 2690
    },
    {
      "entropy": 6.396377086639404,
      "epoch": 0.23551516210783885,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997523603894907,
      "loss": 6.3758,
      "mean_token_accuracy": 0.12924159169197083,
      "num_tokens": 5484886.0,
      "step": 2695
    },
    {
      "entropy": 6.380730533599854,
      "epoch": 0.23595211046054357,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997508966391717,
      "loss": 6.3693,
      "mean_token_accuracy": 0.1304370015859604,
      "num_tokens": 5494803.0,
      "step": 2700
    },
    {
      "entropy": 6.3598357200622555,
      "epoch": 0.2363890588132483,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997494285780129,
      "loss": 6.3124,
      "mean_token_accuracy": 0.12956960424780845,
      "num_tokens": 5504997.0,
      "step": 2705
    },
    {
      "entropy": 6.359509229660034,
      "epoch": 0.23682600716595298,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997479562060421,
      "loss": 6.3054,
      "mean_token_accuracy": 0.13558572456240653,
      "num_tokens": 5514492.0,
      "step": 2710
    },
    {
      "entropy": 6.440937232971192,
      "epoch": 0.2372629555186577,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004997464795232877,
      "loss": 6.361,
      "mean_token_accuracy": 0.12250275984406471,
      "num_tokens": 5524419.0,
      "step": 2715
    },
    {
      "entropy": 6.48492579460144,
      "epoch": 0.23769990387136242,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997449985297782,
      "loss": 6.3825,
      "mean_token_accuracy": 0.13098835572600365,
      "num_tokens": 5534289.0,
      "step": 2720
    },
    {
      "entropy": 6.397220230102539,
      "epoch": 0.2381368522240671,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004997435132255418,
      "loss": 6.2995,
      "mean_token_accuracy": 0.13025546446442604,
      "num_tokens": 5543034.0,
      "step": 2725
    },
    {
      "entropy": 6.376963567733765,
      "epoch": 0.23857380057677183,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997420236106071,
      "loss": 6.375,
      "mean_token_accuracy": 0.12310315221548081,
      "num_tokens": 5553291.0,
      "step": 2730
    },
    {
      "entropy": 6.4181102275848385,
      "epoch": 0.23901074892947655,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997405296850026,
      "loss": 6.3646,
      "mean_token_accuracy": 0.1311754897236824,
      "num_tokens": 5563772.0,
      "step": 2735
    },
    {
      "entropy": 6.461687088012695,
      "epoch": 0.23944769728218124,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997390314487569,
      "loss": 6.4034,
      "mean_token_accuracy": 0.1266815669834614,
      "num_tokens": 5574348.0,
      "step": 2740
    },
    {
      "entropy": 6.408790111541748,
      "epoch": 0.23988464563488596,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004997375289018989,
      "loss": 6.2539,
      "mean_token_accuracy": 0.13011327236890793,
      "num_tokens": 5583491.0,
      "step": 2745
    },
    {
      "entropy": 6.2955926895141605,
      "epoch": 0.24032159398759068,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004997360220444574,
      "loss": 6.4086,
      "mean_token_accuracy": 0.12347037270665169,
      "num_tokens": 5594273.0,
      "step": 2750
    },
    {
      "entropy": 6.434036064147949,
      "epoch": 0.24075854234029537,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004997345108764611,
      "loss": 6.3038,
      "mean_token_accuracy": 0.13056106939911843,
      "num_tokens": 5604529.0,
      "step": 2755
    },
    {
      "entropy": 6.390225028991699,
      "epoch": 0.2411954906930001,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997329953979391,
      "loss": 6.3111,
      "mean_token_accuracy": 0.13571627363562583,
      "num_tokens": 5614036.0,
      "step": 2760
    },
    {
      "entropy": 6.448512220382691,
      "epoch": 0.2416324390457048,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997314756089206,
      "loss": 6.3547,
      "mean_token_accuracy": 0.1281716376543045,
      "num_tokens": 5624036.0,
      "step": 2765
    },
    {
      "entropy": 6.41415810585022,
      "epoch": 0.2420693873984095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997299515094347,
      "loss": 6.3511,
      "mean_token_accuracy": 0.1302356503903866,
      "num_tokens": 5633987.0,
      "step": 2770
    },
    {
      "entropy": 6.347924423217774,
      "epoch": 0.24250633575111422,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997284230995105,
      "loss": 6.2299,
      "mean_token_accuracy": 0.13894670456647873,
      "num_tokens": 5643905.0,
      "step": 2775
    },
    {
      "entropy": 6.3557435989379885,
      "epoch": 0.24294328410381894,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997268903791773,
      "loss": 6.4261,
      "mean_token_accuracy": 0.12468690425157547,
      "num_tokens": 5654922.0,
      "step": 2780
    },
    {
      "entropy": 6.401423168182373,
      "epoch": 0.24338023245652363,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997253533484647,
      "loss": 6.2565,
      "mean_token_accuracy": 0.13697169572114945,
      "num_tokens": 5664407.0,
      "step": 2785
    },
    {
      "entropy": 6.382416534423828,
      "epoch": 0.24381718080922835,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997238120074021,
      "loss": 6.322,
      "mean_token_accuracy": 0.12700313106179237,
      "num_tokens": 5674524.0,
      "step": 2790
    },
    {
      "entropy": 6.416257572174072,
      "epoch": 0.24425412916193306,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997222663560188,
      "loss": 6.3696,
      "mean_token_accuracy": 0.12488905414938926,
      "num_tokens": 5684691.0,
      "step": 2795
    },
    {
      "entropy": 6.357256698608398,
      "epoch": 0.24469107751463778,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997207163943449,
      "loss": 6.2772,
      "mean_token_accuracy": 0.12989294528961182,
      "num_tokens": 5693683.0,
      "step": 2800
    },
    {
      "entropy": 6.347806978225708,
      "epoch": 0.24512802586734248,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004997191621224098,
      "loss": 6.3728,
      "mean_token_accuracy": 0.12666576281189917,
      "num_tokens": 5704394.0,
      "step": 2805
    },
    {
      "entropy": 6.309673023223877,
      "epoch": 0.2455649742200472,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997176035402435,
      "loss": 6.2908,
      "mean_token_accuracy": 0.13173253908753396,
      "num_tokens": 5715480.0,
      "step": 2810
    },
    {
      "entropy": 6.457736778259277,
      "epoch": 0.2460019225727519,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004997160406478758,
      "loss": 6.2801,
      "mean_token_accuracy": 0.13354317992925643,
      "num_tokens": 5725973.0,
      "step": 2815
    },
    {
      "entropy": 6.301509189605713,
      "epoch": 0.2464388709254566,
      "grad_norm": 1.375,
      "learning_rate": 0.0004997144734453367,
      "loss": 6.3082,
      "mean_token_accuracy": 0.13364396914839743,
      "num_tokens": 5735798.0,
      "step": 2820
    },
    {
      "entropy": 6.436800575256347,
      "epoch": 0.24687581927816132,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997129019326562,
      "loss": 6.4021,
      "mean_token_accuracy": 0.12107940614223481,
      "num_tokens": 5746623.0,
      "step": 2825
    },
    {
      "entropy": 6.46128396987915,
      "epoch": 0.24731276763086604,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004997113261098645,
      "loss": 6.3179,
      "mean_token_accuracy": 0.12975894808769226,
      "num_tokens": 5756526.0,
      "step": 2830
    },
    {
      "entropy": 6.367384910583496,
      "epoch": 0.24774971598357073,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997097459769919,
      "loss": 6.3136,
      "mean_token_accuracy": 0.12942242249846458,
      "num_tokens": 5766670.0,
      "step": 2835
    },
    {
      "entropy": 6.435496425628662,
      "epoch": 0.24818666433627545,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997081615340685,
      "loss": 6.3735,
      "mean_token_accuracy": 0.12207994535565377,
      "num_tokens": 5776473.0,
      "step": 2840
    },
    {
      "entropy": 6.467668104171753,
      "epoch": 0.24862361268898017,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997065727811249,
      "loss": 6.3977,
      "mean_token_accuracy": 0.1273019216954708,
      "num_tokens": 5787059.0,
      "step": 2845
    },
    {
      "entropy": 6.383358383178711,
      "epoch": 0.24906056104168486,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997049797181915,
      "loss": 6.3936,
      "mean_token_accuracy": 0.12382100597023964,
      "num_tokens": 5797367.0,
      "step": 2850
    },
    {
      "entropy": 6.367741394042969,
      "epoch": 0.24949750939438958,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004997033823452988,
      "loss": 6.2997,
      "mean_token_accuracy": 0.1279428631067276,
      "num_tokens": 5806704.0,
      "step": 2855
    },
    {
      "entropy": 6.383842611312867,
      "epoch": 0.2499344577470943,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997017806624774,
      "loss": 6.3541,
      "mean_token_accuracy": 0.12971674650907516,
      "num_tokens": 5817797.0,
      "step": 2860
    },
    {
      "entropy": 6.511070632934571,
      "epoch": 0.250371406099799,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997001746697582,
      "loss": 6.3952,
      "mean_token_accuracy": 0.12995700985193254,
      "num_tokens": 5829270.0,
      "step": 2865
    },
    {
      "entropy": 6.367383050918579,
      "epoch": 0.2508083544525037,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996985643671719,
      "loss": 6.3112,
      "mean_token_accuracy": 0.13145541176199912,
      "num_tokens": 5839214.0,
      "step": 2870
    },
    {
      "entropy": 6.328347253799438,
      "epoch": 0.2512453028052084,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004996969497547495,
      "loss": 6.3175,
      "mean_token_accuracy": 0.1253244996070862,
      "num_tokens": 5848171.0,
      "step": 2875
    },
    {
      "entropy": 6.370502042770386,
      "epoch": 0.25168225115791315,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996953308325217,
      "loss": 6.2276,
      "mean_token_accuracy": 0.13664429858326912,
      "num_tokens": 5858399.0,
      "step": 2880
    },
    {
      "entropy": 6.279939651489258,
      "epoch": 0.25211919951061784,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996937076005198,
      "loss": 6.2516,
      "mean_token_accuracy": 0.13539692386984825,
      "num_tokens": 5868212.0,
      "step": 2885
    },
    {
      "entropy": 6.395609283447266,
      "epoch": 0.25255614786332253,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996920800587749,
      "loss": 6.3115,
      "mean_token_accuracy": 0.13084554225206374,
      "num_tokens": 5877883.0,
      "step": 2890
    },
    {
      "entropy": 6.470597410202027,
      "epoch": 0.2529930962160273,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996904482073181,
      "loss": 6.316,
      "mean_token_accuracy": 0.12910602539777755,
      "num_tokens": 5887090.0,
      "step": 2895
    },
    {
      "entropy": 6.354488325119019,
      "epoch": 0.25343004456873197,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996888120461808,
      "loss": 6.3615,
      "mean_token_accuracy": 0.12137048691511154,
      "num_tokens": 5897491.0,
      "step": 2900
    },
    {
      "entropy": 6.437285137176514,
      "epoch": 0.25386699292143666,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004996871715753943,
      "loss": 6.3816,
      "mean_token_accuracy": 0.13091239109635353,
      "num_tokens": 5908494.0,
      "step": 2905
    },
    {
      "entropy": 6.450480365753174,
      "epoch": 0.2543039412741414,
      "grad_norm": 1.15625,
      "learning_rate": 0.00049968552679499,
      "loss": 6.4289,
      "mean_token_accuracy": 0.11489628329873085,
      "num_tokens": 5919708.0,
      "step": 2910
    },
    {
      "entropy": 6.467782783508301,
      "epoch": 0.2547408896268461,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004996838777049998,
      "loss": 6.2838,
      "mean_token_accuracy": 0.1296510674059391,
      "num_tokens": 5930122.0,
      "step": 2915
    },
    {
      "entropy": 6.322751808166504,
      "epoch": 0.2551778379795508,
      "grad_norm": 1.203125,
      "learning_rate": 0.000499682224305455,
      "loss": 6.2882,
      "mean_token_accuracy": 0.12527415454387664,
      "num_tokens": 5940033.0,
      "step": 2920
    },
    {
      "entropy": 6.387962388992309,
      "epoch": 0.25561478633225554,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996805665963874,
      "loss": 6.2917,
      "mean_token_accuracy": 0.12325938194990158,
      "num_tokens": 5949335.0,
      "step": 2925
    },
    {
      "entropy": 6.3954322814941404,
      "epoch": 0.25605173468496023,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004996789045778287,
      "loss": 6.3278,
      "mean_token_accuracy": 0.13017525672912597,
      "num_tokens": 5959299.0,
      "step": 2930
    },
    {
      "entropy": 6.388931512832642,
      "epoch": 0.2564886830376649,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004996772382498111,
      "loss": 6.2909,
      "mean_token_accuracy": 0.12773813903331757,
      "num_tokens": 5970114.0,
      "step": 2935
    },
    {
      "entropy": 6.375958824157715,
      "epoch": 0.25692563139036967,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996755676123662,
      "loss": 6.3937,
      "mean_token_accuracy": 0.12411935850977898,
      "num_tokens": 5979630.0,
      "step": 2940
    },
    {
      "entropy": 6.466895532608032,
      "epoch": 0.25736257974307436,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996738926655262,
      "loss": 6.3775,
      "mean_token_accuracy": 0.12354947850108147,
      "num_tokens": 5990190.0,
      "step": 2945
    },
    {
      "entropy": 6.422803926467895,
      "epoch": 0.25779952809577905,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996722134093231,
      "loss": 6.3873,
      "mean_token_accuracy": 0.12062846720218659,
      "num_tokens": 6000706.0,
      "step": 2950
    },
    {
      "entropy": 6.330030822753907,
      "epoch": 0.2582364764484838,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996705298437894,
      "loss": 6.3495,
      "mean_token_accuracy": 0.1309203699231148,
      "num_tokens": 6011274.0,
      "step": 2955
    },
    {
      "entropy": 6.472089052200317,
      "epoch": 0.2586734248011885,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996688419689572,
      "loss": 6.265,
      "mean_token_accuracy": 0.13017945513129234,
      "num_tokens": 6021481.0,
      "step": 2960
    },
    {
      "entropy": 6.277119541168213,
      "epoch": 0.25911037315389324,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996671497848589,
      "loss": 6.3421,
      "mean_token_accuracy": 0.1263204239308834,
      "num_tokens": 6032091.0,
      "step": 2965
    },
    {
      "entropy": 6.395982933044434,
      "epoch": 0.25954732150659793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996654532915269,
      "loss": 6.3556,
      "mean_token_accuracy": 0.11947837546467781,
      "num_tokens": 6043371.0,
      "step": 2970
    },
    {
      "entropy": 6.389321994781494,
      "epoch": 0.2599842698593026,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004996637524889937,
      "loss": 6.299,
      "mean_token_accuracy": 0.12635288685560225,
      "num_tokens": 6052812.0,
      "step": 2975
    },
    {
      "entropy": 6.3647873878479,
      "epoch": 0.26042121821200737,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996620473772921,
      "loss": 6.2899,
      "mean_token_accuracy": 0.1310460463166237,
      "num_tokens": 6064033.0,
      "step": 2980
    },
    {
      "entropy": 6.37945704460144,
      "epoch": 0.26085816656471206,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996603379564546,
      "loss": 6.3966,
      "mean_token_accuracy": 0.1263035662472248,
      "num_tokens": 6075837.0,
      "step": 2985
    },
    {
      "entropy": 6.492451429367065,
      "epoch": 0.26129511491741675,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004996586242265142,
      "loss": 6.3054,
      "mean_token_accuracy": 0.133855964243412,
      "num_tokens": 6085514.0,
      "step": 2990
    },
    {
      "entropy": 6.431004238128662,
      "epoch": 0.2617320632701215,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004996569061875036,
      "loss": 6.2399,
      "mean_token_accuracy": 0.13191918209195136,
      "num_tokens": 6094992.0,
      "step": 2995
    },
    {
      "entropy": 6.3046787738800045,
      "epoch": 0.2621690116228262,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996551838394559,
      "loss": 6.2998,
      "mean_token_accuracy": 0.1289274014532566,
      "num_tokens": 6104841.0,
      "step": 3000
    },
    {
      "epoch": 0.2621690116228262,
      "eval_entropy": 6.176911606859743,
      "eval_loss": 6.340244770050049,
      "eval_mean_token_accuracy": 0.1347498616543024,
      "eval_num_tokens": 6104841.0,
      "eval_runtime": 28.264,
      "eval_samples_per_second": 1309.582,
      "eval_steps_per_second": 163.707,
      "step": 3000
    },
    {
      "entropy": 6.26025013923645,
      "epoch": 0.2626059599755309,
      "grad_norm": 1.25,
      "learning_rate": 0.000499653457182404,
      "loss": 6.1838,
      "mean_token_accuracy": 0.1339375652372837,
      "num_tokens": 6114830.0,
      "step": 3005
    },
    {
      "entropy": 6.356398057937622,
      "epoch": 0.2630429083282356,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996517262163811,
      "loss": 6.3223,
      "mean_token_accuracy": 0.1326840378344059,
      "num_tokens": 6125165.0,
      "step": 3010
    },
    {
      "entropy": 6.354022789001465,
      "epoch": 0.2634798566809403,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996499909414204,
      "loss": 6.265,
      "mean_token_accuracy": 0.12968195155262946,
      "num_tokens": 6135670.0,
      "step": 3015
    },
    {
      "entropy": 6.269655513763428,
      "epoch": 0.263916805033645,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996482513575551,
      "loss": 6.2436,
      "mean_token_accuracy": 0.14049533605575562,
      "num_tokens": 6146055.0,
      "step": 3020
    },
    {
      "entropy": 6.316104173660278,
      "epoch": 0.26435375338634975,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004996465074648186,
      "loss": 6.2646,
      "mean_token_accuracy": 0.1341840773820877,
      "num_tokens": 6155941.0,
      "step": 3025
    },
    {
      "entropy": 6.4075220108032225,
      "epoch": 0.26479070173905445,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004996447592632444,
      "loss": 6.3538,
      "mean_token_accuracy": 0.13025905638933183,
      "num_tokens": 6166291.0,
      "step": 3030
    },
    {
      "entropy": 6.35784592628479,
      "epoch": 0.26522765009175914,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004996430067528661,
      "loss": 6.3121,
      "mean_token_accuracy": 0.1270274318754673,
      "num_tokens": 6175294.0,
      "step": 3035
    },
    {
      "entropy": 6.434535026550293,
      "epoch": 0.2656645984444639,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996412499337172,
      "loss": 6.2997,
      "mean_token_accuracy": 0.13426331281661988,
      "num_tokens": 6185273.0,
      "step": 3040
    },
    {
      "entropy": 6.3774840354919435,
      "epoch": 0.2661015467971686,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996394888058314,
      "loss": 6.2649,
      "mean_token_accuracy": 0.13806363344192504,
      "num_tokens": 6195075.0,
      "step": 3045
    },
    {
      "entropy": 6.239452505111695,
      "epoch": 0.26653849514987327,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004996377233692425,
      "loss": 6.1567,
      "mean_token_accuracy": 0.14420341700315475,
      "num_tokens": 6206018.0,
      "step": 3050
    },
    {
      "entropy": 6.349570274353027,
      "epoch": 0.266975443502578,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004996359536239843,
      "loss": 6.2962,
      "mean_token_accuracy": 0.12873756289482116,
      "num_tokens": 6216577.0,
      "step": 3055
    },
    {
      "entropy": 6.418041610717774,
      "epoch": 0.2674123918552827,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004996341795700908,
      "loss": 6.3134,
      "mean_token_accuracy": 0.1307532049715519,
      "num_tokens": 6226521.0,
      "step": 3060
    },
    {
      "entropy": 6.285133123397827,
      "epoch": 0.2678493402079874,
      "grad_norm": 1.5,
      "learning_rate": 0.0004996324012075961,
      "loss": 6.2903,
      "mean_token_accuracy": 0.13310011625289916,
      "num_tokens": 6236509.0,
      "step": 3065
    },
    {
      "entropy": 6.4100165367126465,
      "epoch": 0.26828628856069214,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004996306185365341,
      "loss": 6.2967,
      "mean_token_accuracy": 0.13204259872436525,
      "num_tokens": 6246584.0,
      "step": 3070
    },
    {
      "entropy": 6.364015245437622,
      "epoch": 0.26872323691339683,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004996288315569393,
      "loss": 6.313,
      "mean_token_accuracy": 0.12507192716002463,
      "num_tokens": 6256881.0,
      "step": 3075
    },
    {
      "entropy": 6.390865612030029,
      "epoch": 0.2691601852661015,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996270402688457,
      "loss": 6.2919,
      "mean_token_accuracy": 0.130061012506485,
      "num_tokens": 6266380.0,
      "step": 3080
    },
    {
      "entropy": 6.405587100982666,
      "epoch": 0.2695971336188063,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004996252446722877,
      "loss": 6.3237,
      "mean_token_accuracy": 0.13164027556777,
      "num_tokens": 6274872.0,
      "step": 3085
    },
    {
      "entropy": 6.342977666854859,
      "epoch": 0.27003408197151096,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004996234447673,
      "loss": 6.329,
      "mean_token_accuracy": 0.12905460298061372,
      "num_tokens": 6286167.0,
      "step": 3090
    },
    {
      "entropy": 6.405268287658691,
      "epoch": 0.27047103032421566,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004996216405539169,
      "loss": 6.2808,
      "mean_token_accuracy": 0.12965068444609643,
      "num_tokens": 6295898.0,
      "step": 3095
    },
    {
      "entropy": 6.39916033744812,
      "epoch": 0.2709079786769204,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004996198320321728,
      "loss": 6.3046,
      "mean_token_accuracy": 0.13214947730302812,
      "num_tokens": 6306543.0,
      "step": 3100
    },
    {
      "entropy": 6.35726146697998,
      "epoch": 0.2713449270296251,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004996180192021029,
      "loss": 6.2959,
      "mean_token_accuracy": 0.12381328046321868,
      "num_tokens": 6317709.0,
      "step": 3105
    },
    {
      "entropy": 6.319500064849853,
      "epoch": 0.2717818753823298,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004996162020637415,
      "loss": 6.2633,
      "mean_token_accuracy": 0.1328990489244461,
      "num_tokens": 6326874.0,
      "step": 3110
    },
    {
      "entropy": 6.3565386772155765,
      "epoch": 0.27221882373503453,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996143806171238,
      "loss": 6.3577,
      "mean_token_accuracy": 0.12793173417448997,
      "num_tokens": 6337943.0,
      "step": 3115
    },
    {
      "entropy": 6.36138949394226,
      "epoch": 0.2726557720877392,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996125548622845,
      "loss": 6.4066,
      "mean_token_accuracy": 0.12947756350040435,
      "num_tokens": 6348442.0,
      "step": 3120
    },
    {
      "entropy": 6.437826919555664,
      "epoch": 0.2730927204404439,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996107247992587,
      "loss": 6.2951,
      "mean_token_accuracy": 0.13008005246520044,
      "num_tokens": 6358675.0,
      "step": 3125
    },
    {
      "entropy": 6.430121755599975,
      "epoch": 0.27352966879314866,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004996088904280815,
      "loss": 6.3482,
      "mean_token_accuracy": 0.12924067303538322,
      "num_tokens": 6368631.0,
      "step": 3130
    },
    {
      "entropy": 6.361489963531494,
      "epoch": 0.27396661714585335,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004996070517487881,
      "loss": 6.2283,
      "mean_token_accuracy": 0.13231453523039818,
      "num_tokens": 6378709.0,
      "step": 3135
    },
    {
      "entropy": 6.326501083374024,
      "epoch": 0.27440356549855804,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004996052087614138,
      "loss": 6.3593,
      "mean_token_accuracy": 0.12339177057147026,
      "num_tokens": 6389686.0,
      "step": 3140
    },
    {
      "entropy": 6.40691351890564,
      "epoch": 0.2748405138512628,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996033614659938,
      "loss": 6.3156,
      "mean_token_accuracy": 0.1295914389193058,
      "num_tokens": 6399447.0,
      "step": 3145
    },
    {
      "entropy": 6.406366586685181,
      "epoch": 0.2752774622039675,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004996015098625638,
      "loss": 6.3102,
      "mean_token_accuracy": 0.13160308226943015,
      "num_tokens": 6408891.0,
      "step": 3150
    },
    {
      "entropy": 6.349420166015625,
      "epoch": 0.2757144105566722,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000499599653951159,
      "loss": 6.2697,
      "mean_token_accuracy": 0.13136155307292938,
      "num_tokens": 6419471.0,
      "step": 3155
    },
    {
      "entropy": 6.306204700469971,
      "epoch": 0.2761513589093769,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004995977937318153,
      "loss": 6.2657,
      "mean_token_accuracy": 0.12809676453471183,
      "num_tokens": 6430445.0,
      "step": 3160
    },
    {
      "entropy": 6.458633613586426,
      "epoch": 0.2765883072620816,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499595929204568,
      "loss": 6.2973,
      "mean_token_accuracy": 0.12807265892624856,
      "num_tokens": 6440821.0,
      "step": 3165
    },
    {
      "entropy": 6.287018537521362,
      "epoch": 0.27702525561478636,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004995940603694533,
      "loss": 6.2048,
      "mean_token_accuracy": 0.1330143354833126,
      "num_tokens": 6450958.0,
      "step": 3170
    },
    {
      "entropy": 6.275545883178711,
      "epoch": 0.27746220396749105,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004995921872265067,
      "loss": 6.3197,
      "mean_token_accuracy": 0.1292585127055645,
      "num_tokens": 6462106.0,
      "step": 3175
    },
    {
      "entropy": 6.415719652175904,
      "epoch": 0.27789915232019574,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004995903097757643,
      "loss": 6.1978,
      "mean_token_accuracy": 0.13924278244376181,
      "num_tokens": 6471890.0,
      "step": 3180
    },
    {
      "entropy": 6.258716344833374,
      "epoch": 0.2783361006729005,
      "grad_norm": 1.5,
      "learning_rate": 0.0004995884280172622,
      "loss": 6.2407,
      "mean_token_accuracy": 0.13054869547486306,
      "num_tokens": 6481959.0,
      "step": 3185
    },
    {
      "entropy": 6.321152019500732,
      "epoch": 0.2787730490256052,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004995865419510362,
      "loss": 6.2624,
      "mean_token_accuracy": 0.13158614411950112,
      "num_tokens": 6492086.0,
      "step": 3190
    },
    {
      "entropy": 6.370549821853638,
      "epoch": 0.27920999737830987,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004995846515771227,
      "loss": 6.2923,
      "mean_token_accuracy": 0.12675996571779252,
      "num_tokens": 6502530.0,
      "step": 3195
    },
    {
      "entropy": 6.380843448638916,
      "epoch": 0.2796469457310146,
      "grad_norm": 6.15625,
      "learning_rate": 0.000499582756895558,
      "loss": 6.2918,
      "mean_token_accuracy": 0.13517867475748063,
      "num_tokens": 6511600.0,
      "step": 3200
    },
    {
      "entropy": 6.403674364089966,
      "epoch": 0.2800838940837193,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004995808579063782,
      "loss": 6.2769,
      "mean_token_accuracy": 0.12687014937400817,
      "num_tokens": 6522712.0,
      "step": 3205
    },
    {
      "entropy": 6.185982513427734,
      "epoch": 0.280520842436424,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00049957895460962,
      "loss": 6.1342,
      "mean_token_accuracy": 0.14963648468255997,
      "num_tokens": 6532472.0,
      "step": 3210
    },
    {
      "entropy": 6.313288688659668,
      "epoch": 0.28095779078912875,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004995770470053197,
      "loss": 6.1959,
      "mean_token_accuracy": 0.1432195670902729,
      "num_tokens": 6541778.0,
      "step": 3215
    },
    {
      "entropy": 6.362827157974243,
      "epoch": 0.28139473914183344,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004995751350935139,
      "loss": 6.1898,
      "mean_token_accuracy": 0.1278350219130516,
      "num_tokens": 6552368.0,
      "step": 3220
    },
    {
      "entropy": 6.309809732437134,
      "epoch": 0.28183168749453813,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004995732188742395,
      "loss": 6.2401,
      "mean_token_accuracy": 0.13236903101205827,
      "num_tokens": 6562593.0,
      "step": 3225
    },
    {
      "entropy": 6.362755393981933,
      "epoch": 0.2822686358472429,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004995712983475329,
      "loss": 6.2941,
      "mean_token_accuracy": 0.1256416767835617,
      "num_tokens": 6572695.0,
      "step": 3230
    },
    {
      "entropy": 6.330557012557984,
      "epoch": 0.28270558419994757,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004995693735134313,
      "loss": 6.2627,
      "mean_token_accuracy": 0.13055431172251702,
      "num_tokens": 6581671.0,
      "step": 3235
    },
    {
      "entropy": 6.363641738891602,
      "epoch": 0.28314253255265226,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004995674443719713,
      "loss": 6.2622,
      "mean_token_accuracy": 0.13069088533520698,
      "num_tokens": 6590460.0,
      "step": 3240
    },
    {
      "entropy": 6.376384353637695,
      "epoch": 0.283579480905357,
      "grad_norm": 1.3125,
      "learning_rate": 0.00049956551092319,
      "loss": 6.2724,
      "mean_token_accuracy": 0.12941850423812867,
      "num_tokens": 6599892.0,
      "step": 3245
    },
    {
      "entropy": 6.371181440353394,
      "epoch": 0.2840164292580617,
      "grad_norm": 1.5,
      "learning_rate": 0.0004995635731671247,
      "loss": 6.3544,
      "mean_token_accuracy": 0.12331348508596421,
      "num_tokens": 6610530.0,
      "step": 3250
    },
    {
      "entropy": 6.366174936294556,
      "epoch": 0.2844533776107664,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004995616311038124,
      "loss": 6.2512,
      "mean_token_accuracy": 0.13390601947903633,
      "num_tokens": 6621117.0,
      "step": 3255
    },
    {
      "entropy": 6.360169410705566,
      "epoch": 0.28489032596347114,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004995596847332901,
      "loss": 6.3222,
      "mean_token_accuracy": 0.1278347410261631,
      "num_tokens": 6631521.0,
      "step": 3260
    },
    {
      "entropy": 6.463128566741943,
      "epoch": 0.2853272743161758,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004995577340555956,
      "loss": 6.3904,
      "mean_token_accuracy": 0.12619309723377228,
      "num_tokens": 6642180.0,
      "step": 3265
    },
    {
      "entropy": 6.35454044342041,
      "epoch": 0.2857642226688805,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004995557790707661,
      "loss": 6.3048,
      "mean_token_accuracy": 0.13104436099529265,
      "num_tokens": 6652735.0,
      "step": 3270
    },
    {
      "entropy": 6.3466637134552,
      "epoch": 0.28620117102158527,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000499553819778839,
      "loss": 6.2974,
      "mean_token_accuracy": 0.13476421907544137,
      "num_tokens": 6663135.0,
      "step": 3275
    },
    {
      "entropy": 6.359936475753784,
      "epoch": 0.28663811937428996,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000499551856179852,
      "loss": 6.3108,
      "mean_token_accuracy": 0.12338437512516975,
      "num_tokens": 6673375.0,
      "step": 3280
    },
    {
      "entropy": 6.402853918075562,
      "epoch": 0.28707506772699465,
      "grad_norm": 1.25,
      "learning_rate": 0.0004995498882738428,
      "loss": 6.2559,
      "mean_token_accuracy": 0.13144599050283431,
      "num_tokens": 6683101.0,
      "step": 3285
    },
    {
      "entropy": 6.4232560157775875,
      "epoch": 0.2875120160796994,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004995479160608491,
      "loss": 6.3164,
      "mean_token_accuracy": 0.13341468274593354,
      "num_tokens": 6692763.0,
      "step": 3290
    },
    {
      "entropy": 6.3722888946533205,
      "epoch": 0.2879489644324041,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004995459395409085,
      "loss": 6.3002,
      "mean_token_accuracy": 0.12749466225504874,
      "num_tokens": 6702833.0,
      "step": 3295
    },
    {
      "entropy": 6.370219612121582,
      "epoch": 0.2883859127851088,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004995439587140593,
      "loss": 6.3132,
      "mean_token_accuracy": 0.12153316885232926,
      "num_tokens": 6713083.0,
      "step": 3300
    },
    {
      "entropy": 6.3562493324279785,
      "epoch": 0.2888228611378135,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004995419735803392,
      "loss": 6.2323,
      "mean_token_accuracy": 0.12721289396286012,
      "num_tokens": 6722501.0,
      "step": 3305
    },
    {
      "entropy": 6.342270755767823,
      "epoch": 0.2892598094905182,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004995399841397865,
      "loss": 6.2746,
      "mean_token_accuracy": 0.12882607579231262,
      "num_tokens": 6731961.0,
      "step": 3310
    },
    {
      "entropy": 6.394270992279052,
      "epoch": 0.2896967578432229,
      "grad_norm": 1.625,
      "learning_rate": 0.0004995379903924392,
      "loss": 6.2678,
      "mean_token_accuracy": 0.1273388981819153,
      "num_tokens": 6742576.0,
      "step": 3315
    },
    {
      "entropy": 6.378604507446289,
      "epoch": 0.29013370619592765,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004995359923383356,
      "loss": 6.2775,
      "mean_token_accuracy": 0.1297510638833046,
      "num_tokens": 6753066.0,
      "step": 3320
    },
    {
      "entropy": 6.392166423797607,
      "epoch": 0.29057065454863235,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000499533989977514,
      "loss": 6.415,
      "mean_token_accuracy": 0.12183653116226197,
      "num_tokens": 6764455.0,
      "step": 3325
    },
    {
      "entropy": 6.418638849258423,
      "epoch": 0.29100760290133704,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004995319833100128,
      "loss": 6.2895,
      "mean_token_accuracy": 0.13210763931274414,
      "num_tokens": 6774655.0,
      "step": 3330
    },
    {
      "entropy": 6.320852899551392,
      "epoch": 0.2914445512540418,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004995299723358705,
      "loss": 6.2492,
      "mean_token_accuracy": 0.13643589839339257,
      "num_tokens": 6785942.0,
      "step": 3335
    },
    {
      "entropy": 6.265985202789307,
      "epoch": 0.2918814996067465,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995279570551257,
      "loss": 6.2215,
      "mean_token_accuracy": 0.13763746991753578,
      "num_tokens": 6795355.0,
      "step": 3340
    },
    {
      "entropy": 6.3764269828796385,
      "epoch": 0.29231844795945117,
      "grad_norm": 2.125,
      "learning_rate": 0.000499525937467817,
      "loss": 6.28,
      "mean_token_accuracy": 0.12502550855278968,
      "num_tokens": 6805579.0,
      "step": 3345
    },
    {
      "entropy": 6.417794609069825,
      "epoch": 0.2927553963121559,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004995239135739833,
      "loss": 6.2996,
      "mean_token_accuracy": 0.12507596090435982,
      "num_tokens": 6814878.0,
      "step": 3350
    },
    {
      "entropy": 6.373486852645874,
      "epoch": 0.2931923446648606,
      "grad_norm": 1.875,
      "learning_rate": 0.0004995218853736631,
      "loss": 6.1955,
      "mean_token_accuracy": 0.141049724817276,
      "num_tokens": 6824533.0,
      "step": 3355
    },
    {
      "entropy": 6.191241073608398,
      "epoch": 0.29362929301756535,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004995198528668956,
      "loss": 6.1318,
      "mean_token_accuracy": 0.14864774271845818,
      "num_tokens": 6835529.0,
      "step": 3360
    },
    {
      "entropy": 6.3345029830932615,
      "epoch": 0.29406624137027004,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004995178160537197,
      "loss": 6.3096,
      "mean_token_accuracy": 0.13062748312950134,
      "num_tokens": 6845679.0,
      "step": 3365
    },
    {
      "entropy": 6.46401891708374,
      "epoch": 0.29450318972297473,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004995157749341744,
      "loss": 6.2967,
      "mean_token_accuracy": 0.12413412407040596,
      "num_tokens": 6855876.0,
      "step": 3370
    },
    {
      "entropy": 6.382764053344727,
      "epoch": 0.2949401380756795,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004995137295082989,
      "loss": 6.3535,
      "mean_token_accuracy": 0.12331586331129074,
      "num_tokens": 6866200.0,
      "step": 3375
    },
    {
      "entropy": 6.389991807937622,
      "epoch": 0.2953770864283842,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004995116797761324,
      "loss": 6.2298,
      "mean_token_accuracy": 0.13341188728809356,
      "num_tokens": 6875191.0,
      "step": 3380
    },
    {
      "entropy": 6.292295694351196,
      "epoch": 0.29581403478108886,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004995096257377143,
      "loss": 6.2124,
      "mean_token_accuracy": 0.1290462002158165,
      "num_tokens": 6885352.0,
      "step": 3385
    },
    {
      "entropy": 6.323753833770752,
      "epoch": 0.2962509831337936,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004995075673930839,
      "loss": 6.2947,
      "mean_token_accuracy": 0.12870371863245963,
      "num_tokens": 6895585.0,
      "step": 3390
    },
    {
      "entropy": 6.415827417373658,
      "epoch": 0.2966879314864983,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004995055047422808,
      "loss": 6.2397,
      "mean_token_accuracy": 0.1292224906384945,
      "num_tokens": 6905532.0,
      "step": 3395
    },
    {
      "entropy": 6.278207731246948,
      "epoch": 0.297124879839203,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995034377853443,
      "loss": 6.2564,
      "mean_token_accuracy": 0.13268594294786454,
      "num_tokens": 6915835.0,
      "step": 3400
    },
    {
      "entropy": 6.316234302520752,
      "epoch": 0.29756182819190774,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004995013665223143,
      "loss": 6.2921,
      "mean_token_accuracy": 0.1291264720261097,
      "num_tokens": 6925964.0,
      "step": 3405
    },
    {
      "entropy": 6.4064103126525875,
      "epoch": 0.29799877654461243,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004994992909532305,
      "loss": 6.2362,
      "mean_token_accuracy": 0.13458333984017373,
      "num_tokens": 6936145.0,
      "step": 3410
    },
    {
      "entropy": 6.3627848625183105,
      "epoch": 0.2984357248973171,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004994972110781326,
      "loss": 6.2006,
      "mean_token_accuracy": 0.13037536144256592,
      "num_tokens": 6945557.0,
      "step": 3415
    },
    {
      "entropy": 6.272304630279541,
      "epoch": 0.29887267325002187,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004994951268970606,
      "loss": 6.2497,
      "mean_token_accuracy": 0.12965808659791947,
      "num_tokens": 6955147.0,
      "step": 3420
    },
    {
      "entropy": 6.297464752197266,
      "epoch": 0.29930962160272656,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004994930384100545,
      "loss": 6.2003,
      "mean_token_accuracy": 0.13216258585453033,
      "num_tokens": 6964024.0,
      "step": 3425
    },
    {
      "entropy": 6.344718933105469,
      "epoch": 0.29974656995543125,
      "grad_norm": 1.25,
      "learning_rate": 0.0004994909456171541,
      "loss": 6.2799,
      "mean_token_accuracy": 0.13039335906505584,
      "num_tokens": 6974399.0,
      "step": 3430
    },
    {
      "entropy": 6.280664157867432,
      "epoch": 0.300183518308136,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994888485183998,
      "loss": 6.2214,
      "mean_token_accuracy": 0.13267827704548835,
      "num_tokens": 6984324.0,
      "step": 3435
    },
    {
      "entropy": 6.359601449966431,
      "epoch": 0.3006204666608407,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000499486747113832,
      "loss": 6.241,
      "mean_token_accuracy": 0.1327981173992157,
      "num_tokens": 6995659.0,
      "step": 3440
    },
    {
      "entropy": 6.401485824584961,
      "epoch": 0.3010574150135454,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004994846414034905,
      "loss": 6.3482,
      "mean_token_accuracy": 0.1251749128103256,
      "num_tokens": 7006463.0,
      "step": 3445
    },
    {
      "entropy": 6.366568899154663,
      "epoch": 0.30149436336625013,
      "grad_norm": 1.421875,
      "learning_rate": 0.000499482531387416,
      "loss": 6.2367,
      "mean_token_accuracy": 0.13053662180900574,
      "num_tokens": 7016188.0,
      "step": 3450
    },
    {
      "entropy": 6.330869817733765,
      "epoch": 0.3019313117189548,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004994804170656489,
      "loss": 6.2761,
      "mean_token_accuracy": 0.12821143344044686,
      "num_tokens": 7028435.0,
      "step": 3455
    },
    {
      "entropy": 6.3709094524383545,
      "epoch": 0.3023682600716595,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004994782984382299,
      "loss": 6.2637,
      "mean_token_accuracy": 0.13231005817651748,
      "num_tokens": 7037992.0,
      "step": 3460
    },
    {
      "entropy": 6.352482557296753,
      "epoch": 0.30280520842436426,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004994761755051995,
      "loss": 6.2831,
      "mean_token_accuracy": 0.12955223843455316,
      "num_tokens": 7048743.0,
      "step": 3465
    },
    {
      "entropy": 6.381830358505249,
      "epoch": 0.30324215677706895,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004994740482665982,
      "loss": 6.2275,
      "mean_token_accuracy": 0.1312493935227394,
      "num_tokens": 7059524.0,
      "step": 3470
    },
    {
      "entropy": 6.291852378845215,
      "epoch": 0.30367910512977364,
      "grad_norm": 9.5625,
      "learning_rate": 0.0004994719167224672,
      "loss": 6.2339,
      "mean_token_accuracy": 0.12652599215507507,
      "num_tokens": 7069742.0,
      "step": 3475
    },
    {
      "entropy": 6.2860442161560055,
      "epoch": 0.3041160534824784,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004994697808728472,
      "loss": 6.2943,
      "mean_token_accuracy": 0.13188235983252525,
      "num_tokens": 7079909.0,
      "step": 3480
    },
    {
      "entropy": 6.4090958595275875,
      "epoch": 0.3045530018351831,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004994676407177791,
      "loss": 6.3066,
      "mean_token_accuracy": 0.13042071908712388,
      "num_tokens": 7089136.0,
      "step": 3485
    },
    {
      "entropy": 6.4032796859741214,
      "epoch": 0.30498995018788777,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004994654962573043,
      "loss": 6.3154,
      "mean_token_accuracy": 0.13368391543626784,
      "num_tokens": 7099621.0,
      "step": 3490
    },
    {
      "entropy": 6.346981525421143,
      "epoch": 0.3054268985405925,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004994633474914634,
      "loss": 6.1712,
      "mean_token_accuracy": 0.13378375470638276,
      "num_tokens": 7109599.0,
      "step": 3495
    },
    {
      "entropy": 6.18639817237854,
      "epoch": 0.3058638468932972,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004994611944202979,
      "loss": 6.2166,
      "mean_token_accuracy": 0.13542467579245568,
      "num_tokens": 7120733.0,
      "step": 3500
    },
    {
      "entropy": 6.3771179676055905,
      "epoch": 0.3063007952460019,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004994590370438492,
      "loss": 6.2971,
      "mean_token_accuracy": 0.12855091989040374,
      "num_tokens": 7129871.0,
      "step": 3505
    },
    {
      "entropy": 6.4098076820373535,
      "epoch": 0.30673774359870665,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004994568753621585,
      "loss": 6.2136,
      "mean_token_accuracy": 0.1323453053832054,
      "num_tokens": 7139167.0,
      "step": 3510
    },
    {
      "entropy": 6.232396030426026,
      "epoch": 0.30717469195141134,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004994547093752673,
      "loss": 6.2976,
      "mean_token_accuracy": 0.12436885684728623,
      "num_tokens": 7150534.0,
      "step": 3515
    },
    {
      "entropy": 6.342797899246216,
      "epoch": 0.30761164030411603,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004994525390832171,
      "loss": 6.227,
      "mean_token_accuracy": 0.13226970210671424,
      "num_tokens": 7160472.0,
      "step": 3520
    },
    {
      "entropy": 6.419403028488159,
      "epoch": 0.3080485886568208,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004994503644860497,
      "loss": 6.2714,
      "mean_token_accuracy": 0.1302813246846199,
      "num_tokens": 7171636.0,
      "step": 3525
    },
    {
      "entropy": 6.316999244689941,
      "epoch": 0.30848553700952547,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004994481855838067,
      "loss": 6.2156,
      "mean_token_accuracy": 0.1317649282515049,
      "num_tokens": 7182163.0,
      "step": 3530
    },
    {
      "entropy": 6.308993196487426,
      "epoch": 0.30892248536223016,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004994460023765298,
      "loss": 6.2317,
      "mean_token_accuracy": 0.13583657294511794,
      "num_tokens": 7191694.0,
      "step": 3535
    },
    {
      "entropy": 6.351598787307739,
      "epoch": 0.3093594337149349,
      "grad_norm": 2.515625,
      "learning_rate": 0.000499443814864261,
      "loss": 6.2806,
      "mean_token_accuracy": 0.13465951904654502,
      "num_tokens": 7202813.0,
      "step": 3540
    },
    {
      "entropy": 6.349387216567993,
      "epoch": 0.3097963820676396,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004994416230470422,
      "loss": 6.2717,
      "mean_token_accuracy": 0.12884895727038384,
      "num_tokens": 7213226.0,
      "step": 3545
    },
    {
      "entropy": 6.407685232162476,
      "epoch": 0.3102333304203443,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004994394269249155,
      "loss": 6.2859,
      "mean_token_accuracy": 0.12866515889763833,
      "num_tokens": 7223520.0,
      "step": 3550
    },
    {
      "entropy": 6.315626335144043,
      "epoch": 0.31067027877304904,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004994372264979231,
      "loss": 6.2151,
      "mean_token_accuracy": 0.1384662352502346,
      "num_tokens": 7232651.0,
      "step": 3555
    },
    {
      "entropy": 6.338417482376099,
      "epoch": 0.3111072271257537,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004994350217661069,
      "loss": 6.2752,
      "mean_token_accuracy": 0.12966468408703805,
      "num_tokens": 7242064.0,
      "step": 3560
    },
    {
      "entropy": 6.3671448707580565,
      "epoch": 0.3115441754784585,
      "grad_norm": 1.375,
      "learning_rate": 0.0004994328127295095,
      "loss": 6.2833,
      "mean_token_accuracy": 0.1294167585670948,
      "num_tokens": 7253048.0,
      "step": 3565
    },
    {
      "entropy": 6.33545069694519,
      "epoch": 0.31198112383116317,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004994305993881731,
      "loss": 6.2658,
      "mean_token_accuracy": 0.13346811681985854,
      "num_tokens": 7262516.0,
      "step": 3570
    },
    {
      "entropy": 6.290219306945801,
      "epoch": 0.31241807218386786,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004994283817421402,
      "loss": 6.1709,
      "mean_token_accuracy": 0.13820456489920616,
      "num_tokens": 7272079.0,
      "step": 3575
    },
    {
      "entropy": 6.292543888092041,
      "epoch": 0.3128550205365726,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004994261597914534,
      "loss": 6.1505,
      "mean_token_accuracy": 0.14289118051528932,
      "num_tokens": 7282150.0,
      "step": 3580
    },
    {
      "entropy": 6.304704809188843,
      "epoch": 0.3132919688892773,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004994239335361552,
      "loss": 6.2648,
      "mean_token_accuracy": 0.12817902565002443,
      "num_tokens": 7291782.0,
      "step": 3585
    },
    {
      "entropy": 6.298792600631714,
      "epoch": 0.313728917241982,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004994217029762884,
      "loss": 6.2257,
      "mean_token_accuracy": 0.13249474465847016,
      "num_tokens": 7301472.0,
      "step": 3590
    },
    {
      "entropy": 6.3423254013061525,
      "epoch": 0.31416586559468673,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004994194681118957,
      "loss": 6.207,
      "mean_token_accuracy": 0.13881093189120292,
      "num_tokens": 7311456.0,
      "step": 3595
    },
    {
      "entropy": 6.373311948776245,
      "epoch": 0.3146028139473914,
      "grad_norm": 1.46875,
      "learning_rate": 0.00049941722894302,
      "loss": 6.2749,
      "mean_token_accuracy": 0.13226548358798026,
      "num_tokens": 7321106.0,
      "step": 3600
    },
    {
      "entropy": 6.356042814254761,
      "epoch": 0.3150397623000961,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004994149854697042,
      "loss": 6.2768,
      "mean_token_accuracy": 0.13341066539287566,
      "num_tokens": 7331232.0,
      "step": 3605
    },
    {
      "entropy": 6.335075378417969,
      "epoch": 0.31547671065280086,
      "grad_norm": 1.25,
      "learning_rate": 0.0004994127376919916,
      "loss": 6.255,
      "mean_token_accuracy": 0.12598878741264344,
      "num_tokens": 7341560.0,
      "step": 3610
    },
    {
      "entropy": 6.394324016571045,
      "epoch": 0.31591365900550555,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004994104856099249,
      "loss": 6.2575,
      "mean_token_accuracy": 0.12981204688549042,
      "num_tokens": 7350923.0,
      "step": 3615
    },
    {
      "entropy": 6.276216506958008,
      "epoch": 0.31635060735821025,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004994082292235476,
      "loss": 6.1597,
      "mean_token_accuracy": 0.14145579785108567,
      "num_tokens": 7360121.0,
      "step": 3620
    },
    {
      "entropy": 6.232924699783325,
      "epoch": 0.316787555710915,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004994059685329027,
      "loss": 6.254,
      "mean_token_accuracy": 0.13218581676483154,
      "num_tokens": 7370823.0,
      "step": 3625
    },
    {
      "entropy": 6.348323106765747,
      "epoch": 0.3172245040636197,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000499403703538034,
      "loss": 6.2037,
      "mean_token_accuracy": 0.12940399199724198,
      "num_tokens": 7380102.0,
      "step": 3630
    },
    {
      "entropy": 6.337209701538086,
      "epoch": 0.3176614524163244,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004994014342389845,
      "loss": 6.2138,
      "mean_token_accuracy": 0.13238532990217208,
      "num_tokens": 7389909.0,
      "step": 3635
    },
    {
      "entropy": 6.338866519927978,
      "epoch": 0.3180984007690291,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004993991606357979,
      "loss": 6.2981,
      "mean_token_accuracy": 0.13026088103652,
      "num_tokens": 7401092.0,
      "step": 3640
    },
    {
      "entropy": 6.298194074630738,
      "epoch": 0.3185353491217338,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004993968827285179,
      "loss": 6.1915,
      "mean_token_accuracy": 0.12618000954389572,
      "num_tokens": 7412155.0,
      "step": 3645
    },
    {
      "entropy": 6.267037105560303,
      "epoch": 0.3189722974744385,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004993946005171881,
      "loss": 6.1407,
      "mean_token_accuracy": 0.14138761162757874,
      "num_tokens": 7421981.0,
      "step": 3650
    },
    {
      "entropy": 6.311609172821045,
      "epoch": 0.31940924582714325,
      "grad_norm": 2.125,
      "learning_rate": 0.0004993923140018523,
      "loss": 6.1799,
      "mean_token_accuracy": 0.13153257742524146,
      "num_tokens": 7432298.0,
      "step": 3655
    },
    {
      "entropy": 6.261382436752319,
      "epoch": 0.31984619417984794,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004993900231825542,
      "loss": 6.2281,
      "mean_token_accuracy": 0.13518029749393462,
      "num_tokens": 7442575.0,
      "step": 3660
    },
    {
      "entropy": 6.312797546386719,
      "epoch": 0.32028314253255263,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004993877280593381,
      "loss": 6.2169,
      "mean_token_accuracy": 0.13454039469361306,
      "num_tokens": 7453072.0,
      "step": 3665
    },
    {
      "entropy": 6.33916826248169,
      "epoch": 0.3207200908852574,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004993854286322477,
      "loss": 6.1624,
      "mean_token_accuracy": 0.13830483630299567,
      "num_tokens": 7463173.0,
      "step": 3670
    },
    {
      "entropy": 6.2755554676055905,
      "epoch": 0.32115703923796207,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004993831249013272,
      "loss": 6.271,
      "mean_token_accuracy": 0.13103101775050163,
      "num_tokens": 7473751.0,
      "step": 3675
    },
    {
      "entropy": 6.340389108657837,
      "epoch": 0.32159398759066676,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004993808168666207,
      "loss": 6.0924,
      "mean_token_accuracy": 0.14098680019378662,
      "num_tokens": 7483233.0,
      "step": 3680
    },
    {
      "entropy": 6.3110268115997314,
      "epoch": 0.3220309359433715,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004993785045281726,
      "loss": 6.1966,
      "mean_token_accuracy": 0.12528955116868018,
      "num_tokens": 7493152.0,
      "step": 3685
    },
    {
      "entropy": 6.309295558929444,
      "epoch": 0.3224678842960762,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004993761878860272,
      "loss": 6.1858,
      "mean_token_accuracy": 0.13619687259197236,
      "num_tokens": 7502463.0,
      "step": 3690
    },
    {
      "entropy": 6.379939603805542,
      "epoch": 0.3229048326487809,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000499373866940229,
      "loss": 6.235,
      "mean_token_accuracy": 0.1337234690785408,
      "num_tokens": 7512301.0,
      "step": 3695
    },
    {
      "entropy": 6.257325315475464,
      "epoch": 0.32334178100148564,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004993715416908225,
      "loss": 6.2947,
      "mean_token_accuracy": 0.13028813749551774,
      "num_tokens": 7522591.0,
      "step": 3700
    },
    {
      "entropy": 6.353185701370239,
      "epoch": 0.32377872935419033,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004993692121378521,
      "loss": 6.2691,
      "mean_token_accuracy": 0.13103109523653983,
      "num_tokens": 7532400.0,
      "step": 3705
    },
    {
      "entropy": 6.37238621711731,
      "epoch": 0.324215677706895,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004993668782813629,
      "loss": 6.1763,
      "mean_token_accuracy": 0.13341537714004517,
      "num_tokens": 7542163.0,
      "step": 3710
    },
    {
      "entropy": 6.2288731098175045,
      "epoch": 0.32465262605959977,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004993645401213992,
      "loss": 6.1762,
      "mean_token_accuracy": 0.13410028740763663,
      "num_tokens": 7551638.0,
      "step": 3715
    },
    {
      "entropy": 6.326100063323975,
      "epoch": 0.32508957441230446,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004993621976580061,
      "loss": 6.2469,
      "mean_token_accuracy": 0.13389818891882896,
      "num_tokens": 7561230.0,
      "step": 3720
    },
    {
      "entropy": 6.35606951713562,
      "epoch": 0.32552652276500915,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004993598508912285,
      "loss": 6.1313,
      "mean_token_accuracy": 0.13896814957261086,
      "num_tokens": 7570962.0,
      "step": 3725
    },
    {
      "entropy": 6.1932018280029295,
      "epoch": 0.3259634711177139,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004993574998211114,
      "loss": 6.3029,
      "mean_token_accuracy": 0.13531308993697166,
      "num_tokens": 7581454.0,
      "step": 3730
    },
    {
      "entropy": 6.292673969268799,
      "epoch": 0.3264004194704186,
      "grad_norm": 1.25,
      "learning_rate": 0.0004993551444476998,
      "loss": 6.1783,
      "mean_token_accuracy": 0.13669705763459206,
      "num_tokens": 7591497.0,
      "step": 3735
    },
    {
      "entropy": 6.3870453357696535,
      "epoch": 0.3268373678231233,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004993527847710392,
      "loss": 6.197,
      "mean_token_accuracy": 0.1389164596796036,
      "num_tokens": 7600719.0,
      "step": 3740
    },
    {
      "entropy": 6.305917978286743,
      "epoch": 0.32727431617582803,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004993504207911744,
      "loss": 6.2189,
      "mean_token_accuracy": 0.13561010509729385,
      "num_tokens": 7610550.0,
      "step": 3745
    },
    {
      "entropy": 6.306847190856933,
      "epoch": 0.3277112645285327,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004993480525081511,
      "loss": 6.2134,
      "mean_token_accuracy": 0.13858453631401063,
      "num_tokens": 7620472.0,
      "step": 3750
    },
    {
      "entropy": 6.393995046615601,
      "epoch": 0.3281482128812374,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004993456799220145,
      "loss": 6.1824,
      "mean_token_accuracy": 0.1334797091782093,
      "num_tokens": 7630354.0,
      "step": 3755
    },
    {
      "entropy": 6.28503041267395,
      "epoch": 0.32858516123394216,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004993433030328103,
      "loss": 6.1979,
      "mean_token_accuracy": 0.13771358877420425,
      "num_tokens": 7639689.0,
      "step": 3760
    },
    {
      "entropy": 6.265411520004273,
      "epoch": 0.32902210958664685,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004993409218405839,
      "loss": 6.2332,
      "mean_token_accuracy": 0.12891449257731438,
      "num_tokens": 7649319.0,
      "step": 3765
    },
    {
      "entropy": 6.39034948348999,
      "epoch": 0.3294590579393516,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004993385363453812,
      "loss": 6.2413,
      "mean_token_accuracy": 0.13170438036322593,
      "num_tokens": 7659008.0,
      "step": 3770
    },
    {
      "entropy": 6.33780426979065,
      "epoch": 0.3298960062920563,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004993361465472477,
      "loss": 6.1893,
      "mean_token_accuracy": 0.13954115957021712,
      "num_tokens": 7669064.0,
      "step": 3775
    },
    {
      "entropy": 6.2886542797088625,
      "epoch": 0.330332954644761,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004993337524462295,
      "loss": 6.2472,
      "mean_token_accuracy": 0.13983521163463591,
      "num_tokens": 7679122.0,
      "step": 3780
    },
    {
      "entropy": 6.251172590255737,
      "epoch": 0.3307699029974657,
      "grad_norm": 4.5,
      "learning_rate": 0.0004993313540423722,
      "loss": 6.2585,
      "mean_token_accuracy": 0.1355503998696804,
      "num_tokens": 7688373.0,
      "step": 3785
    },
    {
      "entropy": 6.36784176826477,
      "epoch": 0.3312068513501704,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004993289513357221,
      "loss": 6.1903,
      "mean_token_accuracy": 0.1309509187936783,
      "num_tokens": 7697879.0,
      "step": 3790
    },
    {
      "entropy": 6.317149496078491,
      "epoch": 0.3316437997028751,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004993265443263251,
      "loss": 6.2204,
      "mean_token_accuracy": 0.13057684674859046,
      "num_tokens": 7708104.0,
      "step": 3795
    },
    {
      "entropy": 6.325311613082886,
      "epoch": 0.33208074805557986,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004993241330142274,
      "loss": 6.199,
      "mean_token_accuracy": 0.14135780110955237,
      "num_tokens": 7717626.0,
      "step": 3800
    },
    {
      "entropy": 6.313494300842285,
      "epoch": 0.33251769640828455,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004993217173994754,
      "loss": 6.235,
      "mean_token_accuracy": 0.13783995285630227,
      "num_tokens": 7728221.0,
      "step": 3805
    },
    {
      "entropy": 6.333722019195557,
      "epoch": 0.33295464476098924,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004993192974821154,
      "loss": 6.2307,
      "mean_token_accuracy": 0.13345992192626,
      "num_tokens": 7739083.0,
      "step": 3810
    },
    {
      "entropy": 6.377281284332275,
      "epoch": 0.333391593113694,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004993168732621935,
      "loss": 6.2932,
      "mean_token_accuracy": 0.12718051746487619,
      "num_tokens": 7748573.0,
      "step": 3815
    },
    {
      "entropy": 6.328021669387818,
      "epoch": 0.3338285414663987,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004993144447397565,
      "loss": 6.2101,
      "mean_token_accuracy": 0.13788711950182914,
      "num_tokens": 7758741.0,
      "step": 3820
    },
    {
      "entropy": 6.275328159332275,
      "epoch": 0.33426548981910337,
      "grad_norm": 1.53125,
      "learning_rate": 0.000499312011914851,
      "loss": 6.2494,
      "mean_token_accuracy": 0.13237224668264388,
      "num_tokens": 7768504.0,
      "step": 3825
    },
    {
      "entropy": 6.302463436126709,
      "epoch": 0.3347024381718081,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004993095747875234,
      "loss": 6.3879,
      "mean_token_accuracy": 0.13009582236409187,
      "num_tokens": 7779017.0,
      "step": 3830
    },
    {
      "entropy": 6.339192819595337,
      "epoch": 0.3351393865245128,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004993071333578207,
      "loss": 6.1821,
      "mean_token_accuracy": 0.1370094619691372,
      "num_tokens": 7787630.0,
      "step": 3835
    },
    {
      "entropy": 6.278088474273682,
      "epoch": 0.3355763348772175,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004993046876257898,
      "loss": 6.2131,
      "mean_token_accuracy": 0.13482895866036415,
      "num_tokens": 7797105.0,
      "step": 3840
    },
    {
      "entropy": 6.311618185043335,
      "epoch": 0.33601328322992224,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004993022375914773,
      "loss": 6.1937,
      "mean_token_accuracy": 0.13905783295631408,
      "num_tokens": 7807070.0,
      "step": 3845
    },
    {
      "entropy": 6.389101219177246,
      "epoch": 0.33645023158262694,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004992997832549303,
      "loss": 6.2253,
      "mean_token_accuracy": 0.1325555205345154,
      "num_tokens": 7818090.0,
      "step": 3850
    },
    {
      "entropy": 6.221946573257446,
      "epoch": 0.3368871799353316,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004992973246161961,
      "loss": 6.1764,
      "mean_token_accuracy": 0.13771116584539414,
      "num_tokens": 7827562.0,
      "step": 3855
    },
    {
      "entropy": 6.292095041275024,
      "epoch": 0.3373241282880364,
      "grad_norm": 1.75,
      "learning_rate": 0.0004992948616753215,
      "loss": 6.2093,
      "mean_token_accuracy": 0.1278516322374344,
      "num_tokens": 7837581.0,
      "step": 3860
    },
    {
      "entropy": 6.394981861114502,
      "epoch": 0.33776107664074106,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004992923944323541,
      "loss": 6.2634,
      "mean_token_accuracy": 0.13501928597688675,
      "num_tokens": 7848729.0,
      "step": 3865
    },
    {
      "entropy": 6.310138988494873,
      "epoch": 0.33819802499344576,
      "grad_norm": 1.5,
      "learning_rate": 0.000499289922887341,
      "loss": 6.2011,
      "mean_token_accuracy": 0.13485699966549874,
      "num_tokens": 7857379.0,
      "step": 3870
    },
    {
      "entropy": 6.325172996520996,
      "epoch": 0.3386349733461505,
      "grad_norm": 2.0,
      "learning_rate": 0.0004992874470403296,
      "loss": 6.1917,
      "mean_token_accuracy": 0.14079806953668594,
      "num_tokens": 7867200.0,
      "step": 3875
    },
    {
      "entropy": 6.311602544784546,
      "epoch": 0.3390719216988552,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004992849668913674,
      "loss": 6.1704,
      "mean_token_accuracy": 0.1355910912156105,
      "num_tokens": 7876709.0,
      "step": 3880
    },
    {
      "entropy": 6.3373613357543945,
      "epoch": 0.3395088700515599,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004992824824405021,
      "loss": 6.2633,
      "mean_token_accuracy": 0.13315544426441192,
      "num_tokens": 7886256.0,
      "step": 3885
    },
    {
      "entropy": 6.301959228515625,
      "epoch": 0.33994581840426463,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004992799936877813,
      "loss": 6.188,
      "mean_token_accuracy": 0.13398601338267327,
      "num_tokens": 7896497.0,
      "step": 3890
    },
    {
      "entropy": 6.371589469909668,
      "epoch": 0.3403827667569693,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004992775006332526,
      "loss": 6.207,
      "mean_token_accuracy": 0.13721492886543274,
      "num_tokens": 7906608.0,
      "step": 3895
    },
    {
      "entropy": 6.317071485519409,
      "epoch": 0.340819715109674,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000499275003276964,
      "loss": 6.2863,
      "mean_token_accuracy": 0.1299916237592697,
      "num_tokens": 7916359.0,
      "step": 3900
    },
    {
      "entropy": 6.236544132232666,
      "epoch": 0.34125666346237876,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004992725016189632,
      "loss": 6.1603,
      "mean_token_accuracy": 0.13543331772089004,
      "num_tokens": 7925800.0,
      "step": 3905
    },
    {
      "entropy": 6.360721111297607,
      "epoch": 0.34169361181508345,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004992699956592984,
      "loss": 6.2488,
      "mean_token_accuracy": 0.1296587198972702,
      "num_tokens": 7935424.0,
      "step": 3910
    },
    {
      "entropy": 6.361935234069824,
      "epoch": 0.34213056016778814,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004992674853980175,
      "loss": 6.2554,
      "mean_token_accuracy": 0.13166892752051354,
      "num_tokens": 7945766.0,
      "step": 3915
    },
    {
      "entropy": 6.368907356262207,
      "epoch": 0.3425675085204929,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004992649708351689,
      "loss": 6.2776,
      "mean_token_accuracy": 0.1250272959470749,
      "num_tokens": 7955622.0,
      "step": 3920
    },
    {
      "entropy": 6.329561471939087,
      "epoch": 0.3430044568731976,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004992624519708004,
      "loss": 6.2914,
      "mean_token_accuracy": 0.12758758887648583,
      "num_tokens": 7965750.0,
      "step": 3925
    },
    {
      "entropy": 6.343333435058594,
      "epoch": 0.3434414052259023,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004992599288049607,
      "loss": 6.2406,
      "mean_token_accuracy": 0.13924501463770866,
      "num_tokens": 7977568.0,
      "step": 3930
    },
    {
      "entropy": 6.363393783569336,
      "epoch": 0.343878353578607,
      "grad_norm": 1.625,
      "learning_rate": 0.000499257401337698,
      "loss": 6.2872,
      "mean_token_accuracy": 0.12728462964296341,
      "num_tokens": 7988411.0,
      "step": 3935
    },
    {
      "entropy": 6.28168420791626,
      "epoch": 0.3443153019313117,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004992548695690608,
      "loss": 6.1387,
      "mean_token_accuracy": 0.14234383925795555,
      "num_tokens": 7998315.0,
      "step": 3940
    },
    {
      "entropy": 6.276813268661499,
      "epoch": 0.3447522502840164,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004992523334990977,
      "loss": 6.2068,
      "mean_token_accuracy": 0.13106251060962676,
      "num_tokens": 8007133.0,
      "step": 3945
    },
    {
      "entropy": 6.272677087783814,
      "epoch": 0.34518919863672115,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004992497931278574,
      "loss": 6.1528,
      "mean_token_accuracy": 0.14113296419382096,
      "num_tokens": 8016882.0,
      "step": 3950
    },
    {
      "entropy": 6.260069417953491,
      "epoch": 0.34562614698942584,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004992472484553886,
      "loss": 6.1671,
      "mean_token_accuracy": 0.13664042428135872,
      "num_tokens": 8026893.0,
      "step": 3955
    },
    {
      "entropy": 6.308275556564331,
      "epoch": 0.34606309534213053,
      "grad_norm": 1.859375,
      "learning_rate": 0.00049924469948174,
      "loss": 6.1944,
      "mean_token_accuracy": 0.13003704249858855,
      "num_tokens": 8036240.0,
      "step": 3960
    },
    {
      "entropy": 6.345671558380127,
      "epoch": 0.3465000436948353,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004992421462069606,
      "loss": 6.1757,
      "mean_token_accuracy": 0.13144434317946435,
      "num_tokens": 8046086.0,
      "step": 3965
    },
    {
      "entropy": 6.240842390060425,
      "epoch": 0.34693699204753997,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004992395886310993,
      "loss": 6.1895,
      "mean_token_accuracy": 0.1421689584851265,
      "num_tokens": 8056294.0,
      "step": 3970
    },
    {
      "entropy": 6.286834669113159,
      "epoch": 0.3473739404002447,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004992370267542052,
      "loss": 6.2156,
      "mean_token_accuracy": 0.1315187208354473,
      "num_tokens": 8067329.0,
      "step": 3975
    },
    {
      "entropy": 6.3432018756866455,
      "epoch": 0.3478108887529494,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004992344605763275,
      "loss": 6.186,
      "mean_token_accuracy": 0.1398385003209114,
      "num_tokens": 8077106.0,
      "step": 3980
    },
    {
      "entropy": 6.333979988098145,
      "epoch": 0.3482478371056541,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004992318900975153,
      "loss": 6.101,
      "mean_token_accuracy": 0.13900325149297715,
      "num_tokens": 8087766.0,
      "step": 3985
    },
    {
      "entropy": 6.207983016967773,
      "epoch": 0.34868478545835885,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004992293153178179,
      "loss": 6.1902,
      "mean_token_accuracy": 0.14341192990541457,
      "num_tokens": 8098649.0,
      "step": 3990
    },
    {
      "entropy": 6.323047065734864,
      "epoch": 0.34912173381106354,
      "grad_norm": 1.5,
      "learning_rate": 0.0004992267362372848,
      "loss": 6.2092,
      "mean_token_accuracy": 0.13100808262825012,
      "num_tokens": 8109237.0,
      "step": 3995
    },
    {
      "entropy": 6.2751020908355715,
      "epoch": 0.34955868216376823,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004992241528559653,
      "loss": 6.208,
      "mean_token_accuracy": 0.13489847853779793,
      "num_tokens": 8119731.0,
      "step": 4000
    },
    {
      "entropy": 6.318090724945068,
      "epoch": 0.349995630516473,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004992215651739092,
      "loss": 6.2467,
      "mean_token_accuracy": 0.12376242280006408,
      "num_tokens": 8129448.0,
      "step": 4005
    },
    {
      "entropy": 6.28632116317749,
      "epoch": 0.35043257886917767,
      "grad_norm": 1.609375,
      "learning_rate": 0.000499218973191166,
      "loss": 6.1289,
      "mean_token_accuracy": 0.13718134835362433,
      "num_tokens": 8138345.0,
      "step": 4010
    },
    {
      "entropy": 6.301534938812256,
      "epoch": 0.35086952722188236,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004992163769077853,
      "loss": 6.1927,
      "mean_token_accuracy": 0.1334017850458622,
      "num_tokens": 8148380.0,
      "step": 4015
    },
    {
      "entropy": 6.319975233078003,
      "epoch": 0.3513064755745871,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000499213776323817,
      "loss": 6.1918,
      "mean_token_accuracy": 0.14010001271963118,
      "num_tokens": 8159359.0,
      "step": 4020
    },
    {
      "entropy": 6.361384391784668,
      "epoch": 0.3517434239272918,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000499211171439311,
      "loss": 6.2777,
      "mean_token_accuracy": 0.13136567771434784,
      "num_tokens": 8168586.0,
      "step": 4025
    },
    {
      "entropy": 6.306084442138672,
      "epoch": 0.3521803722799965,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004992085622543172,
      "loss": 6.1986,
      "mean_token_accuracy": 0.1354872040450573,
      "num_tokens": 8178836.0,
      "step": 4030
    },
    {
      "entropy": 6.247843790054321,
      "epoch": 0.35261732063270124,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004992059487688859,
      "loss": 6.1853,
      "mean_token_accuracy": 0.13473303467035294,
      "num_tokens": 8188933.0,
      "step": 4035
    },
    {
      "entropy": 6.2794787883758545,
      "epoch": 0.35305426898540593,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004992033309830668,
      "loss": 6.1864,
      "mean_token_accuracy": 0.1396368734538555,
      "num_tokens": 8199366.0,
      "step": 4040
    },
    {
      "entropy": 6.355959844589234,
      "epoch": 0.3534912173381106,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004992007088969105,
      "loss": 6.1731,
      "mean_token_accuracy": 0.1387977495789528,
      "num_tokens": 8209806.0,
      "step": 4045
    },
    {
      "entropy": 6.367386245727539,
      "epoch": 0.35392816569081537,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000499198082510467,
      "loss": 6.2528,
      "mean_token_accuracy": 0.12820714637637137,
      "num_tokens": 8219367.0,
      "step": 4050
    },
    {
      "entropy": 6.205999374389648,
      "epoch": 0.35436511404352006,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004991954518237868,
      "loss": 6.1457,
      "mean_token_accuracy": 0.13604540154337882,
      "num_tokens": 8228918.0,
      "step": 4055
    },
    {
      "entropy": 6.251398229598999,
      "epoch": 0.35480206239622475,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004991928168369205,
      "loss": 6.2326,
      "mean_token_accuracy": 0.13432048931717872,
      "num_tokens": 8239380.0,
      "step": 4060
    },
    {
      "entropy": 6.440735387802124,
      "epoch": 0.3552390107489295,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004991901775499183,
      "loss": 6.299,
      "mean_token_accuracy": 0.12985948398709296,
      "num_tokens": 8249131.0,
      "step": 4065
    },
    {
      "entropy": 6.358709192276001,
      "epoch": 0.3556759591016342,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004991875339628311,
      "loss": 6.1916,
      "mean_token_accuracy": 0.13281939700245857,
      "num_tokens": 8258550.0,
      "step": 4070
    },
    {
      "entropy": 6.2490339279174805,
      "epoch": 0.3561129074543389,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004991848860757095,
      "loss": 6.21,
      "mean_token_accuracy": 0.1398457445204258,
      "num_tokens": 8269089.0,
      "step": 4075
    },
    {
      "entropy": 6.319385766983032,
      "epoch": 0.3565498558070436,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004991822338886044,
      "loss": 6.1852,
      "mean_token_accuracy": 0.13303097859025,
      "num_tokens": 8279735.0,
      "step": 4080
    },
    {
      "entropy": 6.335915517807007,
      "epoch": 0.3569868041597483,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004991795774015665,
      "loss": 6.139,
      "mean_token_accuracy": 0.13977616727352143,
      "num_tokens": 8289693.0,
      "step": 4085
    },
    {
      "entropy": 6.3235859870910645,
      "epoch": 0.357423752512453,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004991769166146469,
      "loss": 6.2341,
      "mean_token_accuracy": 0.13487215265631675,
      "num_tokens": 8300909.0,
      "step": 4090
    },
    {
      "entropy": 6.265295362472534,
      "epoch": 0.35786070086515775,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004991742515278965,
      "loss": 6.2075,
      "mean_token_accuracy": 0.1362832248210907,
      "num_tokens": 8311479.0,
      "step": 4095
    },
    {
      "entropy": 6.325096845626831,
      "epoch": 0.35829764921786245,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004991715821413665,
      "loss": 6.2181,
      "mean_token_accuracy": 0.13479658886790274,
      "num_tokens": 8321267.0,
      "step": 4100
    },
    {
      "entropy": 6.284675645828247,
      "epoch": 0.35873459757056714,
      "grad_norm": 2.0,
      "learning_rate": 0.0004991689084551082,
      "loss": 6.1695,
      "mean_token_accuracy": 0.13774141371250154,
      "num_tokens": 8331445.0,
      "step": 4105
    },
    {
      "entropy": 6.320082473754883,
      "epoch": 0.3591715459232719,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004991662304691727,
      "loss": 6.2593,
      "mean_token_accuracy": 0.12908197790384293,
      "num_tokens": 8340519.0,
      "step": 4110
    },
    {
      "entropy": 6.339802503585815,
      "epoch": 0.3596084942759766,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004991635481836114,
      "loss": 6.2162,
      "mean_token_accuracy": 0.13142214938998223,
      "num_tokens": 8350467.0,
      "step": 4115
    },
    {
      "entropy": 6.34062819480896,
      "epoch": 0.36004544262868127,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004991608615984757,
      "loss": 6.1853,
      "mean_token_accuracy": 0.13231182098388672,
      "num_tokens": 8360067.0,
      "step": 4120
    },
    {
      "entropy": 6.322159385681152,
      "epoch": 0.360482390981386,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004991581707138175,
      "loss": 6.2469,
      "mean_token_accuracy": 0.13345297202467918,
      "num_tokens": 8370509.0,
      "step": 4125
    },
    {
      "entropy": 6.279163980484009,
      "epoch": 0.3609193393340907,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991554755296878,
      "loss": 6.2007,
      "mean_token_accuracy": 0.13149990066885947,
      "num_tokens": 8381489.0,
      "step": 4130
    },
    {
      "entropy": 6.284092140197754,
      "epoch": 0.3613562876867954,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004991527760461389,
      "loss": 6.2099,
      "mean_token_accuracy": 0.1364233486354351,
      "num_tokens": 8392287.0,
      "step": 4135
    },
    {
      "entropy": 6.318317317962647,
      "epoch": 0.36179323603950014,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004991500722632223,
      "loss": 6.1923,
      "mean_token_accuracy": 0.13617967516183854,
      "num_tokens": 8402433.0,
      "step": 4140
    },
    {
      "entropy": 6.276598310470581,
      "epoch": 0.36223018439220483,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004991473641809896,
      "loss": 6.2349,
      "mean_token_accuracy": 0.13364490419626235,
      "num_tokens": 8412944.0,
      "step": 4145
    },
    {
      "entropy": 6.318027210235596,
      "epoch": 0.3626671327449095,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004991446517994932,
      "loss": 6.1439,
      "mean_token_accuracy": 0.13547627851366997,
      "num_tokens": 8422511.0,
      "step": 4150
    },
    {
      "entropy": 6.2953328609466555,
      "epoch": 0.3631040810976143,
      "grad_norm": 1.875,
      "learning_rate": 0.0004991419351187849,
      "loss": 6.1748,
      "mean_token_accuracy": 0.13981883525848388,
      "num_tokens": 8433094.0,
      "step": 4155
    },
    {
      "entropy": 6.268534135818482,
      "epoch": 0.36354102945031896,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004991392141389169,
      "loss": 6.2753,
      "mean_token_accuracy": 0.1309742160141468,
      "num_tokens": 8444430.0,
      "step": 4160
    },
    {
      "entropy": 6.328505039215088,
      "epoch": 0.36397797780302366,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004991364888599412,
      "loss": 6.3247,
      "mean_token_accuracy": 0.12793881446123123,
      "num_tokens": 8455171.0,
      "step": 4165
    },
    {
      "entropy": 6.3519635677337645,
      "epoch": 0.3644149261557284,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004991337592819103,
      "loss": 6.1376,
      "mean_token_accuracy": 0.13894972056150437,
      "num_tokens": 8464724.0,
      "step": 4170
    },
    {
      "entropy": 6.352320718765259,
      "epoch": 0.3648518745084331,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004991310254048763,
      "loss": 6.1937,
      "mean_token_accuracy": 0.13326725661754607,
      "num_tokens": 8474732.0,
      "step": 4175
    },
    {
      "entropy": 6.222937059402466,
      "epoch": 0.36528882286113784,
      "grad_norm": 1.5,
      "learning_rate": 0.0004991282872288919,
      "loss": 6.1661,
      "mean_token_accuracy": 0.13578586652874947,
      "num_tokens": 8484731.0,
      "step": 4180
    },
    {
      "entropy": 6.268366861343384,
      "epoch": 0.36572577121384253,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004991255447540095,
      "loss": 6.1492,
      "mean_token_accuracy": 0.13649590015411378,
      "num_tokens": 8494509.0,
      "step": 4185
    },
    {
      "entropy": 6.322968292236328,
      "epoch": 0.3661627195665472,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004991227979802817,
      "loss": 6.2282,
      "mean_token_accuracy": 0.1378632016479969,
      "num_tokens": 8504260.0,
      "step": 4190
    },
    {
      "entropy": 6.373710203170776,
      "epoch": 0.36659966791925197,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004991200469077612,
      "loss": 6.2754,
      "mean_token_accuracy": 0.1293950892984867,
      "num_tokens": 8514222.0,
      "step": 4195
    },
    {
      "entropy": 6.310275459289551,
      "epoch": 0.36703661627195666,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004991172915365008,
      "loss": 6.2204,
      "mean_token_accuracy": 0.12950511202216147,
      "num_tokens": 8523979.0,
      "step": 4200
    },
    {
      "entropy": 6.316265487670899,
      "epoch": 0.36747356462466135,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004991145318665532,
      "loss": 6.1535,
      "mean_token_accuracy": 0.14113664031028747,
      "num_tokens": 8533516.0,
      "step": 4205
    },
    {
      "entropy": 6.336113405227661,
      "epoch": 0.3679105129773661,
      "grad_norm": 3.25,
      "learning_rate": 0.0004991117678979716,
      "loss": 6.2776,
      "mean_token_accuracy": 0.12696770057082177,
      "num_tokens": 8544444.0,
      "step": 4210
    },
    {
      "entropy": 6.2604944705963135,
      "epoch": 0.3683474613300708,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004991089996308089,
      "loss": 6.1988,
      "mean_token_accuracy": 0.13401636481285095,
      "num_tokens": 8555571.0,
      "step": 4215
    },
    {
      "entropy": 6.359772682189941,
      "epoch": 0.3687844096827755,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004991062270651179,
      "loss": 6.2705,
      "mean_token_accuracy": 0.12754399403929711,
      "num_tokens": 8566486.0,
      "step": 4220
    },
    {
      "entropy": 6.3696788311004635,
      "epoch": 0.36922135803548023,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004991034502009522,
      "loss": 6.1645,
      "mean_token_accuracy": 0.1367284558713436,
      "num_tokens": 8575324.0,
      "step": 4225
    },
    {
      "entropy": 6.254921817779541,
      "epoch": 0.3696583063881849,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004991006690383649,
      "loss": 6.1641,
      "mean_token_accuracy": 0.13787002265453338,
      "num_tokens": 8586015.0,
      "step": 4230
    },
    {
      "entropy": 6.297842502593994,
      "epoch": 0.3700952547408896,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004990978835774093,
      "loss": 6.1391,
      "mean_token_accuracy": 0.13815010488033294,
      "num_tokens": 8594748.0,
      "step": 4235
    },
    {
      "entropy": 6.2729315757751465,
      "epoch": 0.37053220309359436,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004990950938181389,
      "loss": 6.211,
      "mean_token_accuracy": 0.12907496616244316,
      "num_tokens": 8605805.0,
      "step": 4240
    },
    {
      "entropy": 6.246347951889038,
      "epoch": 0.37096915144629905,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004990922997606071,
      "loss": 6.1785,
      "mean_token_accuracy": 0.13529932647943496,
      "num_tokens": 8616591.0,
      "step": 4245
    },
    {
      "entropy": 6.321091175079346,
      "epoch": 0.37140609979900374,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004990895014048677,
      "loss": 6.2076,
      "mean_token_accuracy": 0.13347845450043677,
      "num_tokens": 8627063.0,
      "step": 4250
    },
    {
      "entropy": 6.415533828735351,
      "epoch": 0.3718430481517085,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004990866987509741,
      "loss": 6.2818,
      "mean_token_accuracy": 0.12526709884405135,
      "num_tokens": 8637041.0,
      "step": 4255
    },
    {
      "entropy": 6.27305645942688,
      "epoch": 0.3722799965044132,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004990838917989802,
      "loss": 6.2054,
      "mean_token_accuracy": 0.1385042130947113,
      "num_tokens": 8647485.0,
      "step": 4260
    },
    {
      "entropy": 6.278818893432617,
      "epoch": 0.37271694485711787,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00049908108054894,
      "loss": 6.1707,
      "mean_token_accuracy": 0.13312357068061828,
      "num_tokens": 8657185.0,
      "step": 4265
    },
    {
      "entropy": 6.30776538848877,
      "epoch": 0.3731538932098226,
      "grad_norm": 1.75,
      "learning_rate": 0.0004990782650009071,
      "loss": 6.1924,
      "mean_token_accuracy": 0.1385415717959404,
      "num_tokens": 8667639.0,
      "step": 4270
    },
    {
      "entropy": 6.320880556106568,
      "epoch": 0.3735908415625273,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004990754451549358,
      "loss": 6.2078,
      "mean_token_accuracy": 0.13354232385754586,
      "num_tokens": 8678003.0,
      "step": 4275
    },
    {
      "entropy": 6.328972673416137,
      "epoch": 0.374027789915232,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004990726210110798,
      "loss": 6.2554,
      "mean_token_accuracy": 0.13083540871739388,
      "num_tokens": 8688837.0,
      "step": 4280
    },
    {
      "entropy": 6.31743278503418,
      "epoch": 0.37446473826793675,
      "grad_norm": 1.75,
      "learning_rate": 0.0004990697925693938,
      "loss": 6.0607,
      "mean_token_accuracy": 0.1439003139734268,
      "num_tokens": 8698395.0,
      "step": 4285
    },
    {
      "entropy": 6.311701154708862,
      "epoch": 0.37490168662064144,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004990669598299316,
      "loss": 6.188,
      "mean_token_accuracy": 0.13384513854980468,
      "num_tokens": 8708031.0,
      "step": 4290
    },
    {
      "entropy": 6.264176940917968,
      "epoch": 0.37533863497334613,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004990641227927478,
      "loss": 6.1554,
      "mean_token_accuracy": 0.14129097983241082,
      "num_tokens": 8717825.0,
      "step": 4295
    },
    {
      "entropy": 6.35695514678955,
      "epoch": 0.3757755833260509,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004990612814578967,
      "loss": 6.2345,
      "mean_token_accuracy": 0.12975349202752112,
      "num_tokens": 8728873.0,
      "step": 4300
    },
    {
      "entropy": 6.237163925170899,
      "epoch": 0.37621253167875557,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004990584358254327,
      "loss": 6.1579,
      "mean_token_accuracy": 0.1332879640161991,
      "num_tokens": 8737917.0,
      "step": 4305
    },
    {
      "entropy": 6.223076629638672,
      "epoch": 0.37664948003146026,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004990555858954107,
      "loss": 6.128,
      "mean_token_accuracy": 0.13721137419342994,
      "num_tokens": 8747466.0,
      "step": 4310
    },
    {
      "entropy": 6.261527442932129,
      "epoch": 0.377086428384165,
      "grad_norm": 1.453125,
      "learning_rate": 0.000499052731667885,
      "loss": 6.1715,
      "mean_token_accuracy": 0.13348276689648628,
      "num_tokens": 8757037.0,
      "step": 4315
    },
    {
      "entropy": 6.295954370498658,
      "epoch": 0.3775233767368697,
      "grad_norm": 1.625,
      "learning_rate": 0.0004990498731429105,
      "loss": 6.1968,
      "mean_token_accuracy": 0.1333417996764183,
      "num_tokens": 8766780.0,
      "step": 4320
    },
    {
      "entropy": 6.292055940628051,
      "epoch": 0.3779603250895744,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004990470103205421,
      "loss": 6.1637,
      "mean_token_accuracy": 0.13308844119310378,
      "num_tokens": 8777063.0,
      "step": 4325
    },
    {
      "entropy": 6.216593027114868,
      "epoch": 0.37839727344227914,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004990441432008347,
      "loss": 6.204,
      "mean_token_accuracy": 0.13550992980599402,
      "num_tokens": 8786479.0,
      "step": 4330
    },
    {
      "entropy": 6.37866063117981,
      "epoch": 0.3788342217949838,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004990412717838431,
      "loss": 6.2101,
      "mean_token_accuracy": 0.13552200198173522,
      "num_tokens": 8795663.0,
      "step": 4335
    },
    {
      "entropy": 6.282534646987915,
      "epoch": 0.3792711701476885,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004990383960696226,
      "loss": 6.1561,
      "mean_token_accuracy": 0.1367136836051941,
      "num_tokens": 8805854.0,
      "step": 4340
    },
    {
      "entropy": 6.236309766769409,
      "epoch": 0.37970811850039327,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004990355160582283,
      "loss": 6.119,
      "mean_token_accuracy": 0.13313476443290712,
      "num_tokens": 8816068.0,
      "step": 4345
    },
    {
      "entropy": 6.224233531951905,
      "epoch": 0.38014506685309796,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004990326317497153,
      "loss": 6.1522,
      "mean_token_accuracy": 0.13832384943962098,
      "num_tokens": 8826301.0,
      "step": 4350
    },
    {
      "entropy": 6.314862632751465,
      "epoch": 0.38058201520580265,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004990297431441391,
      "loss": 6.1756,
      "mean_token_accuracy": 0.1411406196653843,
      "num_tokens": 8837482.0,
      "step": 4355
    },
    {
      "entropy": 6.324439620971679,
      "epoch": 0.3810189635585074,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000499026850241555,
      "loss": 6.2086,
      "mean_token_accuracy": 0.13611030280590058,
      "num_tokens": 8847041.0,
      "step": 4360
    },
    {
      "entropy": 6.2874983787536625,
      "epoch": 0.3814559119112121,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004990239530420185,
      "loss": 6.2079,
      "mean_token_accuracy": 0.13298215717077255,
      "num_tokens": 8856963.0,
      "step": 4365
    },
    {
      "entropy": 6.320013093948364,
      "epoch": 0.3818928602639168,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004990210515455853,
      "loss": 6.1642,
      "mean_token_accuracy": 0.13463006168603897,
      "num_tokens": 8867318.0,
      "step": 4370
    },
    {
      "entropy": 6.29173731803894,
      "epoch": 0.3823298086166215,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004990181457523109,
      "loss": 6.1803,
      "mean_token_accuracy": 0.14037566334009172,
      "num_tokens": 8877634.0,
      "step": 4375
    },
    {
      "entropy": 6.282214641571045,
      "epoch": 0.3827667569693262,
      "grad_norm": 2.0625,
      "learning_rate": 0.000499015235662251,
      "loss": 6.2365,
      "mean_token_accuracy": 0.13639468774199487,
      "num_tokens": 8887708.0,
      "step": 4380
    },
    {
      "entropy": 6.314670038223267,
      "epoch": 0.38320370532203096,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004990123212754616,
      "loss": 6.1379,
      "mean_token_accuracy": 0.13615118563175202,
      "num_tokens": 8897510.0,
      "step": 4385
    },
    {
      "entropy": 6.3385828018188475,
      "epoch": 0.38364065367473565,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004990094025919985,
      "loss": 6.1851,
      "mean_token_accuracy": 0.1292119763791561,
      "num_tokens": 8908282.0,
      "step": 4390
    },
    {
      "entropy": 6.322610807418823,
      "epoch": 0.38407760202744035,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004990064796119175,
      "loss": 6.2251,
      "mean_token_accuracy": 0.1304896928369999,
      "num_tokens": 8918332.0,
      "step": 4395
    },
    {
      "entropy": 6.242078018188477,
      "epoch": 0.3845145503801451,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004990035523352751,
      "loss": 6.1793,
      "mean_token_accuracy": 0.1364331990480423,
      "num_tokens": 8929138.0,
      "step": 4400
    },
    {
      "entropy": 6.288182020187378,
      "epoch": 0.3849514987328498,
      "grad_norm": 1.46875,
      "learning_rate": 0.000499000620762127,
      "loss": 6.232,
      "mean_token_accuracy": 0.130785021930933,
      "num_tokens": 8940871.0,
      "step": 4405
    },
    {
      "entropy": 6.344573259353638,
      "epoch": 0.3853884470855545,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004989976848925297,
      "loss": 6.1573,
      "mean_token_accuracy": 0.1379668101668358,
      "num_tokens": 8951199.0,
      "step": 4410
    },
    {
      "entropy": 6.306144666671753,
      "epoch": 0.3858253954382592,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004989947447265394,
      "loss": 6.3012,
      "mean_token_accuracy": 0.1280881054699421,
      "num_tokens": 8961482.0,
      "step": 4415
    },
    {
      "entropy": 6.357047700881958,
      "epoch": 0.3862623437909639,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004989918002642126,
      "loss": 6.1941,
      "mean_token_accuracy": 0.13600103184580803,
      "num_tokens": 8971884.0,
      "step": 4420
    },
    {
      "entropy": 6.260739088058472,
      "epoch": 0.3866992921436686,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004989888515056057,
      "loss": 6.1486,
      "mean_token_accuracy": 0.13692375347018243,
      "num_tokens": 8982086.0,
      "step": 4425
    },
    {
      "entropy": 6.221055173873902,
      "epoch": 0.38713624049637335,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004989858984507752,
      "loss": 6.2043,
      "mean_token_accuracy": 0.13385815545916557,
      "num_tokens": 8991555.0,
      "step": 4430
    },
    {
      "entropy": 6.318227338790893,
      "epoch": 0.38757318884907804,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004989829410997778,
      "loss": 6.1941,
      "mean_token_accuracy": 0.13073932528495788,
      "num_tokens": 9002459.0,
      "step": 4435
    },
    {
      "entropy": 6.283407688140869,
      "epoch": 0.38801013720178273,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004989799794526702,
      "loss": 6.2485,
      "mean_token_accuracy": 0.13009282052516938,
      "num_tokens": 9012514.0,
      "step": 4440
    },
    {
      "entropy": 6.27811188697815,
      "epoch": 0.3884470855544875,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004989770135095093,
      "loss": 6.147,
      "mean_token_accuracy": 0.13882900923490524,
      "num_tokens": 9022636.0,
      "step": 4445
    },
    {
      "entropy": 6.311051321029663,
      "epoch": 0.3888840339071922,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004989740432703519,
      "loss": 6.1784,
      "mean_token_accuracy": 0.13938639983534812,
      "num_tokens": 9032868.0,
      "step": 4450
    },
    {
      "entropy": 6.3481199741363525,
      "epoch": 0.38932098225989686,
      "grad_norm": 1.859375,
      "learning_rate": 0.000498971068735255,
      "loss": 6.1694,
      "mean_token_accuracy": 0.13330444023013116,
      "num_tokens": 9043057.0,
      "step": 4455
    },
    {
      "entropy": 6.2306193828582765,
      "epoch": 0.3897579306126016,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004989680899042756,
      "loss": 6.1749,
      "mean_token_accuracy": 0.1374656155705452,
      "num_tokens": 9053179.0,
      "step": 4460
    },
    {
      "entropy": 6.302839851379394,
      "epoch": 0.3901948789653063,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004989651067774708,
      "loss": 6.1927,
      "mean_token_accuracy": 0.13686879202723504,
      "num_tokens": 9063112.0,
      "step": 4465
    },
    {
      "entropy": 6.364594268798828,
      "epoch": 0.390631827318011,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004989621193548981,
      "loss": 6.2015,
      "mean_token_accuracy": 0.13447140827775,
      "num_tokens": 9073512.0,
      "step": 4470
    },
    {
      "entropy": 6.2928688526153564,
      "epoch": 0.39106877567071574,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004989591276366145,
      "loss": 6.1931,
      "mean_token_accuracy": 0.12548629194498062,
      "num_tokens": 9083369.0,
      "step": 4475
    },
    {
      "entropy": 6.308518123626709,
      "epoch": 0.39150572402342043,
      "grad_norm": 2.25,
      "learning_rate": 0.0004989561316226774,
      "loss": 6.1756,
      "mean_token_accuracy": 0.13745871260762216,
      "num_tokens": 9092814.0,
      "step": 4480
    },
    {
      "entropy": 6.209445142745972,
      "epoch": 0.3919426723761251,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004989531313131444,
      "loss": 6.1086,
      "mean_token_accuracy": 0.1465831182897091,
      "num_tokens": 9102816.0,
      "step": 4485
    },
    {
      "entropy": 6.265080356597901,
      "epoch": 0.39237962072882987,
      "grad_norm": 2.140625,
      "learning_rate": 0.000498950126708073,
      "loss": 6.2491,
      "mean_token_accuracy": 0.1286209739744663,
      "num_tokens": 9114470.0,
      "step": 4490
    },
    {
      "entropy": 6.401240396499634,
      "epoch": 0.39281656908153456,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004989471178075208,
      "loss": 6.2179,
      "mean_token_accuracy": 0.13008468151092528,
      "num_tokens": 9125604.0,
      "step": 4495
    },
    {
      "entropy": 6.2590978145599365,
      "epoch": 0.39325351743423925,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004989441046115456,
      "loss": 6.1499,
      "mean_token_accuracy": 0.1401989698410034,
      "num_tokens": 9135088.0,
      "step": 4500
    },
    {
      "entropy": 6.247618913650513,
      "epoch": 0.393690465786944,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004989410871202051,
      "loss": 6.2266,
      "mean_token_accuracy": 0.13349834457039833,
      "num_tokens": 9145302.0,
      "step": 4505
    },
    {
      "entropy": 6.270593976974487,
      "epoch": 0.3941274141396487,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004989380653335571,
      "loss": 6.1515,
      "mean_token_accuracy": 0.13583647683262826,
      "num_tokens": 9155104.0,
      "step": 4510
    },
    {
      "entropy": 6.252223634719849,
      "epoch": 0.3945643624923534,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004989350392516598,
      "loss": 6.0822,
      "mean_token_accuracy": 0.1374653048813343,
      "num_tokens": 9165055.0,
      "step": 4515
    },
    {
      "entropy": 6.263556861877442,
      "epoch": 0.39500131084505813,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004989320088745711,
      "loss": 6.1041,
      "mean_token_accuracy": 0.1383694276213646,
      "num_tokens": 9175305.0,
      "step": 4520
    },
    {
      "entropy": 6.23392162322998,
      "epoch": 0.3954382591977628,
      "grad_norm": 2.140625,
      "learning_rate": 0.000498928974202349,
      "loss": 6.1748,
      "mean_token_accuracy": 0.13336197286844254,
      "num_tokens": 9185153.0,
      "step": 4525
    },
    {
      "entropy": 6.2877270698547365,
      "epoch": 0.3958752075504675,
      "grad_norm": 1.640625,
      "learning_rate": 0.000498925935235052,
      "loss": 6.1521,
      "mean_token_accuracy": 0.13427923619747162,
      "num_tokens": 9195385.0,
      "step": 4530
    },
    {
      "entropy": 6.242312908172607,
      "epoch": 0.39631215590317226,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004989228919727382,
      "loss": 6.1075,
      "mean_token_accuracy": 0.13958708792924882,
      "num_tokens": 9205009.0,
      "step": 4535
    },
    {
      "entropy": 6.280454111099243,
      "epoch": 0.39674910425587695,
      "grad_norm": 1.75,
      "learning_rate": 0.000498919844415466,
      "loss": 6.1749,
      "mean_token_accuracy": 0.13809840381145477,
      "num_tokens": 9214684.0,
      "step": 4540
    },
    {
      "entropy": 6.232513427734375,
      "epoch": 0.39718605260858164,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004989167925632937,
      "loss": 6.0956,
      "mean_token_accuracy": 0.13999973982572556,
      "num_tokens": 9225721.0,
      "step": 4545
    },
    {
      "entropy": 6.2594568729400635,
      "epoch": 0.3976230009612864,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004989137364162801,
      "loss": 6.204,
      "mean_token_accuracy": 0.13365002050995828,
      "num_tokens": 9236167.0,
      "step": 4550
    },
    {
      "entropy": 6.248045921325684,
      "epoch": 0.3980599493139911,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004989106759744837,
      "loss": 6.1494,
      "mean_token_accuracy": 0.13835562020540237,
      "num_tokens": 9245716.0,
      "step": 4555
    },
    {
      "entropy": 6.303405904769898,
      "epoch": 0.39849689766669577,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004989076112379633,
      "loss": 6.229,
      "mean_token_accuracy": 0.12724980786442758,
      "num_tokens": 9256897.0,
      "step": 4560
    },
    {
      "entropy": 6.376533365249633,
      "epoch": 0.3989338460194005,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004989045422067775,
      "loss": 6.2398,
      "mean_token_accuracy": 0.12131529077887535,
      "num_tokens": 9266080.0,
      "step": 4565
    },
    {
      "entropy": 6.336807346343994,
      "epoch": 0.3993707943721052,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004989014688809853,
      "loss": 6.1559,
      "mean_token_accuracy": 0.13339983820915222,
      "num_tokens": 9276815.0,
      "step": 4570
    },
    {
      "entropy": 6.242786026000976,
      "epoch": 0.3998077427248099,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004988983912606454,
      "loss": 6.2034,
      "mean_token_accuracy": 0.13579384982585907,
      "num_tokens": 9287339.0,
      "step": 4575
    },
    {
      "entropy": 6.177998495101929,
      "epoch": 0.40024469107751465,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004988953093458173,
      "loss": 6.0911,
      "mean_token_accuracy": 0.13939642161130905,
      "num_tokens": 9298002.0,
      "step": 4580
    },
    {
      "entropy": 6.274811935424805,
      "epoch": 0.40068163943021934,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004988922231365597,
      "loss": 6.2056,
      "mean_token_accuracy": 0.13488566651940345,
      "num_tokens": 9309008.0,
      "step": 4585
    },
    {
      "entropy": 6.313817262649536,
      "epoch": 0.4011185877829241,
      "grad_norm": 1.9296875,
      "learning_rate": 0.000498889132632932,
      "loss": 6.2136,
      "mean_token_accuracy": 0.12972528114914894,
      "num_tokens": 9319391.0,
      "step": 4590
    },
    {
      "entropy": 6.283246612548828,
      "epoch": 0.4015555361356288,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004988860378349934,
      "loss": 6.0693,
      "mean_token_accuracy": 0.14501725882291794,
      "num_tokens": 9329808.0,
      "step": 4595
    },
    {
      "entropy": 6.2188276767730715,
      "epoch": 0.40199248448833347,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004988829387428033,
      "loss": 6.1365,
      "mean_token_accuracy": 0.1380341224372387,
      "num_tokens": 9339580.0,
      "step": 4600
    },
    {
      "entropy": 6.261927127838135,
      "epoch": 0.4024294328410382,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004988798353564211,
      "loss": 6.1482,
      "mean_token_accuracy": 0.14019528031349182,
      "num_tokens": 9350008.0,
      "step": 4605
    },
    {
      "entropy": 6.351348400115967,
      "epoch": 0.4028663811937429,
      "grad_norm": 2.0,
      "learning_rate": 0.0004988767276759064,
      "loss": 6.1556,
      "mean_token_accuracy": 0.1406958892941475,
      "num_tokens": 9359026.0,
      "step": 4610
    },
    {
      "entropy": 6.299345588684082,
      "epoch": 0.4033033295464476,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004988736157013186,
      "loss": 6.1164,
      "mean_token_accuracy": 0.1405692771077156,
      "num_tokens": 9369084.0,
      "step": 4615
    },
    {
      "entropy": 6.181832313537598,
      "epoch": 0.40374027789915234,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004988704994327178,
      "loss": 6.0993,
      "mean_token_accuracy": 0.14575967118144034,
      "num_tokens": 9378273.0,
      "step": 4620
    },
    {
      "entropy": 6.229097747802735,
      "epoch": 0.40417722625185704,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004988673788701633,
      "loss": 6.1288,
      "mean_token_accuracy": 0.13718436360359193,
      "num_tokens": 9388434.0,
      "step": 4625
    },
    {
      "entropy": 6.275507068634033,
      "epoch": 0.4046141746045617,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004988642540137152,
      "loss": 6.229,
      "mean_token_accuracy": 0.13442543298006057,
      "num_tokens": 9398484.0,
      "step": 4630
    },
    {
      "entropy": 6.276987504959107,
      "epoch": 0.4050511229572665,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004988611248634335,
      "loss": 6.098,
      "mean_token_accuracy": 0.13898643031716346,
      "num_tokens": 9409544.0,
      "step": 4635
    },
    {
      "entropy": 6.310326623916626,
      "epoch": 0.40548807130997117,
      "grad_norm": 1.5625,
      "learning_rate": 0.000498857991419378,
      "loss": 6.1672,
      "mean_token_accuracy": 0.13725512996315956,
      "num_tokens": 9419391.0,
      "step": 4640
    },
    {
      "entropy": 6.253142261505127,
      "epoch": 0.40592501966267586,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004988548536816091,
      "loss": 6.1333,
      "mean_token_accuracy": 0.14332313239574432,
      "num_tokens": 9429342.0,
      "step": 4645
    },
    {
      "entropy": 6.221396780014038,
      "epoch": 0.4063619680153806,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004988517116501867,
      "loss": 6.2374,
      "mean_token_accuracy": 0.12361930385231971,
      "num_tokens": 9440643.0,
      "step": 4650
    },
    {
      "entropy": 6.3487019538879395,
      "epoch": 0.4067989163680853,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004988485653251711,
      "loss": 6.1879,
      "mean_token_accuracy": 0.1392652027308941,
      "num_tokens": 9450674.0,
      "step": 4655
    },
    {
      "entropy": 6.317318248748779,
      "epoch": 0.40723586472079,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004988454147066228,
      "loss": 6.1493,
      "mean_token_accuracy": 0.13783033117651938,
      "num_tokens": 9460075.0,
      "step": 4660
    },
    {
      "entropy": 6.2390021800994875,
      "epoch": 0.40767281307349473,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004988422597946022,
      "loss": 6.1456,
      "mean_token_accuracy": 0.14160386994481086,
      "num_tokens": 9469327.0,
      "step": 4665
    },
    {
      "entropy": 6.297602844238281,
      "epoch": 0.4081097614261994,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004988391005891698,
      "loss": 6.1608,
      "mean_token_accuracy": 0.13125161826610565,
      "num_tokens": 9478304.0,
      "step": 4670
    },
    {
      "entropy": 6.253264236450195,
      "epoch": 0.4085467097789041,
      "grad_norm": 2.125,
      "learning_rate": 0.000498835937090386,
      "loss": 6.1152,
      "mean_token_accuracy": 0.13788354843854905,
      "num_tokens": 9487453.0,
      "step": 4675
    },
    {
      "entropy": 6.231164598464966,
      "epoch": 0.40898365813160886,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004988327692983116,
      "loss": 6.0639,
      "mean_token_accuracy": 0.14462961927056311,
      "num_tokens": 9496623.0,
      "step": 4680
    },
    {
      "entropy": 6.245018386840821,
      "epoch": 0.40942060648431355,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004988295972130076,
      "loss": 6.1898,
      "mean_token_accuracy": 0.13496496230363847,
      "num_tokens": 9506293.0,
      "step": 4685
    },
    {
      "entropy": 6.317003297805786,
      "epoch": 0.40985755483701825,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004988264208345346,
      "loss": 6.0992,
      "mean_token_accuracy": 0.13748762384057045,
      "num_tokens": 9516127.0,
      "step": 4690
    },
    {
      "entropy": 6.200523328781128,
      "epoch": 0.410294503189723,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004988232401629535,
      "loss": 6.0927,
      "mean_token_accuracy": 0.1418273501098156,
      "num_tokens": 9525597.0,
      "step": 4695
    },
    {
      "entropy": 6.208586359024048,
      "epoch": 0.4107314515424277,
      "grad_norm": 1.75,
      "learning_rate": 0.0004988200551983254,
      "loss": 6.0739,
      "mean_token_accuracy": 0.13941014930605888,
      "num_tokens": 9535208.0,
      "step": 4700
    },
    {
      "entropy": 6.236052703857422,
      "epoch": 0.4111683998951324,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004988168659407113,
      "loss": 6.1417,
      "mean_token_accuracy": 0.1489814765751362,
      "num_tokens": 9545078.0,
      "step": 4705
    },
    {
      "entropy": 6.321736574172974,
      "epoch": 0.4116053482478371,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004988136723901725,
      "loss": 6.1652,
      "mean_token_accuracy": 0.1318232461810112,
      "num_tokens": 9555074.0,
      "step": 4710
    },
    {
      "entropy": 6.245498085021973,
      "epoch": 0.4120422966005418,
      "grad_norm": 2.125,
      "learning_rate": 0.0004988104745467702,
      "loss": 6.1654,
      "mean_token_accuracy": 0.1366843894124031,
      "num_tokens": 9564777.0,
      "step": 4715
    },
    {
      "entropy": 6.245613765716553,
      "epoch": 0.4124792449532465,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004988072724105657,
      "loss": 6.0925,
      "mean_token_accuracy": 0.13342124447226525,
      "num_tokens": 9574551.0,
      "step": 4720
    },
    {
      "entropy": 6.277015590667725,
      "epoch": 0.41291619330595125,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004988040659816203,
      "loss": 6.2486,
      "mean_token_accuracy": 0.12661304399371148,
      "num_tokens": 9585162.0,
      "step": 4725
    },
    {
      "entropy": 6.323986434936524,
      "epoch": 0.41335314165865594,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004988008552599958,
      "loss": 6.1269,
      "mean_token_accuracy": 0.1459247313439846,
      "num_tokens": 9595422.0,
      "step": 4730
    },
    {
      "entropy": 6.303846168518066,
      "epoch": 0.41379009001136063,
      "grad_norm": 2.0,
      "learning_rate": 0.0004987976402457536,
      "loss": 6.1264,
      "mean_token_accuracy": 0.14018649980425835,
      "num_tokens": 9604315.0,
      "step": 4735
    },
    {
      "entropy": 6.219915056228638,
      "epoch": 0.4142270383640654,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004987944209389555,
      "loss": 6.1425,
      "mean_token_accuracy": 0.1348230868577957,
      "num_tokens": 9614369.0,
      "step": 4740
    },
    {
      "entropy": 6.2782059669494625,
      "epoch": 0.41466398671677007,
      "grad_norm": 1.9140625,
      "learning_rate": 0.000498791197339663,
      "loss": 6.177,
      "mean_token_accuracy": 0.13660474270582199,
      "num_tokens": 9624154.0,
      "step": 4745
    },
    {
      "entropy": 6.327009057998657,
      "epoch": 0.41510093506947476,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000498787969447938,
      "loss": 6.1719,
      "mean_token_accuracy": 0.1309115432202816,
      "num_tokens": 9634903.0,
      "step": 4750
    },
    {
      "entropy": 6.216856050491333,
      "epoch": 0.4155378834221795,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004987847372638427,
      "loss": 6.2105,
      "mean_token_accuracy": 0.1377333827316761,
      "num_tokens": 9644873.0,
      "step": 4755
    },
    {
      "entropy": 6.254890203475952,
      "epoch": 0.4159748317748842,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004987815007874388,
      "loss": 6.1216,
      "mean_token_accuracy": 0.14137759655714036,
      "num_tokens": 9654661.0,
      "step": 4760
    },
    {
      "entropy": 6.321761226654052,
      "epoch": 0.4164117801275889,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004987782600187884,
      "loss": 6.1429,
      "mean_token_accuracy": 0.14375444427132605,
      "num_tokens": 9664980.0,
      "step": 4765
    },
    {
      "entropy": 6.302230596542358,
      "epoch": 0.41684872848029364,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004987750149579537,
      "loss": 6.1302,
      "mean_token_accuracy": 0.13495260551571847,
      "num_tokens": 9675117.0,
      "step": 4770
    },
    {
      "entropy": 6.239761018753052,
      "epoch": 0.41728567683299833,
      "grad_norm": 2.265625,
      "learning_rate": 0.000498771765604997,
      "loss": 6.16,
      "mean_token_accuracy": 0.1374809578061104,
      "num_tokens": 9684900.0,
      "step": 4775
    },
    {
      "entropy": 6.26757173538208,
      "epoch": 0.417722625185703,
      "grad_norm": 3.0,
      "learning_rate": 0.0004987685119599806,
      "loss": 6.1657,
      "mean_token_accuracy": 0.13259884789586068,
      "num_tokens": 9694586.0,
      "step": 4780
    },
    {
      "entropy": 6.345765399932861,
      "epoch": 0.41815957353840777,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004987652540229668,
      "loss": 6.1779,
      "mean_token_accuracy": 0.14059032276272773,
      "num_tokens": 9705676.0,
      "step": 4785
    },
    {
      "entropy": 6.305094909667969,
      "epoch": 0.41859652189111246,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004987619917940183,
      "loss": 6.2076,
      "mean_token_accuracy": 0.14054128676652908,
      "num_tokens": 9715442.0,
      "step": 4790
    },
    {
      "entropy": 6.331025552749634,
      "epoch": 0.4190334702438172,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004987587252731975,
      "loss": 6.2761,
      "mean_token_accuracy": 0.1291821740567684,
      "num_tokens": 9726621.0,
      "step": 4795
    },
    {
      "entropy": 6.396036100387573,
      "epoch": 0.4194704185965219,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004987554544605671,
      "loss": 6.264,
      "mean_token_accuracy": 0.1289959356188774,
      "num_tokens": 9736688.0,
      "step": 4800
    },
    {
      "entropy": 6.341685104370117,
      "epoch": 0.4199073669492266,
      "grad_norm": 1.625,
      "learning_rate": 0.0004987521793561898,
      "loss": 6.1406,
      "mean_token_accuracy": 0.13719259053468705,
      "num_tokens": 9746478.0,
      "step": 4805
    },
    {
      "entropy": 6.214406061172485,
      "epoch": 0.42034431530193134,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004987488999601285,
      "loss": 6.179,
      "mean_token_accuracy": 0.13636296689510347,
      "num_tokens": 9756370.0,
      "step": 4810
    },
    {
      "entropy": 6.324111032485962,
      "epoch": 0.42078126365463603,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000498745616272446,
      "loss": 6.1979,
      "mean_token_accuracy": 0.13605131432414055,
      "num_tokens": 9766293.0,
      "step": 4815
    },
    {
      "entropy": 6.287442398071289,
      "epoch": 0.4212182120073407,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004987423282932054,
      "loss": 6.1521,
      "mean_token_accuracy": 0.1371973216533661,
      "num_tokens": 9777172.0,
      "step": 4820
    },
    {
      "entropy": 6.228835916519165,
      "epoch": 0.42165516036004547,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004987390360224697,
      "loss": 6.1417,
      "mean_token_accuracy": 0.13556611761450768,
      "num_tokens": 9787194.0,
      "step": 4825
    },
    {
      "entropy": 6.231557416915893,
      "epoch": 0.42209210871275016,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000498735739460302,
      "loss": 6.1746,
      "mean_token_accuracy": 0.13963005244731902,
      "num_tokens": 9797528.0,
      "step": 4830
    },
    {
      "entropy": 6.281035709381103,
      "epoch": 0.42252905706545485,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004987324386067656,
      "loss": 6.1915,
      "mean_token_accuracy": 0.136855348944664,
      "num_tokens": 9808412.0,
      "step": 4835
    },
    {
      "entropy": 6.375536823272705,
      "epoch": 0.4229660054181596,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004987291334619239,
      "loss": 6.2603,
      "mean_token_accuracy": 0.13053438663482667,
      "num_tokens": 9820047.0,
      "step": 4840
    },
    {
      "entropy": 6.282112407684326,
      "epoch": 0.4234029537708643,
      "grad_norm": 1.71875,
      "learning_rate": 0.00049872582402584,
      "loss": 6.1031,
      "mean_token_accuracy": 0.13801912888884543,
      "num_tokens": 9828812.0,
      "step": 4845
    },
    {
      "entropy": 6.20565824508667,
      "epoch": 0.423839902123569,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004987225102985778,
      "loss": 6.1333,
      "mean_token_accuracy": 0.1396217368543148,
      "num_tokens": 9838288.0,
      "step": 4850
    },
    {
      "entropy": 6.323318767547607,
      "epoch": 0.4242768504762737,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004987191922802003,
      "loss": 6.1694,
      "mean_token_accuracy": 0.1373145028948784,
      "num_tokens": 9848536.0,
      "step": 4855
    },
    {
      "entropy": 6.334697198867798,
      "epoch": 0.4247137988289784,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004987158699707717,
      "loss": 6.2124,
      "mean_token_accuracy": 0.13281583562493324,
      "num_tokens": 9859741.0,
      "step": 4860
    },
    {
      "entropy": 6.3007416248321535,
      "epoch": 0.4251507471816831,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004987125433703554,
      "loss": 6.1828,
      "mean_token_accuracy": 0.13512975797057153,
      "num_tokens": 9869173.0,
      "step": 4865
    },
    {
      "entropy": 6.326250267028809,
      "epoch": 0.42558769553438786,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004987092124790155,
      "loss": 6.1798,
      "mean_token_accuracy": 0.13396698832511902,
      "num_tokens": 9879619.0,
      "step": 4870
    },
    {
      "entropy": 6.256536626815796,
      "epoch": 0.42602464388709255,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004987058772968154,
      "loss": 6.0923,
      "mean_token_accuracy": 0.15063711032271385,
      "num_tokens": 9890214.0,
      "step": 4875
    },
    {
      "entropy": 6.207323265075684,
      "epoch": 0.42646159223979724,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004987025378238195,
      "loss": 6.1358,
      "mean_token_accuracy": 0.1356499522924423,
      "num_tokens": 9900266.0,
      "step": 4880
    },
    {
      "entropy": 6.373274326324463,
      "epoch": 0.426898540592502,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004986991940600917,
      "loss": 6.1614,
      "mean_token_accuracy": 0.14057112634181976,
      "num_tokens": 9910790.0,
      "step": 4885
    },
    {
      "entropy": 6.35953106880188,
      "epoch": 0.4273354889452067,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004986958460056961,
      "loss": 6.1699,
      "mean_token_accuracy": 0.12968755811452864,
      "num_tokens": 9921167.0,
      "step": 4890
    },
    {
      "entropy": 6.220204162597656,
      "epoch": 0.42777243729791137,
      "grad_norm": 1.90625,
      "learning_rate": 0.000498692493660697,
      "loss": 6.1137,
      "mean_token_accuracy": 0.14389638230204582,
      "num_tokens": 9931674.0,
      "step": 4895
    },
    {
      "entropy": 6.240059232711792,
      "epoch": 0.4282093856506161,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004986891370251587,
      "loss": 6.0591,
      "mean_token_accuracy": 0.13982971683144568,
      "num_tokens": 9941484.0,
      "step": 4900
    },
    {
      "entropy": 6.3408478736877445,
      "epoch": 0.4286463340033208,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004986857760991454,
      "loss": 6.2198,
      "mean_token_accuracy": 0.13376458287239074,
      "num_tokens": 9951885.0,
      "step": 4905
    },
    {
      "entropy": 6.308237314224243,
      "epoch": 0.4290832823560255,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004986824108827218,
      "loss": 6.0576,
      "mean_token_accuracy": 0.14093493595719336,
      "num_tokens": 9961658.0,
      "step": 4910
    },
    {
      "entropy": 6.158659744262695,
      "epoch": 0.42952023070873024,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004986790413759523,
      "loss": 6.1409,
      "mean_token_accuracy": 0.13590563461184502,
      "num_tokens": 9971971.0,
      "step": 4915
    },
    {
      "entropy": 6.2474815368652346,
      "epoch": 0.42995717906143494,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004986756675789016,
      "loss": 6.057,
      "mean_token_accuracy": 0.1382055960595608,
      "num_tokens": 9982211.0,
      "step": 4920
    },
    {
      "entropy": 6.227798938751221,
      "epoch": 0.4303941274141396,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004986722894916344,
      "loss": 6.0999,
      "mean_token_accuracy": 0.14289458617568015,
      "num_tokens": 9991935.0,
      "step": 4925
    },
    {
      "entropy": 6.233926773071289,
      "epoch": 0.4308310757668444,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004986689071142155,
      "loss": 6.1476,
      "mean_token_accuracy": 0.13406068161129953,
      "num_tokens": 10002014.0,
      "step": 4930
    },
    {
      "entropy": 6.267605781555176,
      "epoch": 0.43126802411954906,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004986655204467097,
      "loss": 6.1681,
      "mean_token_accuracy": 0.13497984185814857,
      "num_tokens": 10012030.0,
      "step": 4935
    },
    {
      "entropy": 6.2325609683990475,
      "epoch": 0.43170497247225376,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004986621294891822,
      "loss": 6.1395,
      "mean_token_accuracy": 0.13742314130067826,
      "num_tokens": 10022188.0,
      "step": 4940
    },
    {
      "entropy": 6.2819342613220215,
      "epoch": 0.4321419208249585,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004986587342416976,
      "loss": 6.0424,
      "mean_token_accuracy": 0.14051671847701072,
      "num_tokens": 10032131.0,
      "step": 4945
    },
    {
      "entropy": 6.3012300491333,
      "epoch": 0.4325788691776632,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004986553347043214,
      "loss": 6.1026,
      "mean_token_accuracy": 0.1372190795838833,
      "num_tokens": 10043204.0,
      "step": 4950
    },
    {
      "entropy": 6.27725133895874,
      "epoch": 0.4330158175303679,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004986519308771188,
      "loss": 6.1602,
      "mean_token_accuracy": 0.1372821055352688,
      "num_tokens": 10052863.0,
      "step": 4955
    },
    {
      "entropy": 6.238871479034424,
      "epoch": 0.43345276588307263,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004986485227601548,
      "loss": 6.1281,
      "mean_token_accuracy": 0.14254157543182372,
      "num_tokens": 10063212.0,
      "step": 4960
    },
    {
      "entropy": 6.304278421401977,
      "epoch": 0.4338897142357773,
      "grad_norm": 3.359375,
      "learning_rate": 0.000498645110353495,
      "loss": 6.1642,
      "mean_token_accuracy": 0.1352584458887577,
      "num_tokens": 10074180.0,
      "step": 4965
    },
    {
      "entropy": 6.254854822158814,
      "epoch": 0.434326662588482,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004986416936572049,
      "loss": 6.115,
      "mean_token_accuracy": 0.14317337051033974,
      "num_tokens": 10084513.0,
      "step": 4970
    },
    {
      "entropy": 6.285315895080567,
      "epoch": 0.43476361094118676,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004986382726713497,
      "loss": 6.1416,
      "mean_token_accuracy": 0.13702046871185303,
      "num_tokens": 10094352.0,
      "step": 4975
    },
    {
      "entropy": 6.238975763320923,
      "epoch": 0.43520055929389145,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004986348473959955,
      "loss": 6.0051,
      "mean_token_accuracy": 0.15185373425483703,
      "num_tokens": 10103677.0,
      "step": 4980
    },
    {
      "entropy": 6.216447448730468,
      "epoch": 0.43563750764659614,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004986314178312076,
      "loss": 6.2146,
      "mean_token_accuracy": 0.13050349950790405,
      "num_tokens": 10113479.0,
      "step": 4985
    },
    {
      "entropy": 6.266159963607788,
      "epoch": 0.4360744559993009,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004986279839770519,
      "loss": 6.0954,
      "mean_token_accuracy": 0.14220530316233634,
      "num_tokens": 10123526.0,
      "step": 4990
    },
    {
      "entropy": 6.2446376323699955,
      "epoch": 0.4365114043520056,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004986245458335944,
      "loss": 6.1378,
      "mean_token_accuracy": 0.13541444465518,
      "num_tokens": 10134176.0,
      "step": 4995
    },
    {
      "entropy": 6.265491676330567,
      "epoch": 0.43694835270471033,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004986211034009008,
      "loss": 6.0889,
      "mean_token_accuracy": 0.1447139300405979,
      "num_tokens": 10144763.0,
      "step": 5000
    },
    {
      "entropy": 6.22551908493042,
      "epoch": 0.437385301057415,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004986176566790375,
      "loss": 6.0839,
      "mean_token_accuracy": 0.14613140970468522,
      "num_tokens": 10155011.0,
      "step": 5005
    },
    {
      "entropy": 6.261688280105591,
      "epoch": 0.4378222494101197,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004986142056680701,
      "loss": 6.1545,
      "mean_token_accuracy": 0.13823105841875077,
      "num_tokens": 10165474.0,
      "step": 5010
    },
    {
      "entropy": 6.294175291061402,
      "epoch": 0.43825919776282446,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004986107503680652,
      "loss": 6.184,
      "mean_token_accuracy": 0.14294707179069518,
      "num_tokens": 10175137.0,
      "step": 5015
    },
    {
      "entropy": 6.259722518920898,
      "epoch": 0.43869614611552915,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004986072907790889,
      "loss": 6.1862,
      "mean_token_accuracy": 0.1319163627922535,
      "num_tokens": 10185623.0,
      "step": 5020
    },
    {
      "entropy": 6.3086286067962645,
      "epoch": 0.43913309446823384,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004986038269012076,
      "loss": 6.15,
      "mean_token_accuracy": 0.14134794026613234,
      "num_tokens": 10195201.0,
      "step": 5025
    },
    {
      "entropy": 6.29203200340271,
      "epoch": 0.4395700428209386,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004986003587344877,
      "loss": 6.1807,
      "mean_token_accuracy": 0.13764383196830748,
      "num_tokens": 10205951.0,
      "step": 5030
    },
    {
      "entropy": 6.234759187698364,
      "epoch": 0.4400069911736433,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004985968862789957,
      "loss": 6.0516,
      "mean_token_accuracy": 0.14456723183393477,
      "num_tokens": 10215665.0,
      "step": 5035
    },
    {
      "entropy": 6.230546188354492,
      "epoch": 0.44044393952634797,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004985934095347985,
      "loss": 6.1461,
      "mean_token_accuracy": 0.13692135065793992,
      "num_tokens": 10225037.0,
      "step": 5040
    },
    {
      "entropy": 6.164027881622315,
      "epoch": 0.4408808878790527,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004985899285019624,
      "loss": 6.1115,
      "mean_token_accuracy": 0.13943026587367058,
      "num_tokens": 10234581.0,
      "step": 5045
    },
    {
      "entropy": 6.272631359100342,
      "epoch": 0.4413178362317574,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004985864431805543,
      "loss": 6.0153,
      "mean_token_accuracy": 0.14523213654756545,
      "num_tokens": 10243716.0,
      "step": 5050
    },
    {
      "entropy": 6.294983530044556,
      "epoch": 0.4417547845844621,
      "grad_norm": 2.0,
      "learning_rate": 0.0004985829535706411,
      "loss": 6.1241,
      "mean_token_accuracy": 0.1396915651857853,
      "num_tokens": 10253198.0,
      "step": 5055
    },
    {
      "entropy": 6.240324974060059,
      "epoch": 0.44219173293716685,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004985794596722897,
      "loss": 6.1935,
      "mean_token_accuracy": 0.1337872676551342,
      "num_tokens": 10262912.0,
      "step": 5060
    },
    {
      "entropy": 6.27632966041565,
      "epoch": 0.44262868128987154,
      "grad_norm": 1.65625,
      "learning_rate": 0.000498575961485567,
      "loss": 6.1393,
      "mean_token_accuracy": 0.13864117488265038,
      "num_tokens": 10272960.0,
      "step": 5065
    },
    {
      "entropy": 6.33082594871521,
      "epoch": 0.44306562964257623,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004985724590105403,
      "loss": 6.1777,
      "mean_token_accuracy": 0.13294639587402343,
      "num_tokens": 10283034.0,
      "step": 5070
    },
    {
      "entropy": 6.242563962936401,
      "epoch": 0.443502577995281,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004985689522472767,
      "loss": 6.1406,
      "mean_token_accuracy": 0.13814122676849366,
      "num_tokens": 10292553.0,
      "step": 5075
    },
    {
      "entropy": 6.212743759155273,
      "epoch": 0.44393952634798567,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004985654411958434,
      "loss": 6.1061,
      "mean_token_accuracy": 0.13814549073576926,
      "num_tokens": 10302475.0,
      "step": 5080
    },
    {
      "entropy": 6.28053388595581,
      "epoch": 0.44437647470069036,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004985619258563078,
      "loss": 6.1142,
      "mean_token_accuracy": 0.14118221551179885,
      "num_tokens": 10311693.0,
      "step": 5085
    },
    {
      "entropy": 6.312252950668335,
      "epoch": 0.4448134230533951,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004985584062287372,
      "loss": 6.1182,
      "mean_token_accuracy": 0.14231281131505966,
      "num_tokens": 10323052.0,
      "step": 5090
    },
    {
      "entropy": 6.227076244354248,
      "epoch": 0.4452503714060998,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004985548823131994,
      "loss": 6.1118,
      "mean_token_accuracy": 0.13764646723866464,
      "num_tokens": 10332997.0,
      "step": 5095
    },
    {
      "entropy": 6.322070789337158,
      "epoch": 0.4456873197588045,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004985513541097618,
      "loss": 6.2201,
      "mean_token_accuracy": 0.13421619758009912,
      "num_tokens": 10343298.0,
      "step": 5100
    },
    {
      "entropy": 6.336534023284912,
      "epoch": 0.44612426811150924,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000498547821618492,
      "loss": 6.1144,
      "mean_token_accuracy": 0.14139558374881744,
      "num_tokens": 10352485.0,
      "step": 5105
    },
    {
      "entropy": 6.177110052108764,
      "epoch": 0.44656121646421393,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004985442848394579,
      "loss": 6.1703,
      "mean_token_accuracy": 0.13453608453273774,
      "num_tokens": 10362810.0,
      "step": 5110
    },
    {
      "entropy": 6.2077878475189205,
      "epoch": 0.4469981648169186,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004985407437727273,
      "loss": 6.1211,
      "mean_token_accuracy": 0.1392608255147934,
      "num_tokens": 10372817.0,
      "step": 5115
    },
    {
      "entropy": 6.309681034088134,
      "epoch": 0.44743511316962337,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004985371984183681,
      "loss": 6.157,
      "mean_token_accuracy": 0.1435435265302658,
      "num_tokens": 10383046.0,
      "step": 5120
    },
    {
      "entropy": 6.246765279769898,
      "epoch": 0.44787206152232806,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004985336487764483,
      "loss": 6.1826,
      "mean_token_accuracy": 0.1371629036962986,
      "num_tokens": 10393498.0,
      "step": 5125
    },
    {
      "entropy": 6.284682369232177,
      "epoch": 0.44830900987503275,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004985300948470359,
      "loss": 6.1789,
      "mean_token_accuracy": 0.13662788644433022,
      "num_tokens": 10403595.0,
      "step": 5130
    },
    {
      "entropy": 6.321662712097168,
      "epoch": 0.4487459582277375,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004985265366301992,
      "loss": 6.0978,
      "mean_token_accuracy": 0.13992622643709182,
      "num_tokens": 10413475.0,
      "step": 5135
    },
    {
      "entropy": 6.256285905838013,
      "epoch": 0.4491829065804422,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004985229741260065,
      "loss": 6.1482,
      "mean_token_accuracy": 0.14397763162851335,
      "num_tokens": 10423512.0,
      "step": 5140
    },
    {
      "entropy": 6.2504840850830075,
      "epoch": 0.4496198549331469,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004985194073345259,
      "loss": 6.164,
      "mean_token_accuracy": 0.1343572497367859,
      "num_tokens": 10434198.0,
      "step": 5145
    },
    {
      "entropy": 6.243315315246582,
      "epoch": 0.4500568032858516,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004985158362558261,
      "loss": 6.1113,
      "mean_token_accuracy": 0.1347861848771572,
      "num_tokens": 10444077.0,
      "step": 5150
    },
    {
      "entropy": 6.259370422363281,
      "epoch": 0.4504937516385563,
      "grad_norm": 1.875,
      "learning_rate": 0.0004985122608899752,
      "loss": 6.1479,
      "mean_token_accuracy": 0.1376808151602745,
      "num_tokens": 10454481.0,
      "step": 5155
    },
    {
      "entropy": 6.303308010101318,
      "epoch": 0.450930699991261,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004985086812370422,
      "loss": 6.1446,
      "mean_token_accuracy": 0.13569988757371904,
      "num_tokens": 10464453.0,
      "step": 5160
    },
    {
      "entropy": 6.338668537139893,
      "epoch": 0.45136764834396575,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004985050972970956,
      "loss": 6.2315,
      "mean_token_accuracy": 0.1327189899981022,
      "num_tokens": 10474932.0,
      "step": 5165
    },
    {
      "entropy": 6.316284084320069,
      "epoch": 0.45180459669667045,
      "grad_norm": 1.7890625,
      "learning_rate": 0.000498501509070204,
      "loss": 6.2224,
      "mean_token_accuracy": 0.1338010162115097,
      "num_tokens": 10484719.0,
      "step": 5170
    },
    {
      "entropy": 6.248991584777832,
      "epoch": 0.45224154504937514,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004984979165564364,
      "loss": 6.2699,
      "mean_token_accuracy": 0.12927524521946906,
      "num_tokens": 10495808.0,
      "step": 5175
    },
    {
      "entropy": 6.309340620040894,
      "epoch": 0.4526784934020799,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004984943197558616,
      "loss": 6.1642,
      "mean_token_accuracy": 0.13639893680810927,
      "num_tokens": 10506528.0,
      "step": 5180
    },
    {
      "entropy": 6.324915504455566,
      "epoch": 0.4531154417547846,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004984907186685487,
      "loss": 6.1391,
      "mean_token_accuracy": 0.13927296698093414,
      "num_tokens": 10517117.0,
      "step": 5185
    },
    {
      "entropy": 6.337198209762573,
      "epoch": 0.45355239010748927,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004984871132945667,
      "loss": 6.1291,
      "mean_token_accuracy": 0.1367869570851326,
      "num_tokens": 10527394.0,
      "step": 5190
    },
    {
      "entropy": 6.266608858108521,
      "epoch": 0.453989338460194,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004984835036339848,
      "loss": 6.1256,
      "mean_token_accuracy": 0.13677924424409865,
      "num_tokens": 10536779.0,
      "step": 5195
    },
    {
      "entropy": 6.243874835968017,
      "epoch": 0.4544262868128987,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004984798896868722,
      "loss": 6.2,
      "mean_token_accuracy": 0.1301602840423584,
      "num_tokens": 10547116.0,
      "step": 5200
    },
    {
      "entropy": 6.328261041641236,
      "epoch": 0.45486323516560345,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004984762714532981,
      "loss": 6.0815,
      "mean_token_accuracy": 0.14150492921471597,
      "num_tokens": 10556396.0,
      "step": 5205
    },
    {
      "entropy": 6.2097395896911625,
      "epoch": 0.45530018351830814,
      "grad_norm": 2.109375,
      "learning_rate": 0.000498472648933332,
      "loss": 6.2847,
      "mean_token_accuracy": 0.13041267469525336,
      "num_tokens": 10568127.0,
      "step": 5210
    },
    {
      "entropy": 6.29595627784729,
      "epoch": 0.45573713187101283,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004984690221270436,
      "loss": 6.1136,
      "mean_token_accuracy": 0.14152851030230523,
      "num_tokens": 10578565.0,
      "step": 5215
    },
    {
      "entropy": 6.28857102394104,
      "epoch": 0.4561740802237176,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004984653910345021,
      "loss": 6.1523,
      "mean_token_accuracy": 0.1274610109627247,
      "num_tokens": 10588633.0,
      "step": 5220
    },
    {
      "entropy": 6.25370512008667,
      "epoch": 0.4566110285764223,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004984617556557774,
      "loss": 6.1017,
      "mean_token_accuracy": 0.1470041058957577,
      "num_tokens": 10597917.0,
      "step": 5225
    },
    {
      "entropy": 6.3347279071807865,
      "epoch": 0.45704797692912696,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000498458115990939,
      "loss": 6.1662,
      "mean_token_accuracy": 0.13205529153347015,
      "num_tokens": 10608396.0,
      "step": 5230
    },
    {
      "entropy": 6.224194192886353,
      "epoch": 0.4574849252818317,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004984544720400568,
      "loss": 6.1553,
      "mean_token_accuracy": 0.14002828896045685,
      "num_tokens": 10618191.0,
      "step": 5235
    },
    {
      "entropy": 6.242114734649658,
      "epoch": 0.4579218736345364,
      "grad_norm": 1.859375,
      "learning_rate": 0.000498450823803201,
      "loss": 6.1463,
      "mean_token_accuracy": 0.13393353223800658,
      "num_tokens": 10627594.0,
      "step": 5240
    },
    {
      "entropy": 6.292330074310303,
      "epoch": 0.4583588219872411,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004984471712804412,
      "loss": 6.135,
      "mean_token_accuracy": 0.14734850078821182,
      "num_tokens": 10637260.0,
      "step": 5245
    },
    {
      "entropy": 6.3297521591186525,
      "epoch": 0.45879577033994584,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004984435144718475,
      "loss": 6.1519,
      "mean_token_accuracy": 0.14019931629300117,
      "num_tokens": 10647968.0,
      "step": 5250
    },
    {
      "entropy": 6.210675239562988,
      "epoch": 0.45923271869265053,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004984398533774901,
      "loss": 6.0935,
      "mean_token_accuracy": 0.14126433730125426,
      "num_tokens": 10657760.0,
      "step": 5255
    },
    {
      "entropy": 6.232064247131348,
      "epoch": 0.4596696670453552,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004984361879974393,
      "loss": 6.1274,
      "mean_token_accuracy": 0.1337346188724041,
      "num_tokens": 10667577.0,
      "step": 5260
    },
    {
      "entropy": 6.291175508499146,
      "epoch": 0.46010661539805997,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004984325183317654,
      "loss": 6.1139,
      "mean_token_accuracy": 0.135456882417202,
      "num_tokens": 10677984.0,
      "step": 5265
    },
    {
      "entropy": 6.245900678634643,
      "epoch": 0.46054356375076466,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004984288443805386,
      "loss": 6.1471,
      "mean_token_accuracy": 0.14233410954475403,
      "num_tokens": 10687700.0,
      "step": 5270
    },
    {
      "entropy": 6.252521085739136,
      "epoch": 0.46098051210346935,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004984251661438295,
      "loss": 6.1306,
      "mean_token_accuracy": 0.1394471637904644,
      "num_tokens": 10697351.0,
      "step": 5275
    },
    {
      "entropy": 6.27062954902649,
      "epoch": 0.4614174604561741,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004984214836217087,
      "loss": 6.1384,
      "mean_token_accuracy": 0.1371995449066162,
      "num_tokens": 10707498.0,
      "step": 5280
    },
    {
      "entropy": 6.301582670211792,
      "epoch": 0.4618544088088788,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004984177968142467,
      "loss": 6.1647,
      "mean_token_accuracy": 0.1353110708296299,
      "num_tokens": 10717724.0,
      "step": 5285
    },
    {
      "entropy": 6.294982767105102,
      "epoch": 0.4622913571615835,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004984141057215143,
      "loss": 6.1396,
      "mean_token_accuracy": 0.13943758979439735,
      "num_tokens": 10727030.0,
      "step": 5290
    },
    {
      "entropy": 6.221749544143677,
      "epoch": 0.46272830551428823,
      "grad_norm": 2.25,
      "learning_rate": 0.0004984104103435824,
      "loss": 6.0332,
      "mean_token_accuracy": 0.1422347217798233,
      "num_tokens": 10736861.0,
      "step": 5295
    },
    {
      "entropy": 6.303720426559448,
      "epoch": 0.4631652538669929,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004984067106805217,
      "loss": 6.2312,
      "mean_token_accuracy": 0.13308657109737396,
      "num_tokens": 10748310.0,
      "step": 5300
    },
    {
      "entropy": 6.329783916473389,
      "epoch": 0.4636022022196976,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004984030067324032,
      "loss": 6.0846,
      "mean_token_accuracy": 0.1387430287897587,
      "num_tokens": 10758110.0,
      "step": 5305
    },
    {
      "entropy": 6.263466262817383,
      "epoch": 0.46403915057240236,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004983992984992979,
      "loss": 6.1352,
      "mean_token_accuracy": 0.14275783076882362,
      "num_tokens": 10767919.0,
      "step": 5310
    },
    {
      "entropy": 6.255305910110474,
      "epoch": 0.46447609892510705,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004983955859812771,
      "loss": 6.0637,
      "mean_token_accuracy": 0.1412653312087059,
      "num_tokens": 10776888.0,
      "step": 5315
    },
    {
      "entropy": 6.233955955505371,
      "epoch": 0.46491304727781174,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004983918691784119,
      "loss": 6.1344,
      "mean_token_accuracy": 0.13571209982037544,
      "num_tokens": 10786806.0,
      "step": 5320
    },
    {
      "entropy": 6.285499954223633,
      "epoch": 0.4653499956305165,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004983881480907736,
      "loss": 6.1211,
      "mean_token_accuracy": 0.13879325687885286,
      "num_tokens": 10796452.0,
      "step": 5325
    },
    {
      "entropy": 6.286596488952637,
      "epoch": 0.4657869439832212,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004983844227184334,
      "loss": 6.0932,
      "mean_token_accuracy": 0.1357939973473549,
      "num_tokens": 10806718.0,
      "step": 5330
    },
    {
      "entropy": 6.296868705749512,
      "epoch": 0.46622389233592587,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004983806930614631,
      "loss": 6.2042,
      "mean_token_accuracy": 0.13334835544228554,
      "num_tokens": 10817978.0,
      "step": 5335
    },
    {
      "entropy": 6.271781158447266,
      "epoch": 0.4666608406886306,
      "grad_norm": 2.390625,
      "learning_rate": 0.000498376959119934,
      "loss": 6.0604,
      "mean_token_accuracy": 0.14710889458656312,
      "num_tokens": 10827818.0,
      "step": 5340
    },
    {
      "entropy": 6.251382207870483,
      "epoch": 0.4670977890413353,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004983732208939178,
      "loss": 6.2205,
      "mean_token_accuracy": 0.1345639370381832,
      "num_tokens": 10838492.0,
      "step": 5345
    },
    {
      "entropy": 6.263595485687256,
      "epoch": 0.46753473739404,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004983694783834863,
      "loss": 6.0753,
      "mean_token_accuracy": 0.13865476250648498,
      "num_tokens": 10847807.0,
      "step": 5350
    },
    {
      "entropy": 6.2714066982269285,
      "epoch": 0.46797168574674475,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004983657315887109,
      "loss": 6.1925,
      "mean_token_accuracy": 0.1346592463552952,
      "num_tokens": 10858637.0,
      "step": 5355
    },
    {
      "entropy": 6.325130271911621,
      "epoch": 0.46840863409944944,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004983619805096639,
      "loss": 6.1113,
      "mean_token_accuracy": 0.14088876843452453,
      "num_tokens": 10868042.0,
      "step": 5360
    },
    {
      "entropy": 6.261090326309204,
      "epoch": 0.46884558245215413,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004983582251464171,
      "loss": 6.1164,
      "mean_token_accuracy": 0.13843616396188735,
      "num_tokens": 10877171.0,
      "step": 5365
    },
    {
      "entropy": 6.176885032653809,
      "epoch": 0.4692825308048589,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004983544654990423,
      "loss": 6.1067,
      "mean_token_accuracy": 0.1395420491695404,
      "num_tokens": 10888434.0,
      "step": 5370
    },
    {
      "entropy": 6.2764442443847654,
      "epoch": 0.46971947915756357,
      "grad_norm": 1.625,
      "learning_rate": 0.000498350701567612,
      "loss": 6.1631,
      "mean_token_accuracy": 0.14059600606560707,
      "num_tokens": 10898467.0,
      "step": 5375
    },
    {
      "entropy": 6.308957958221436,
      "epoch": 0.47015642751026826,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004983469333521983,
      "loss": 6.1202,
      "mean_token_accuracy": 0.14148796796798707,
      "num_tokens": 10908178.0,
      "step": 5380
    },
    {
      "entropy": 6.2301304817199705,
      "epoch": 0.470593375862973,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004983431608528732,
      "loss": 6.2115,
      "mean_token_accuracy": 0.12733783274888993,
      "num_tokens": 10919957.0,
      "step": 5385
    },
    {
      "entropy": 6.302609539031982,
      "epoch": 0.4710303242156777,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004983393840697092,
      "loss": 6.1579,
      "mean_token_accuracy": 0.13929501324892044,
      "num_tokens": 10929728.0,
      "step": 5390
    },
    {
      "entropy": 6.291434621810913,
      "epoch": 0.4714672725683824,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004983356030027789,
      "loss": 6.1957,
      "mean_token_accuracy": 0.13646872639656066,
      "num_tokens": 10940427.0,
      "step": 5395
    },
    {
      "entropy": 6.217552328109742,
      "epoch": 0.47190422092108714,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004983318176521546,
      "loss": 6.1575,
      "mean_token_accuracy": 0.13504248932003976,
      "num_tokens": 10950641.0,
      "step": 5400
    },
    {
      "entropy": 6.335336399078369,
      "epoch": 0.4723411692737918,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004983280280179092,
      "loss": 6.0819,
      "mean_token_accuracy": 0.1392090730369091,
      "num_tokens": 10959341.0,
      "step": 5405
    },
    {
      "entropy": 6.327978801727295,
      "epoch": 0.4727781176264966,
      "grad_norm": 2.1875,
      "learning_rate": 0.000498324234100115,
      "loss": 6.1075,
      "mean_token_accuracy": 0.13960520774126053,
      "num_tokens": 10967745.0,
      "step": 5410
    },
    {
      "entropy": 6.209249353408813,
      "epoch": 0.47321506597920127,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004983204358988451,
      "loss": 6.1168,
      "mean_token_accuracy": 0.13568175733089446,
      "num_tokens": 10978345.0,
      "step": 5415
    },
    {
      "entropy": 6.214342164993286,
      "epoch": 0.47365201433190596,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004983166334141722,
      "loss": 6.1723,
      "mean_token_accuracy": 0.13103256821632386,
      "num_tokens": 10988030.0,
      "step": 5420
    },
    {
      "entropy": 6.362129402160645,
      "epoch": 0.4740889626846107,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004983128266461693,
      "loss": 6.1685,
      "mean_token_accuracy": 0.12891222313046455,
      "num_tokens": 10999169.0,
      "step": 5425
    },
    {
      "entropy": 6.280517530441284,
      "epoch": 0.4745259110373154,
      "grad_norm": 5.125,
      "learning_rate": 0.0004983090155949093,
      "loss": 6.2207,
      "mean_token_accuracy": 0.13183309584856034,
      "num_tokens": 11009608.0,
      "step": 5430
    },
    {
      "entropy": 6.205153751373291,
      "epoch": 0.4749628593900201,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004983052002604653,
      "loss": 6.1465,
      "mean_token_accuracy": 0.13711439147591592,
      "num_tokens": 11020187.0,
      "step": 5435
    },
    {
      "entropy": 6.298795366287232,
      "epoch": 0.47539980774272483,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004983013806429107,
      "loss": 6.1303,
      "mean_token_accuracy": 0.1389899007976055,
      "num_tokens": 11030459.0,
      "step": 5440
    },
    {
      "entropy": 6.219760274887085,
      "epoch": 0.4758367560954295,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004982975567423185,
      "loss": 6.0758,
      "mean_token_accuracy": 0.1416230507194996,
      "num_tokens": 11040880.0,
      "step": 5445
    },
    {
      "entropy": 6.261538934707642,
      "epoch": 0.4762737044481342,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004982937285587623,
      "loss": 6.1041,
      "mean_token_accuracy": 0.14153316840529442,
      "num_tokens": 11050901.0,
      "step": 5450
    },
    {
      "entropy": 6.320458269119262,
      "epoch": 0.47671065280083896,
      "grad_norm": 2.25,
      "learning_rate": 0.0004982898960923152,
      "loss": 6.2696,
      "mean_token_accuracy": 0.1284100241959095,
      "num_tokens": 11061274.0,
      "step": 5455
    },
    {
      "entropy": 6.34821310043335,
      "epoch": 0.47714760115354365,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004982860593430509,
      "loss": 6.076,
      "mean_token_accuracy": 0.1355578728020191,
      "num_tokens": 11070644.0,
      "step": 5460
    },
    {
      "entropy": 6.249191522598267,
      "epoch": 0.47758454950624835,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004982822183110431,
      "loss": 6.136,
      "mean_token_accuracy": 0.1370291531085968,
      "num_tokens": 11081154.0,
      "step": 5465
    },
    {
      "entropy": 6.270349931716919,
      "epoch": 0.4780214978589531,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004982783729963654,
      "loss": 6.2228,
      "mean_token_accuracy": 0.1320676937699318,
      "num_tokens": 11091929.0,
      "step": 5470
    },
    {
      "entropy": 6.239210939407348,
      "epoch": 0.4784584462116578,
      "grad_norm": 2.125,
      "learning_rate": 0.0004982745233990914,
      "loss": 6.0728,
      "mean_token_accuracy": 0.14582818001508713,
      "num_tokens": 11102075.0,
      "step": 5475
    },
    {
      "entropy": 6.25434103012085,
      "epoch": 0.4788953945643625,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004982706695192949,
      "loss": 6.1428,
      "mean_token_accuracy": 0.1331348791718483,
      "num_tokens": 11111886.0,
      "step": 5480
    },
    {
      "entropy": 6.288838911056518,
      "epoch": 0.4793323429170672,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004982668113570503,
      "loss": 6.1392,
      "mean_token_accuracy": 0.13364839926362038,
      "num_tokens": 11122329.0,
      "step": 5485
    },
    {
      "entropy": 6.253954172134399,
      "epoch": 0.4797692912697719,
      "grad_norm": 2.78125,
      "learning_rate": 0.000498262948912431,
      "loss": 6.0818,
      "mean_token_accuracy": 0.13713959529995917,
      "num_tokens": 11132348.0,
      "step": 5490
    },
    {
      "entropy": 6.224376440048218,
      "epoch": 0.4802062396224766,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004982590821855115,
      "loss": 6.1627,
      "mean_token_accuracy": 0.13753138929605485,
      "num_tokens": 11141975.0,
      "step": 5495
    },
    {
      "entropy": 6.271784734725952,
      "epoch": 0.48064318797518135,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004982552111763657,
      "loss": 6.1531,
      "mean_token_accuracy": 0.13205370903015137,
      "num_tokens": 11152174.0,
      "step": 5500
    },
    {
      "entropy": 6.328236293792725,
      "epoch": 0.48108013632788604,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004982513358850681,
      "loss": 6.1503,
      "mean_token_accuracy": 0.13584645465016365,
      "num_tokens": 11162563.0,
      "step": 5505
    },
    {
      "entropy": 6.294511032104492,
      "epoch": 0.48151708468059073,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004982474563116928,
      "loss": 6.0588,
      "mean_token_accuracy": 0.1419669657945633,
      "num_tokens": 11172572.0,
      "step": 5510
    },
    {
      "entropy": 6.245644664764404,
      "epoch": 0.4819540330332955,
      "grad_norm": 2.25,
      "learning_rate": 0.0004982435724563143,
      "loss": 6.1391,
      "mean_token_accuracy": 0.14169370457530023,
      "num_tokens": 11183404.0,
      "step": 5515
    },
    {
      "entropy": 6.188567304611206,
      "epoch": 0.4823909813860002,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004982396843190072,
      "loss": 6.1603,
      "mean_token_accuracy": 0.13115517497062684,
      "num_tokens": 11193905.0,
      "step": 5520
    },
    {
      "entropy": 6.244078588485718,
      "epoch": 0.48282792973870486,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004982357918998459,
      "loss": 6.0617,
      "mean_token_accuracy": 0.15290317088365554,
      "num_tokens": 11203127.0,
      "step": 5525
    },
    {
      "entropy": 6.258895540237427,
      "epoch": 0.4832648780914096,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004982318951989053,
      "loss": 6.0768,
      "mean_token_accuracy": 0.14411210864782334,
      "num_tokens": 11213468.0,
      "step": 5530
    },
    {
      "entropy": 6.245413780212402,
      "epoch": 0.4837018264441143,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004982279942162597,
      "loss": 6.171,
      "mean_token_accuracy": 0.14047423750162125,
      "num_tokens": 11224722.0,
      "step": 5535
    },
    {
      "entropy": 6.271645212173462,
      "epoch": 0.484138774796819,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004982240889519844,
      "loss": 6.1466,
      "mean_token_accuracy": 0.1361476957798004,
      "num_tokens": 11234565.0,
      "step": 5540
    },
    {
      "entropy": 6.279338216781616,
      "epoch": 0.48457572314952374,
      "grad_norm": 2.453125,
      "learning_rate": 0.000498220179406154,
      "loss": 6.1516,
      "mean_token_accuracy": 0.13795690312981607,
      "num_tokens": 11244496.0,
      "step": 5545
    },
    {
      "entropy": 6.319229602813721,
      "epoch": 0.48501267150222843,
      "grad_norm": 1.875,
      "learning_rate": 0.0004982162655788436,
      "loss": 6.1097,
      "mean_token_accuracy": 0.1403142899274826,
      "num_tokens": 11254187.0,
      "step": 5550
    },
    {
      "entropy": 6.252913284301758,
      "epoch": 0.4854496198549331,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004982123474701283,
      "loss": 6.0967,
      "mean_token_accuracy": 0.14224717393517494,
      "num_tokens": 11264580.0,
      "step": 5555
    },
    {
      "entropy": 6.238586473464966,
      "epoch": 0.48588656820763787,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004982084250800832,
      "loss": 6.2267,
      "mean_token_accuracy": 0.13188387230038642,
      "num_tokens": 11275104.0,
      "step": 5560
    },
    {
      "entropy": 6.348457193374633,
      "epoch": 0.48632351656034256,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004982044984087836,
      "loss": 6.2423,
      "mean_token_accuracy": 0.13024362698197364,
      "num_tokens": 11285983.0,
      "step": 5565
    },
    {
      "entropy": 6.368983268737793,
      "epoch": 0.48676046491304725,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004982005674563048,
      "loss": 6.1208,
      "mean_token_accuracy": 0.1400579683482647,
      "num_tokens": 11295378.0,
      "step": 5570
    },
    {
      "entropy": 6.3164036750793455,
      "epoch": 0.487197413265752,
      "grad_norm": 2.515625,
      "learning_rate": 0.000498196632222722,
      "loss": 6.2217,
      "mean_token_accuracy": 0.1353451281785965,
      "num_tokens": 11305500.0,
      "step": 5575
    },
    {
      "entropy": 6.254426193237305,
      "epoch": 0.4876343616184567,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004981926927081109,
      "loss": 6.1479,
      "mean_token_accuracy": 0.13708797097206116,
      "num_tokens": 11316455.0,
      "step": 5580
    },
    {
      "entropy": 6.273268747329712,
      "epoch": 0.4880713099711614,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004981887489125471,
      "loss": 6.1121,
      "mean_token_accuracy": 0.1336928963661194,
      "num_tokens": 11325865.0,
      "step": 5585
    },
    {
      "entropy": 6.25109896659851,
      "epoch": 0.48850825832386613,
      "grad_norm": 2.234375,
      "learning_rate": 0.000498184800836106,
      "loss": 6.0871,
      "mean_token_accuracy": 0.14778236225247382,
      "num_tokens": 11337368.0,
      "step": 5590
    },
    {
      "entropy": 6.20948600769043,
      "epoch": 0.4889452066765708,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004981808484788635,
      "loss": 6.0673,
      "mean_token_accuracy": 0.1448922872543335,
      "num_tokens": 11347129.0,
      "step": 5595
    },
    {
      "entropy": 6.27021336555481,
      "epoch": 0.48938215502927557,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004981768918408955,
      "loss": 6.1304,
      "mean_token_accuracy": 0.13713421151041985,
      "num_tokens": 11356641.0,
      "step": 5600
    },
    {
      "entropy": 6.252780914306641,
      "epoch": 0.48981910338198026,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004981729309222777,
      "loss": 6.0043,
      "mean_token_accuracy": 0.1457279920578003,
      "num_tokens": 11366485.0,
      "step": 5605
    },
    {
      "entropy": 6.238640213012696,
      "epoch": 0.49025605173468495,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004981689657230861,
      "loss": 6.1711,
      "mean_token_accuracy": 0.1374865762889385,
      "num_tokens": 11376305.0,
      "step": 5610
    },
    {
      "entropy": 6.294531774520874,
      "epoch": 0.4906930000873897,
      "grad_norm": 2.9375,
      "learning_rate": 0.000498164996243397,
      "loss": 6.1401,
      "mean_token_accuracy": 0.13293426483869553,
      "num_tokens": 11385837.0,
      "step": 5615
    },
    {
      "entropy": 6.267136859893799,
      "epoch": 0.4911299484400944,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004981610224832861,
      "loss": 6.1751,
      "mean_token_accuracy": 0.14129467606544494,
      "num_tokens": 11396719.0,
      "step": 5620
    },
    {
      "entropy": 6.326460218429565,
      "epoch": 0.4915668967927991,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004981570444428299,
      "loss": 6.191,
      "mean_token_accuracy": 0.13250873759388923,
      "num_tokens": 11408107.0,
      "step": 5625
    },
    {
      "entropy": 6.281957674026489,
      "epoch": 0.4920038451455038,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004981530621221048,
      "loss": 6.1149,
      "mean_token_accuracy": 0.14190632849931717,
      "num_tokens": 11417547.0,
      "step": 5630
    },
    {
      "entropy": 6.336714124679565,
      "epoch": 0.4924407934982085,
      "grad_norm": 2.625,
      "learning_rate": 0.0004981490755211869,
      "loss": 6.1658,
      "mean_token_accuracy": 0.14127280861139296,
      "num_tokens": 11427660.0,
      "step": 5635
    },
    {
      "entropy": 6.287770366668701,
      "epoch": 0.4928777418509132,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004981450846401528,
      "loss": 6.0589,
      "mean_token_accuracy": 0.14569119960069657,
      "num_tokens": 11438032.0,
      "step": 5640
    },
    {
      "entropy": 6.2903404712677,
      "epoch": 0.49331469020361796,
      "grad_norm": 2.15625,
      "learning_rate": 0.000498141089479079,
      "loss": 6.1505,
      "mean_token_accuracy": 0.13444297164678573,
      "num_tokens": 11448987.0,
      "step": 5645
    },
    {
      "entropy": 6.238702011108399,
      "epoch": 0.49375163855632265,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004981370900380423,
      "loss": 6.1149,
      "mean_token_accuracy": 0.14375114664435387,
      "num_tokens": 11458659.0,
      "step": 5650
    },
    {
      "entropy": 6.276248025894165,
      "epoch": 0.49418858690902734,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004981330863171194,
      "loss": 6.0849,
      "mean_token_accuracy": 0.1386402390897274,
      "num_tokens": 11468104.0,
      "step": 5655
    },
    {
      "entropy": 6.289006233215332,
      "epoch": 0.4946255352617321,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004981290783163867,
      "loss": 6.0704,
      "mean_token_accuracy": 0.1435806483030319,
      "num_tokens": 11479579.0,
      "step": 5660
    },
    {
      "entropy": 6.295569610595703,
      "epoch": 0.4950624836144368,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004981250660359216,
      "loss": 6.1365,
      "mean_token_accuracy": 0.14263879805803298,
      "num_tokens": 11489202.0,
      "step": 5665
    },
    {
      "entropy": 6.295229196548462,
      "epoch": 0.49549943196714147,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004981210494758007,
      "loss": 6.1683,
      "mean_token_accuracy": 0.13030786961317062,
      "num_tokens": 11499425.0,
      "step": 5670
    },
    {
      "entropy": 6.279474401473999,
      "epoch": 0.4959363803198462,
      "grad_norm": 2.921875,
      "learning_rate": 0.000498117028636101,
      "loss": 6.1286,
      "mean_token_accuracy": 0.13767644241452218,
      "num_tokens": 11508983.0,
      "step": 5675
    },
    {
      "entropy": 6.254884052276611,
      "epoch": 0.4963733286725509,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004981130035169,
      "loss": 6.1943,
      "mean_token_accuracy": 0.14120121970772742,
      "num_tokens": 11519791.0,
      "step": 5680
    },
    {
      "entropy": 6.310758495330811,
      "epoch": 0.4968102770252556,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004981089741182747,
      "loss": 6.1183,
      "mean_token_accuracy": 0.14048743546009063,
      "num_tokens": 11530537.0,
      "step": 5685
    },
    {
      "entropy": 6.284348869323731,
      "epoch": 0.49724722537796034,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004981049404403023,
      "loss": 6.1598,
      "mean_token_accuracy": 0.13990036472678186,
      "num_tokens": 11540591.0,
      "step": 5690
    },
    {
      "entropy": 6.2591174125671385,
      "epoch": 0.49768417373066504,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004981009024830602,
      "loss": 6.0942,
      "mean_token_accuracy": 0.13814121037721633,
      "num_tokens": 11550719.0,
      "step": 5695
    },
    {
      "entropy": 6.222075080871582,
      "epoch": 0.4981211220833697,
      "grad_norm": 2.390625,
      "learning_rate": 0.000498096860246626,
      "loss": 6.1473,
      "mean_token_accuracy": 0.1384321503341198,
      "num_tokens": 11561805.0,
      "step": 5700
    },
    {
      "entropy": 6.352153635025024,
      "epoch": 0.4985580704360745,
      "grad_norm": 2.328125,
      "learning_rate": 0.000498092813731077,
      "loss": 6.1634,
      "mean_token_accuracy": 0.13272807076573373,
      "num_tokens": 11571380.0,
      "step": 5705
    },
    {
      "entropy": 6.310147285461426,
      "epoch": 0.49899501878877917,
      "grad_norm": 2.796875,
      "learning_rate": 0.000498088762936491,
      "loss": 6.1564,
      "mean_token_accuracy": 0.1386110305786133,
      "num_tokens": 11580853.0,
      "step": 5710
    },
    {
      "entropy": 6.148832035064697,
      "epoch": 0.49943196714148386,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004980847078629457,
      "loss": 6.0639,
      "mean_token_accuracy": 0.1558942325413227,
      "num_tokens": 11591966.0,
      "step": 5715
    },
    {
      "entropy": 6.198760461807251,
      "epoch": 0.4998689154941886,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004980806485105186,
      "loss": 6.1139,
      "mean_token_accuracy": 0.14199392199516297,
      "num_tokens": 11601239.0,
      "step": 5720
    },
    {
      "entropy": 6.2181836605072025,
      "epoch": 0.5003058638468934,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004980765848792879,
      "loss": 6.0396,
      "mean_token_accuracy": 0.1473376788198948,
      "num_tokens": 11609857.0,
      "step": 5725
    },
    {
      "entropy": 6.266326093673706,
      "epoch": 0.500742812199598,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004980725169693314,
      "loss": 6.0735,
      "mean_token_accuracy": 0.14394057914614677,
      "num_tokens": 11618952.0,
      "step": 5730
    },
    {
      "entropy": 6.269810056686401,
      "epoch": 0.5011797605523027,
      "grad_norm": 2.5,
      "learning_rate": 0.0004980684447807272,
      "loss": 6.0361,
      "mean_token_accuracy": 0.14204780012369156,
      "num_tokens": 11627638.0,
      "step": 5735
    },
    {
      "entropy": 6.2179922580719,
      "epoch": 0.5016167089050074,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004980643683135533,
      "loss": 6.1444,
      "mean_token_accuracy": 0.1408068686723709,
      "num_tokens": 11637842.0,
      "step": 5740
    },
    {
      "entropy": 6.215860080718994,
      "epoch": 0.5020536572577121,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004980602875678878,
      "loss": 6.0222,
      "mean_token_accuracy": 0.14877952188253402,
      "num_tokens": 11647217.0,
      "step": 5745
    },
    {
      "entropy": 6.198777055740356,
      "epoch": 0.5024906056104168,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004980562025438093,
      "loss": 6.0429,
      "mean_token_accuracy": 0.1428973264992237,
      "num_tokens": 11657558.0,
      "step": 5750
    },
    {
      "entropy": 6.150891494750977,
      "epoch": 0.5029275539631216,
      "grad_norm": 2.5,
      "learning_rate": 0.0004980521132413958,
      "loss": 6.0492,
      "mean_token_accuracy": 0.14264654815196992,
      "num_tokens": 11668054.0,
      "step": 5755
    },
    {
      "entropy": 6.298615503311157,
      "epoch": 0.5033645023158263,
      "grad_norm": 2.390625,
      "learning_rate": 0.000498048019660726,
      "loss": 6.0764,
      "mean_token_accuracy": 0.1434045687317848,
      "num_tokens": 11677779.0,
      "step": 5760
    },
    {
      "entropy": 6.304033184051514,
      "epoch": 0.503801450668531,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004980439218018783,
      "loss": 6.1201,
      "mean_token_accuracy": 0.13853600099682808,
      "num_tokens": 11687577.0,
      "step": 5765
    },
    {
      "entropy": 6.269702291488647,
      "epoch": 0.5042383990212357,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004980398196649312,
      "loss": 6.2192,
      "mean_token_accuracy": 0.13097881749272347,
      "num_tokens": 11699630.0,
      "step": 5770
    },
    {
      "entropy": 6.266751432418824,
      "epoch": 0.5046753473739404,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004980357132499636,
      "loss": 6.1513,
      "mean_token_accuracy": 0.13628338277339935,
      "num_tokens": 11710139.0,
      "step": 5775
    },
    {
      "entropy": 6.301572465896607,
      "epoch": 0.5051122957266451,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004980316025570542,
      "loss": 6.1026,
      "mean_token_accuracy": 0.13941313028335572,
      "num_tokens": 11719070.0,
      "step": 5780
    },
    {
      "entropy": 6.239738321304321,
      "epoch": 0.5055492440793499,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004980274875862816,
      "loss": 6.1406,
      "mean_token_accuracy": 0.1385067269206047,
      "num_tokens": 11728770.0,
      "step": 5785
    },
    {
      "entropy": 6.282432889938354,
      "epoch": 0.5059861924320546,
      "grad_norm": 2.921875,
      "learning_rate": 0.000498023368337725,
      "loss": 6.0922,
      "mean_token_accuracy": 0.14360614940524102,
      "num_tokens": 11737753.0,
      "step": 5790
    },
    {
      "entropy": 6.236385011672974,
      "epoch": 0.5064231407847593,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004980192448114635,
      "loss": 6.0672,
      "mean_token_accuracy": 0.14035124331712723,
      "num_tokens": 11747837.0,
      "step": 5795
    },
    {
      "entropy": 6.165898418426513,
      "epoch": 0.5068600891374639,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004980151170075758,
      "loss": 6.1914,
      "mean_token_accuracy": 0.1387983664870262,
      "num_tokens": 11757911.0,
      "step": 5800
    },
    {
      "entropy": 6.33115816116333,
      "epoch": 0.5072970374901686,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004980109849261414,
      "loss": 6.051,
      "mean_token_accuracy": 0.14173925220966338,
      "num_tokens": 11766213.0,
      "step": 5805
    },
    {
      "entropy": 6.249571466445923,
      "epoch": 0.5077339858428733,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004980068485672394,
      "loss": 6.0191,
      "mean_token_accuracy": 0.1495368130505085,
      "num_tokens": 11775640.0,
      "step": 5810
    },
    {
      "entropy": 6.280322885513305,
      "epoch": 0.5081709341955781,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004980027079309493,
      "loss": 6.0965,
      "mean_token_accuracy": 0.14636014476418496,
      "num_tokens": 11785664.0,
      "step": 5815
    },
    {
      "entropy": 6.23998990058899,
      "epoch": 0.5086078825482828,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004979985630173503,
      "loss": 6.1902,
      "mean_token_accuracy": 0.12907436937093736,
      "num_tokens": 11795402.0,
      "step": 5820
    },
    {
      "entropy": 6.258623647689819,
      "epoch": 0.5090448309009875,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004979944138265221,
      "loss": 6.0445,
      "mean_token_accuracy": 0.14042759463191032,
      "num_tokens": 11805923.0,
      "step": 5825
    },
    {
      "entropy": 6.231123399734497,
      "epoch": 0.5094817792536922,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004979902603585442,
      "loss": 6.1287,
      "mean_token_accuracy": 0.14135455191135407,
      "num_tokens": 11816097.0,
      "step": 5830
    },
    {
      "entropy": 6.2674267292022705,
      "epoch": 0.5099187276063969,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004979861026134961,
      "loss": 6.0798,
      "mean_token_accuracy": 0.14172066152095794,
      "num_tokens": 11826889.0,
      "step": 5835
    },
    {
      "entropy": 6.2974059104919435,
      "epoch": 0.5103556759591016,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004979819405914579,
      "loss": 6.1233,
      "mean_token_accuracy": 0.1364697828888893,
      "num_tokens": 11837935.0,
      "step": 5840
    },
    {
      "entropy": 6.264686822891235,
      "epoch": 0.5107926243118064,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004979777742925092,
      "loss": 6.1301,
      "mean_token_accuracy": 0.1356620818376541,
      "num_tokens": 11847715.0,
      "step": 5845
    },
    {
      "entropy": 6.263647890090942,
      "epoch": 0.5112295726645111,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004979736037167298,
      "loss": 6.1405,
      "mean_token_accuracy": 0.1327194854617119,
      "num_tokens": 11857841.0,
      "step": 5850
    },
    {
      "entropy": 6.285010814666748,
      "epoch": 0.5116665210172158,
      "grad_norm": 2.375,
      "learning_rate": 0.0004979694288642,
      "loss": 6.1182,
      "mean_token_accuracy": 0.14218535423278808,
      "num_tokens": 11868135.0,
      "step": 5855
    },
    {
      "entropy": 6.3095512866973875,
      "epoch": 0.5121034693699205,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004979652497349997,
      "loss": 6.0558,
      "mean_token_accuracy": 0.14099670946598053,
      "num_tokens": 11877637.0,
      "step": 5860
    },
    {
      "entropy": 6.186911582946777,
      "epoch": 0.5125404177226252,
      "grad_norm": 1.8203125,
      "learning_rate": 0.000497961066329209,
      "loss": 6.1142,
      "mean_token_accuracy": 0.13389617949724197,
      "num_tokens": 11887128.0,
      "step": 5865
    },
    {
      "entropy": 6.2536896705627445,
      "epoch": 0.5129773660753298,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004979568786469083,
      "loss": 6.1368,
      "mean_token_accuracy": 0.12886641845107077,
      "num_tokens": 11897885.0,
      "step": 5870
    },
    {
      "entropy": 6.281183671951294,
      "epoch": 0.5134143144280346,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004979526866881777,
      "loss": 6.0867,
      "mean_token_accuracy": 0.13957348465919495,
      "num_tokens": 11907919.0,
      "step": 5875
    },
    {
      "entropy": 6.24418797492981,
      "epoch": 0.5138512627807393,
      "grad_norm": 3.59375,
      "learning_rate": 0.000497948490453098,
      "loss": 6.078,
      "mean_token_accuracy": 0.1437679760158062,
      "num_tokens": 11918612.0,
      "step": 5880
    },
    {
      "entropy": 6.180129671096802,
      "epoch": 0.514288211133444,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004979442899417493,
      "loss": 6.1513,
      "mean_token_accuracy": 0.13741420656442643,
      "num_tokens": 11927759.0,
      "step": 5885
    },
    {
      "entropy": 6.280853271484375,
      "epoch": 0.5147251594861487,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004979400851542122,
      "loss": 6.0726,
      "mean_token_accuracy": 0.1479244738817215,
      "num_tokens": 11937522.0,
      "step": 5890
    },
    {
      "entropy": 6.3057684898376465,
      "epoch": 0.5151621078388534,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004979358760905675,
      "loss": 6.0803,
      "mean_token_accuracy": 0.13870553597807883,
      "num_tokens": 11947787.0,
      "step": 5895
    },
    {
      "entropy": 6.171240282058716,
      "epoch": 0.5155990561915581,
      "grad_norm": 2.84375,
      "learning_rate": 0.000497931662750896,
      "loss": 6.1081,
      "mean_token_accuracy": 0.1385062739253044,
      "num_tokens": 11958929.0,
      "step": 5900
    },
    {
      "entropy": 6.199537706375122,
      "epoch": 0.5160360045442629,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004979274451352783,
      "loss": 6.0583,
      "mean_token_accuracy": 0.13892978727817534,
      "num_tokens": 11969242.0,
      "step": 5905
    },
    {
      "entropy": 6.292746734619141,
      "epoch": 0.5164729528969676,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004979232232437954,
      "loss": 6.1336,
      "mean_token_accuracy": 0.1345326527953148,
      "num_tokens": 11980149.0,
      "step": 5910
    },
    {
      "entropy": 6.410525369644165,
      "epoch": 0.5169099012496723,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004979189970765282,
      "loss": 6.1351,
      "mean_token_accuracy": 0.13244713917374612,
      "num_tokens": 11990636.0,
      "step": 5915
    },
    {
      "entropy": 6.33321442604065,
      "epoch": 0.517346849602377,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004979147666335579,
      "loss": 6.1639,
      "mean_token_accuracy": 0.14218005761504174,
      "num_tokens": 12000615.0,
      "step": 5920
    },
    {
      "entropy": 6.229039573669434,
      "epoch": 0.5177837979550817,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004979105319149656,
      "loss": 6.1872,
      "mean_token_accuracy": 0.1388854667544365,
      "num_tokens": 12011072.0,
      "step": 5925
    },
    {
      "entropy": 6.2475790023803714,
      "epoch": 0.5182207463077865,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004979062929208324,
      "loss": 6.0716,
      "mean_token_accuracy": 0.14881260618567466,
      "num_tokens": 12021726.0,
      "step": 5930
    },
    {
      "entropy": 6.287198257446289,
      "epoch": 0.5186576946604912,
      "grad_norm": 2.125,
      "learning_rate": 0.0004979020496512398,
      "loss": 6.2044,
      "mean_token_accuracy": 0.13563393875956536,
      "num_tokens": 12031353.0,
      "step": 5935
    },
    {
      "entropy": 6.357402753829956,
      "epoch": 0.5190946430131959,
      "grad_norm": 2.71875,
      "learning_rate": 0.000497897802106269,
      "loss": 6.1369,
      "mean_token_accuracy": 0.13450635075569153,
      "num_tokens": 12041234.0,
      "step": 5940
    },
    {
      "entropy": 6.225025939941406,
      "epoch": 0.5195315913659005,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004978935502860017,
      "loss": 6.0755,
      "mean_token_accuracy": 0.14126456677913665,
      "num_tokens": 12051100.0,
      "step": 5945
    },
    {
      "entropy": 6.214692163467407,
      "epoch": 0.5199685397186052,
      "grad_norm": 2.25,
      "learning_rate": 0.0004978892941905191,
      "loss": 6.1442,
      "mean_token_accuracy": 0.1400235928595066,
      "num_tokens": 12062012.0,
      "step": 5950
    },
    {
      "entropy": 6.325677824020386,
      "epoch": 0.5204054880713099,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004978850338199033,
      "loss": 6.1463,
      "mean_token_accuracy": 0.1371977373957634,
      "num_tokens": 12072763.0,
      "step": 5955
    },
    {
      "entropy": 6.259446907043457,
      "epoch": 0.5208424364240147,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004978807691742356,
      "loss": 5.9974,
      "mean_token_accuracy": 0.148126420378685,
      "num_tokens": 12081892.0,
      "step": 5960
    },
    {
      "entropy": 6.177407932281494,
      "epoch": 0.5212793847767194,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000497876500253598,
      "loss": 6.0896,
      "mean_token_accuracy": 0.14243873953819275,
      "num_tokens": 12091428.0,
      "step": 5965
    },
    {
      "entropy": 6.217768621444702,
      "epoch": 0.5217163331294241,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004978722270580722,
      "loss": 6.1084,
      "mean_token_accuracy": 0.1395185723900795,
      "num_tokens": 12101687.0,
      "step": 5970
    },
    {
      "entropy": 6.190403652191162,
      "epoch": 0.5221532814821288,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004978679495877405,
      "loss": 6.0414,
      "mean_token_accuracy": 0.1401352971792221,
      "num_tokens": 12111726.0,
      "step": 5975
    },
    {
      "entropy": 6.2400407791137695,
      "epoch": 0.5225902298348335,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004978636678426847,
      "loss": 6.0739,
      "mean_token_accuracy": 0.14529430642724037,
      "num_tokens": 12122171.0,
      "step": 5980
    },
    {
      "entropy": 6.371889734268189,
      "epoch": 0.5230271781875382,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004978593818229869,
      "loss": 6.1141,
      "mean_token_accuracy": 0.1365119956433773,
      "num_tokens": 12131414.0,
      "step": 5985
    },
    {
      "entropy": 6.23081226348877,
      "epoch": 0.523464126540243,
      "grad_norm": 2.375,
      "learning_rate": 0.0004978550915287294,
      "loss": 6.0954,
      "mean_token_accuracy": 0.1337629109621048,
      "num_tokens": 12141665.0,
      "step": 5990
    },
    {
      "entropy": 6.2171844959259035,
      "epoch": 0.5239010748929477,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004978507969599945,
      "loss": 6.1587,
      "mean_token_accuracy": 0.1341513715684414,
      "num_tokens": 12151263.0,
      "step": 5995
    },
    {
      "entropy": 6.334724473953247,
      "epoch": 0.5243380232456524,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004978464981168645,
      "loss": 6.1197,
      "mean_token_accuracy": 0.13718220144510268,
      "num_tokens": 12161070.0,
      "step": 6000
    },
    {
      "epoch": 0.5243380232456524,
      "eval_entropy": 6.185264414372392,
      "eval_loss": 6.144855499267578,
      "eval_mean_token_accuracy": 0.14467228393698192,
      "eval_num_tokens": 12161070.0,
      "eval_runtime": 28.2471,
      "eval_samples_per_second": 1310.366,
      "eval_steps_per_second": 163.805,
      "step": 6000
    },
    {
      "entropy": 6.30408787727356,
      "epoch": 0.5247749715983571,
      "grad_norm": 2.5,
      "learning_rate": 0.000497842194999422,
      "loss": 6.1043,
      "mean_token_accuracy": 0.13960754871368408,
      "num_tokens": 12171197.0,
      "step": 6005
    },
    {
      "entropy": 6.211764478683472,
      "epoch": 0.5252119199510618,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004978378876077493,
      "loss": 6.1377,
      "mean_token_accuracy": 0.13761500567197799,
      "num_tokens": 12181462.0,
      "step": 6010
    },
    {
      "entropy": 6.263395357131958,
      "epoch": 0.5256488683037664,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004978335759419293,
      "loss": 6.1135,
      "mean_token_accuracy": 0.14520457834005357,
      "num_tokens": 12191473.0,
      "step": 6015
    },
    {
      "entropy": 6.347568559646606,
      "epoch": 0.5260858166564713,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004978292600020444,
      "loss": 6.0729,
      "mean_token_accuracy": 0.14109345898032188,
      "num_tokens": 12200920.0,
      "step": 6020
    },
    {
      "entropy": 6.300686883926391,
      "epoch": 0.5265227650091759,
      "grad_norm": 2.125,
      "learning_rate": 0.0004978249397881776,
      "loss": 6.1262,
      "mean_token_accuracy": 0.1390794076025486,
      "num_tokens": 12210823.0,
      "step": 6025
    },
    {
      "entropy": 6.263337230682373,
      "epoch": 0.5269597133618806,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004978206153004117,
      "loss": 6.1859,
      "mean_token_accuracy": 0.13617446646094322,
      "num_tokens": 12221967.0,
      "step": 6030
    },
    {
      "entropy": 6.28148627281189,
      "epoch": 0.5273966617145853,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004978162865388297,
      "loss": 6.1163,
      "mean_token_accuracy": 0.13850070759654046,
      "num_tokens": 12232519.0,
      "step": 6035
    },
    {
      "entropy": 6.273348712921143,
      "epoch": 0.52783361006729,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004978119535035144,
      "loss": 6.0763,
      "mean_token_accuracy": 0.13540117889642717,
      "num_tokens": 12243193.0,
      "step": 6040
    },
    {
      "entropy": 6.2839103698730465,
      "epoch": 0.5282705584199947,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004978076161945492,
      "loss": 6.0884,
      "mean_token_accuracy": 0.14195277616381646,
      "num_tokens": 12253640.0,
      "step": 6045
    },
    {
      "entropy": 6.409676742553711,
      "epoch": 0.5287075067726995,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004978032746120173,
      "loss": 6.2042,
      "mean_token_accuracy": 0.12442022413015366,
      "num_tokens": 12265072.0,
      "step": 6050
    },
    {
      "entropy": 6.2757384300231935,
      "epoch": 0.5291444551254042,
      "grad_norm": 2.5,
      "learning_rate": 0.0004977989287560016,
      "loss": 6.136,
      "mean_token_accuracy": 0.13877920806407928,
      "num_tokens": 12274449.0,
      "step": 6055
    },
    {
      "entropy": 6.271896934509277,
      "epoch": 0.5295814034781089,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004977945786265857,
      "loss": 6.1416,
      "mean_token_accuracy": 0.1312025398015976,
      "num_tokens": 12284910.0,
      "step": 6060
    },
    {
      "entropy": 6.38758020401001,
      "epoch": 0.5300183518308136,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004977902242238533,
      "loss": 6.1486,
      "mean_token_accuracy": 0.1351582631468773,
      "num_tokens": 12294046.0,
      "step": 6065
    },
    {
      "entropy": 6.280045509338379,
      "epoch": 0.5304553001835183,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004977858655478874,
      "loss": 6.1488,
      "mean_token_accuracy": 0.13523225262761115,
      "num_tokens": 12303900.0,
      "step": 6070
    },
    {
      "entropy": 6.238226747512817,
      "epoch": 0.530892248536223,
      "grad_norm": 1.671875,
      "learning_rate": 0.000497781502598772,
      "loss": 6.0946,
      "mean_token_accuracy": 0.14537760987877846,
      "num_tokens": 12313326.0,
      "step": 6075
    },
    {
      "entropy": 6.217903280258179,
      "epoch": 0.5313291968889278,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004977771353765906,
      "loss": 6.0816,
      "mean_token_accuracy": 0.13752787113189696,
      "num_tokens": 12322792.0,
      "step": 6080
    },
    {
      "entropy": 6.297841787338257,
      "epoch": 0.5317661452416325,
      "grad_norm": 2.109375,
      "learning_rate": 0.000497772763881427,
      "loss": 6.2045,
      "mean_token_accuracy": 0.13359650894999503,
      "num_tokens": 12333417.0,
      "step": 6085
    },
    {
      "entropy": 6.292433023452759,
      "epoch": 0.5322030935943372,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004977683881133651,
      "loss": 6.1106,
      "mean_token_accuracy": 0.13563139513134956,
      "num_tokens": 12343944.0,
      "step": 6090
    },
    {
      "entropy": 6.2595867156982425,
      "epoch": 0.5326400419470418,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004977640080724888,
      "loss": 6.0801,
      "mean_token_accuracy": 0.1347056582570076,
      "num_tokens": 12353838.0,
      "step": 6095
    },
    {
      "entropy": 6.292349815368652,
      "epoch": 0.5330769902997465,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004977596237588821,
      "loss": 6.1777,
      "mean_token_accuracy": 0.1316857986152172,
      "num_tokens": 12364996.0,
      "step": 6100
    },
    {
      "entropy": 6.327098417282104,
      "epoch": 0.5335139386524512,
      "grad_norm": 1.859375,
      "learning_rate": 0.000497755235172629,
      "loss": 6.1415,
      "mean_token_accuracy": 0.1362926870584488,
      "num_tokens": 12375111.0,
      "step": 6105
    },
    {
      "entropy": 6.296980190277099,
      "epoch": 0.533950887005156,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004977508423138139,
      "loss": 6.1166,
      "mean_token_accuracy": 0.13595320358872415,
      "num_tokens": 12384376.0,
      "step": 6110
    },
    {
      "entropy": 6.287497091293335,
      "epoch": 0.5343878353578607,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004977464451825208,
      "loss": 6.1722,
      "mean_token_accuracy": 0.13291269168257713,
      "num_tokens": 12395248.0,
      "step": 6115
    },
    {
      "entropy": 6.29342999458313,
      "epoch": 0.5348247837105654,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004977420437788341,
      "loss": 6.1345,
      "mean_token_accuracy": 0.13793925195932388,
      "num_tokens": 12405083.0,
      "step": 6120
    },
    {
      "entropy": 6.214169359207153,
      "epoch": 0.5352617320632701,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004977376381028385,
      "loss": 6.0607,
      "mean_token_accuracy": 0.14968093633651733,
      "num_tokens": 12415787.0,
      "step": 6125
    },
    {
      "entropy": 6.330326509475708,
      "epoch": 0.5356986804159748,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004977332281546181,
      "loss": 6.1069,
      "mean_token_accuracy": 0.13906310126185417,
      "num_tokens": 12425911.0,
      "step": 6130
    },
    {
      "entropy": 6.279489517211914,
      "epoch": 0.5361356287686796,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004977288139342578,
      "loss": 6.1376,
      "mean_token_accuracy": 0.1360112026333809,
      "num_tokens": 12437270.0,
      "step": 6135
    },
    {
      "entropy": 6.254675626754761,
      "epoch": 0.5365725771213843,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004977243954418422,
      "loss": 6.0827,
      "mean_token_accuracy": 0.13789032846689225,
      "num_tokens": 12446778.0,
      "step": 6140
    },
    {
      "entropy": 6.228334951400757,
      "epoch": 0.537009525474089,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004977199726774558,
      "loss": 6.0745,
      "mean_token_accuracy": 0.14418192505836486,
      "num_tokens": 12456550.0,
      "step": 6145
    },
    {
      "entropy": 6.249480247497559,
      "epoch": 0.5374464738267937,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004977155456411837,
      "loss": 6.0808,
      "mean_token_accuracy": 0.13996390253305435,
      "num_tokens": 12465562.0,
      "step": 6150
    },
    {
      "entropy": 6.334767532348633,
      "epoch": 0.5378834221794984,
      "grad_norm": 2.5,
      "learning_rate": 0.0004977111143331108,
      "loss": 6.1114,
      "mean_token_accuracy": 0.13722820729017257,
      "num_tokens": 12475277.0,
      "step": 6155
    },
    {
      "entropy": 6.240406560897827,
      "epoch": 0.538320370532203,
      "grad_norm": 1.75,
      "learning_rate": 0.0004977066787533219,
      "loss": 6.0562,
      "mean_token_accuracy": 0.1435964971780777,
      "num_tokens": 12485478.0,
      "step": 6160
    },
    {
      "entropy": 6.293786430358887,
      "epoch": 0.5387573188849079,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004977022389019023,
      "loss": 6.1654,
      "mean_token_accuracy": 0.13637920394539832,
      "num_tokens": 12495714.0,
      "step": 6165
    },
    {
      "entropy": 6.393072557449341,
      "epoch": 0.5391942672376125,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004976977947789371,
      "loss": 6.1033,
      "mean_token_accuracy": 0.14155342802405357,
      "num_tokens": 12505125.0,
      "step": 6170
    },
    {
      "entropy": 6.261453914642334,
      "epoch": 0.5396312155903172,
      "grad_norm": 4.375,
      "learning_rate": 0.0004976933463845115,
      "loss": 6.0887,
      "mean_token_accuracy": 0.14094004184007644,
      "num_tokens": 12515051.0,
      "step": 6175
    },
    {
      "entropy": 6.120384788513183,
      "epoch": 0.5400681639430219,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004976888937187107,
      "loss": 6.0326,
      "mean_token_accuracy": 0.14698556512594224,
      "num_tokens": 12525389.0,
      "step": 6180
    },
    {
      "entropy": 6.238776445388794,
      "epoch": 0.5405051122957266,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004976844367816203,
      "loss": 6.1414,
      "mean_token_accuracy": 0.13036047220230101,
      "num_tokens": 12535454.0,
      "step": 6185
    },
    {
      "entropy": 6.278572750091553,
      "epoch": 0.5409420606484313,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004976799755733257,
      "loss": 5.9896,
      "mean_token_accuracy": 0.14532420784235,
      "num_tokens": 12546092.0,
      "step": 6190
    },
    {
      "entropy": 6.220078992843628,
      "epoch": 0.5413790090011361,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004976755100939126,
      "loss": 6.0993,
      "mean_token_accuracy": 0.13647042587399483,
      "num_tokens": 12555462.0,
      "step": 6195
    },
    {
      "entropy": 6.2149238109588625,
      "epoch": 0.5418159573538408,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004976710403434663,
      "loss": 6.1188,
      "mean_token_accuracy": 0.14242732897400856,
      "num_tokens": 12564506.0,
      "step": 6200
    },
    {
      "entropy": 6.283080196380615,
      "epoch": 0.5422529057065455,
      "grad_norm": 1.796875,
      "learning_rate": 0.000497666566322073,
      "loss": 6.1043,
      "mean_token_accuracy": 0.1390668608248234,
      "num_tokens": 12574887.0,
      "step": 6205
    },
    {
      "entropy": 6.363029432296753,
      "epoch": 0.5426898540592502,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004976620880298182,
      "loss": 6.1301,
      "mean_token_accuracy": 0.13858333677053453,
      "num_tokens": 12584499.0,
      "step": 6210
    },
    {
      "entropy": 6.247929334640503,
      "epoch": 0.5431268024119549,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004976576054667878,
      "loss": 6.0664,
      "mean_token_accuracy": 0.14071348756551744,
      "num_tokens": 12593796.0,
      "step": 6215
    },
    {
      "entropy": 6.159736394882202,
      "epoch": 0.5435637507646596,
      "grad_norm": 2.59375,
      "learning_rate": 0.000497653118633068,
      "loss": 6.1181,
      "mean_token_accuracy": 0.13587007597088813,
      "num_tokens": 12603719.0,
      "step": 6220
    },
    {
      "entropy": 6.186320877075195,
      "epoch": 0.5440006991173644,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004976486275287446,
      "loss": 6.1136,
      "mean_token_accuracy": 0.1381325379014015,
      "num_tokens": 12614407.0,
      "step": 6225
    },
    {
      "entropy": 6.255827951431274,
      "epoch": 0.5444376474700691,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004976441321539038,
      "loss": 6.1109,
      "mean_token_accuracy": 0.14508781358599662,
      "num_tokens": 12624414.0,
      "step": 6230
    },
    {
      "entropy": 6.289930629730224,
      "epoch": 0.5448745958227738,
      "grad_norm": 2.375,
      "learning_rate": 0.0004976396325086321,
      "loss": 6.1272,
      "mean_token_accuracy": 0.1362790621817112,
      "num_tokens": 12635679.0,
      "step": 6235
    },
    {
      "entropy": 6.303273487091064,
      "epoch": 0.5453115441754784,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004976351285930154,
      "loss": 6.0497,
      "mean_token_accuracy": 0.1359069027006626,
      "num_tokens": 12645253.0,
      "step": 6240
    },
    {
      "entropy": 6.189432621002197,
      "epoch": 0.5457484925281831,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004976306204071404,
      "loss": 6.0268,
      "mean_token_accuracy": 0.1512258157134056,
      "num_tokens": 12655815.0,
      "step": 6245
    },
    {
      "entropy": 6.201519250869751,
      "epoch": 0.5461854408808878,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004976261079510934,
      "loss": 6.1203,
      "mean_token_accuracy": 0.142380790412426,
      "num_tokens": 12666173.0,
      "step": 6250
    },
    {
      "entropy": 6.226180744171143,
      "epoch": 0.5466223892335926,
      "grad_norm": 2.03125,
      "learning_rate": 0.000497621591224961,
      "loss": 6.0546,
      "mean_token_accuracy": 0.14446704164147378,
      "num_tokens": 12676008.0,
      "step": 6255
    },
    {
      "entropy": 6.300201177597046,
      "epoch": 0.5470593375862973,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004976170702288298,
      "loss": 6.1232,
      "mean_token_accuracy": 0.13573255613446236,
      "num_tokens": 12686217.0,
      "step": 6260
    },
    {
      "entropy": 6.262121248245239,
      "epoch": 0.547496285939002,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004976125449627866,
      "loss": 6.1142,
      "mean_token_accuracy": 0.13377300202846526,
      "num_tokens": 12696809.0,
      "step": 6265
    },
    {
      "entropy": 6.294864988327026,
      "epoch": 0.5479332342917067,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004976080154269181,
      "loss": 6.1849,
      "mean_token_accuracy": 0.1314307503402233,
      "num_tokens": 12708328.0,
      "step": 6270
    },
    {
      "entropy": 6.319700622558594,
      "epoch": 0.5483701826444114,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004976034816213112,
      "loss": 6.027,
      "mean_token_accuracy": 0.13883933499455453,
      "num_tokens": 12718478.0,
      "step": 6275
    },
    {
      "entropy": 6.179039812088012,
      "epoch": 0.5488071309971161,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004975989435460528,
      "loss": 6.0823,
      "mean_token_accuracy": 0.14150369465351104,
      "num_tokens": 12728853.0,
      "step": 6280
    },
    {
      "entropy": 6.204391956329346,
      "epoch": 0.5492440793498209,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004975944012012302,
      "loss": 6.148,
      "mean_token_accuracy": 0.14521700143814087,
      "num_tokens": 12739036.0,
      "step": 6285
    },
    {
      "entropy": 6.317832612991333,
      "epoch": 0.5496810277025256,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004975898545869303,
      "loss": 6.0781,
      "mean_token_accuracy": 0.14573338478803635,
      "num_tokens": 12748908.0,
      "step": 6290
    },
    {
      "entropy": 6.303751611709595,
      "epoch": 0.5501179760552303,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004975853037032403,
      "loss": 6.1436,
      "mean_token_accuracy": 0.13691058456897737,
      "num_tokens": 12758671.0,
      "step": 6295
    },
    {
      "entropy": 6.165650987625122,
      "epoch": 0.550554924407935,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004975807485502476,
      "loss": 6.1135,
      "mean_token_accuracy": 0.13557566404342652,
      "num_tokens": 12769552.0,
      "step": 6300
    },
    {
      "entropy": 6.271069431304932,
      "epoch": 0.5509918727606397,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004975761891280394,
      "loss": 6.2495,
      "mean_token_accuracy": 0.12941641584038735,
      "num_tokens": 12781032.0,
      "step": 6305
    },
    {
      "entropy": 6.3879612445831295,
      "epoch": 0.5514288211133443,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004975716254367033,
      "loss": 6.1693,
      "mean_token_accuracy": 0.13546326085925103,
      "num_tokens": 12790818.0,
      "step": 6310
    },
    {
      "entropy": 6.262982940673828,
      "epoch": 0.5518657694660491,
      "grad_norm": 1.9765625,
      "learning_rate": 0.000497567057476327,
      "loss": 6.0399,
      "mean_token_accuracy": 0.14123047664761543,
      "num_tokens": 12799701.0,
      "step": 6315
    },
    {
      "entropy": 6.253837537765503,
      "epoch": 0.5523027178187538,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004975624852469976,
      "loss": 6.1763,
      "mean_token_accuracy": 0.14434364140033723,
      "num_tokens": 12809575.0,
      "step": 6320
    },
    {
      "entropy": 6.324242067337036,
      "epoch": 0.5527396661714585,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004975579087488033,
      "loss": 6.1198,
      "mean_token_accuracy": 0.13896695226430894,
      "num_tokens": 12820278.0,
      "step": 6325
    },
    {
      "entropy": 6.2966714859008786,
      "epoch": 0.5531766145241632,
      "grad_norm": 2.875,
      "learning_rate": 0.0004975533279818317,
      "loss": 6.1065,
      "mean_token_accuracy": 0.13830024525523185,
      "num_tokens": 12830774.0,
      "step": 6330
    },
    {
      "entropy": 6.179326057434082,
      "epoch": 0.5536135628768679,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004975487429461706,
      "loss": 6.0738,
      "mean_token_accuracy": 0.14779838770627976,
      "num_tokens": 12839487.0,
      "step": 6335
    },
    {
      "entropy": 6.211460542678833,
      "epoch": 0.5540505112295727,
      "grad_norm": 1.96875,
      "learning_rate": 0.000497544153641908,
      "loss": 6.0443,
      "mean_token_accuracy": 0.13948678150773047,
      "num_tokens": 12849614.0,
      "step": 6340
    },
    {
      "entropy": 6.282157468795776,
      "epoch": 0.5544874595822774,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004975395600691319,
      "loss": 6.1558,
      "mean_token_accuracy": 0.14338527768850326,
      "num_tokens": 12859428.0,
      "step": 6345
    },
    {
      "entropy": 6.241535282135009,
      "epoch": 0.5549244079349821,
      "grad_norm": 2.625,
      "learning_rate": 0.0004975349622279305,
      "loss": 6.1502,
      "mean_token_accuracy": 0.1375342957675457,
      "num_tokens": 12869167.0,
      "step": 6350
    },
    {
      "entropy": 6.230395698547364,
      "epoch": 0.5553613562876868,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004975303601183919,
      "loss": 6.0852,
      "mean_token_accuracy": 0.14296441450715064,
      "num_tokens": 12878410.0,
      "step": 6355
    },
    {
      "entropy": 6.313714265823364,
      "epoch": 0.5557983046403915,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004975257537406043,
      "loss": 6.0971,
      "mean_token_accuracy": 0.13664701357483863,
      "num_tokens": 12888372.0,
      "step": 6360
    },
    {
      "entropy": 6.228673601150513,
      "epoch": 0.5562352529930962,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004975211430946563,
      "loss": 6.1475,
      "mean_token_accuracy": 0.1400043822824955,
      "num_tokens": 12897958.0,
      "step": 6365
    },
    {
      "entropy": 6.231792783737182,
      "epoch": 0.556672201345801,
      "grad_norm": 2.21875,
      "learning_rate": 0.000497516528180636,
      "loss": 6.1409,
      "mean_token_accuracy": 0.1417572058737278,
      "num_tokens": 12907890.0,
      "step": 6370
    },
    {
      "entropy": 6.292871379852295,
      "epoch": 0.5571091496985057,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004975119089986322,
      "loss": 6.0825,
      "mean_token_accuracy": 0.13979636877775192,
      "num_tokens": 12917914.0,
      "step": 6375
    },
    {
      "entropy": 6.238379096984863,
      "epoch": 0.5575460980512104,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004975072855487333,
      "loss": 6.1133,
      "mean_token_accuracy": 0.14305043667554856,
      "num_tokens": 12927090.0,
      "step": 6380
    },
    {
      "entropy": 6.2708868980407715,
      "epoch": 0.557983046403915,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004975026578310282,
      "loss": 6.0967,
      "mean_token_accuracy": 0.1360911414027214,
      "num_tokens": 12937236.0,
      "step": 6385
    },
    {
      "entropy": 6.29693112373352,
      "epoch": 0.5584199947566197,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004974980258456055,
      "loss": 6.0469,
      "mean_token_accuracy": 0.1430341236293316,
      "num_tokens": 12947635.0,
      "step": 6390
    },
    {
      "entropy": 6.304951333999634,
      "epoch": 0.5588569431093244,
      "grad_norm": 2.515625,
      "learning_rate": 0.000497493389592554,
      "loss": 6.0122,
      "mean_token_accuracy": 0.14151546210050583,
      "num_tokens": 12957236.0,
      "step": 6395
    },
    {
      "entropy": 6.261232042312622,
      "epoch": 0.5592938914620292,
      "grad_norm": 2.25,
      "learning_rate": 0.0004974887490719627,
      "loss": 6.1415,
      "mean_token_accuracy": 0.13431976288557052,
      "num_tokens": 12967497.0,
      "step": 6400
    },
    {
      "entropy": 6.239714336395264,
      "epoch": 0.5597308398147339,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004974841042839207,
      "loss": 6.0611,
      "mean_token_accuracy": 0.14347777962684632,
      "num_tokens": 12977467.0,
      "step": 6405
    },
    {
      "entropy": 6.184810781478882,
      "epoch": 0.5601677881674386,
      "grad_norm": 1.828125,
      "learning_rate": 0.000497479455228517,
      "loss": 6.1154,
      "mean_token_accuracy": 0.1409076377749443,
      "num_tokens": 12987867.0,
      "step": 6410
    },
    {
      "entropy": 6.262886953353882,
      "epoch": 0.5606047365201433,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004974748019058407,
      "loss": 6.1303,
      "mean_token_accuracy": 0.13358882069587708,
      "num_tokens": 12998081.0,
      "step": 6415
    },
    {
      "entropy": 6.270763158798218,
      "epoch": 0.561041684872848,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004974701443159812,
      "loss": 6.121,
      "mean_token_accuracy": 0.1384132459759712,
      "num_tokens": 13008047.0,
      "step": 6420
    },
    {
      "entropy": 6.324219751358032,
      "epoch": 0.5614786332255527,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004974654824590277,
      "loss": 6.106,
      "mean_token_accuracy": 0.13544400185346603,
      "num_tokens": 13018048.0,
      "step": 6425
    },
    {
      "entropy": 6.262833309173584,
      "epoch": 0.5619155815782575,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004974608163350697,
      "loss": 6.0846,
      "mean_token_accuracy": 0.1436422161757946,
      "num_tokens": 13028156.0,
      "step": 6430
    },
    {
      "entropy": 6.209911346435547,
      "epoch": 0.5623525299309622,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004974561459441967,
      "loss": 6.1046,
      "mean_token_accuracy": 0.13928409665822983,
      "num_tokens": 13038191.0,
      "step": 6435
    },
    {
      "entropy": 6.233520221710205,
      "epoch": 0.5627894782836669,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004974514712864981,
      "loss": 6.1073,
      "mean_token_accuracy": 0.138884087651968,
      "num_tokens": 13049085.0,
      "step": 6440
    },
    {
      "entropy": 6.2541821002960205,
      "epoch": 0.5632264266363716,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004974467923620639,
      "loss": 6.1268,
      "mean_token_accuracy": 0.1373557224869728,
      "num_tokens": 13060010.0,
      "step": 6445
    },
    {
      "entropy": 6.349029874801635,
      "epoch": 0.5636633749890763,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004974421091709836,
      "loss": 6.0994,
      "mean_token_accuracy": 0.1365789458155632,
      "num_tokens": 13069768.0,
      "step": 6450
    },
    {
      "entropy": 6.315038061141967,
      "epoch": 0.564100323341781,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004974374217133471,
      "loss": 6.1204,
      "mean_token_accuracy": 0.135160880535841,
      "num_tokens": 13079842.0,
      "step": 6455
    },
    {
      "entropy": 6.214017534255982,
      "epoch": 0.5645372716944858,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004974327299892443,
      "loss": 6.0578,
      "mean_token_accuracy": 0.14208607152104377,
      "num_tokens": 13089235.0,
      "step": 6460
    },
    {
      "entropy": 6.231755542755127,
      "epoch": 0.5649742200471904,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004974280339987651,
      "loss": 6.0935,
      "mean_token_accuracy": 0.13741404563188553,
      "num_tokens": 13098364.0,
      "step": 6465
    },
    {
      "entropy": 6.2427702903747555,
      "epoch": 0.5654111683998951,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004974233337419997,
      "loss": 6.2171,
      "mean_token_accuracy": 0.12881077006459235,
      "num_tokens": 13108876.0,
      "step": 6470
    },
    {
      "entropy": 6.329076623916626,
      "epoch": 0.5658481167525998,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004974186292190383,
      "loss": 6.2221,
      "mean_token_accuracy": 0.13388051688671113,
      "num_tokens": 13119591.0,
      "step": 6475
    },
    {
      "entropy": 6.32397837638855,
      "epoch": 0.5662850651053045,
      "grad_norm": 2.25,
      "learning_rate": 0.000497413920429971,
      "loss": 6.1107,
      "mean_token_accuracy": 0.14448174014687537,
      "num_tokens": 13129796.0,
      "step": 6480
    },
    {
      "entropy": 6.2648690700531,
      "epoch": 0.5667220134580092,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004974092073748882,
      "loss": 6.0587,
      "mean_token_accuracy": 0.142123019695282,
      "num_tokens": 13139302.0,
      "step": 6485
    },
    {
      "entropy": 6.228122711181641,
      "epoch": 0.567158961810714,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004974044900538802,
      "loss": 6.1044,
      "mean_token_accuracy": 0.13770891204476357,
      "num_tokens": 13148983.0,
      "step": 6490
    },
    {
      "entropy": 6.3073498725891115,
      "epoch": 0.5675959101634187,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004973997684670375,
      "loss": 6.1079,
      "mean_token_accuracy": 0.13494180366396905,
      "num_tokens": 13159016.0,
      "step": 6495
    },
    {
      "entropy": 6.2961560726165775,
      "epoch": 0.5680328585161234,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004973950426144508,
      "loss": 6.1339,
      "mean_token_accuracy": 0.13156452551484107,
      "num_tokens": 13169705.0,
      "step": 6500
    },
    {
      "entropy": 6.255215787887574,
      "epoch": 0.5684698068688281,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004973903124962106,
      "loss": 6.1086,
      "mean_token_accuracy": 0.13032817989587783,
      "num_tokens": 13181372.0,
      "step": 6505
    },
    {
      "entropy": 6.190559673309326,
      "epoch": 0.5689067552215328,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004973855781124077,
      "loss": 6.171,
      "mean_token_accuracy": 0.13174699544906615,
      "num_tokens": 13192581.0,
      "step": 6510
    },
    {
      "entropy": 6.302378129959107,
      "epoch": 0.5693437035742375,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004973808394631329,
      "loss": 6.192,
      "mean_token_accuracy": 0.135031296312809,
      "num_tokens": 13203269.0,
      "step": 6515
    },
    {
      "entropy": 6.276038312911988,
      "epoch": 0.5697806519269423,
      "grad_norm": 2.1875,
      "learning_rate": 0.000497376096548477,
      "loss": 6.083,
      "mean_token_accuracy": 0.13695611134171487,
      "num_tokens": 13213183.0,
      "step": 6520
    },
    {
      "entropy": 6.272439908981323,
      "epoch": 0.570217600279647,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004973713493685311,
      "loss": 6.1495,
      "mean_token_accuracy": 0.13435291722416878,
      "num_tokens": 13223196.0,
      "step": 6525
    },
    {
      "entropy": 6.315483617782593,
      "epoch": 0.5706545486323517,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004973665979233861,
      "loss": 6.0792,
      "mean_token_accuracy": 0.14457561075687408,
      "num_tokens": 13233915.0,
      "step": 6530
    },
    {
      "entropy": 6.278499937057495,
      "epoch": 0.5710914969850563,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004973618422131334,
      "loss": 6.0987,
      "mean_token_accuracy": 0.13833219408988953,
      "num_tokens": 13245108.0,
      "step": 6535
    },
    {
      "entropy": 6.187056827545166,
      "epoch": 0.571528445337761,
      "grad_norm": 2.21875,
      "learning_rate": 0.000497357082237864,
      "loss": 6.1186,
      "mean_token_accuracy": 0.13869462460279464,
      "num_tokens": 13255508.0,
      "step": 6540
    },
    {
      "entropy": 6.274870586395264,
      "epoch": 0.5719653936904658,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004973523179976692,
      "loss": 6.15,
      "mean_token_accuracy": 0.14257874116301536,
      "num_tokens": 13265118.0,
      "step": 6545
    },
    {
      "entropy": 6.288551330566406,
      "epoch": 0.5724023420431705,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004973475494926404,
      "loss": 6.0539,
      "mean_token_accuracy": 0.14106326252222062,
      "num_tokens": 13275947.0,
      "step": 6550
    },
    {
      "entropy": 6.276229333877564,
      "epoch": 0.5728392903958752,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004973427767228692,
      "loss": 6.0666,
      "mean_token_accuracy": 0.14221703708171846,
      "num_tokens": 13284662.0,
      "step": 6555
    },
    {
      "entropy": 6.192042350769043,
      "epoch": 0.5732762387485799,
      "grad_norm": 2.0,
      "learning_rate": 0.000497337999688447,
      "loss": 6.1372,
      "mean_token_accuracy": 0.1390274167060852,
      "num_tokens": 13295627.0,
      "step": 6560
    },
    {
      "entropy": 6.261907720565796,
      "epoch": 0.5737131871012846,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004973332183894653,
      "loss": 6.0564,
      "mean_token_accuracy": 0.1444385051727295,
      "num_tokens": 13305205.0,
      "step": 6565
    },
    {
      "entropy": 6.24415340423584,
      "epoch": 0.5741501354539893,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004973284328260162,
      "loss": 6.0837,
      "mean_token_accuracy": 0.13924380391836166,
      "num_tokens": 13315464.0,
      "step": 6570
    },
    {
      "entropy": 6.187749910354614,
      "epoch": 0.5745870838066941,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004973236429981912,
      "loss": 5.9966,
      "mean_token_accuracy": 0.1391263373196125,
      "num_tokens": 13325468.0,
      "step": 6575
    },
    {
      "entropy": 6.241669321060181,
      "epoch": 0.5750240321593988,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004973188489060821,
      "loss": 6.1219,
      "mean_token_accuracy": 0.1372533343732357,
      "num_tokens": 13334719.0,
      "step": 6580
    },
    {
      "entropy": 6.264699459075928,
      "epoch": 0.5754609805121035,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004973140505497812,
      "loss": 6.0766,
      "mean_token_accuracy": 0.13733603060245514,
      "num_tokens": 13344545.0,
      "step": 6585
    },
    {
      "entropy": 6.309505987167358,
      "epoch": 0.5758979288648082,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004973092479293802,
      "loss": 6.15,
      "mean_token_accuracy": 0.13204444646835328,
      "num_tokens": 13355212.0,
      "step": 6590
    },
    {
      "entropy": 6.189426136016846,
      "epoch": 0.5763348772175129,
      "grad_norm": 2.125,
      "learning_rate": 0.0004973044410449713,
      "loss": 6.0509,
      "mean_token_accuracy": 0.1430459849536419,
      "num_tokens": 13364869.0,
      "step": 6595
    },
    {
      "entropy": 6.240922355651856,
      "epoch": 0.5767718255702176,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004972996298966468,
      "loss": 6.0806,
      "mean_token_accuracy": 0.14123327955603598,
      "num_tokens": 13376511.0,
      "step": 6600
    },
    {
      "entropy": 6.2815958023071286,
      "epoch": 0.5772087739229224,
      "grad_norm": 2.5,
      "learning_rate": 0.000497294814484499,
      "loss": 6.1152,
      "mean_token_accuracy": 0.14026200994849206,
      "num_tokens": 13385470.0,
      "step": 6605
    },
    {
      "entropy": 6.271773958206177,
      "epoch": 0.577645722275627,
      "grad_norm": 1.90625,
      "learning_rate": 0.00049728999480862,
      "loss": 6.17,
      "mean_token_accuracy": 0.14066533967852593,
      "num_tokens": 13395328.0,
      "step": 6610
    },
    {
      "entropy": 6.319705438613892,
      "epoch": 0.5780826706283317,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004972851708691025,
      "loss": 6.0677,
      "mean_token_accuracy": 0.1438952326774597,
      "num_tokens": 13405319.0,
      "step": 6615
    },
    {
      "entropy": 6.200738954544067,
      "epoch": 0.5785196189810364,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004972803426660389,
      "loss": 6.0668,
      "mean_token_accuracy": 0.13934995457530022,
      "num_tokens": 13416119.0,
      "step": 6620
    },
    {
      "entropy": 6.258353757858276,
      "epoch": 0.5789565673337411,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004972755101995219,
      "loss": 6.071,
      "mean_token_accuracy": 0.1352447398006916,
      "num_tokens": 13425824.0,
      "step": 6625
    },
    {
      "entropy": 6.244670963287353,
      "epoch": 0.5793935156864458,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004972706734696442,
      "loss": 6.0222,
      "mean_token_accuracy": 0.14621324539184571,
      "num_tokens": 13435506.0,
      "step": 6630
    },
    {
      "entropy": 6.228378963470459,
      "epoch": 0.5798304640391506,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004972658324764984,
      "loss": 6.0556,
      "mean_token_accuracy": 0.14824110716581346,
      "num_tokens": 13445313.0,
      "step": 6635
    },
    {
      "entropy": 6.13565616607666,
      "epoch": 0.5802674123918553,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004972609872201775,
      "loss": 5.9809,
      "mean_token_accuracy": 0.14731499776244164,
      "num_tokens": 13455312.0,
      "step": 6640
    },
    {
      "entropy": 6.213436412811279,
      "epoch": 0.58070436074456,
      "grad_norm": 1.75,
      "learning_rate": 0.0004972561377007744,
      "loss": 6.1353,
      "mean_token_accuracy": 0.13471954986453055,
      "num_tokens": 13465721.0,
      "step": 6645
    },
    {
      "entropy": 6.3089337825775145,
      "epoch": 0.5811413090972647,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004972512839183822,
      "loss": 6.1479,
      "mean_token_accuracy": 0.14083665534853934,
      "num_tokens": 13477185.0,
      "step": 6650
    },
    {
      "entropy": 6.232909917831421,
      "epoch": 0.5815782574499694,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004972464258730939,
      "loss": 6.0548,
      "mean_token_accuracy": 0.15199510008096695,
      "num_tokens": 13487846.0,
      "step": 6655
    },
    {
      "entropy": 6.185512924194336,
      "epoch": 0.5820152058026741,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004972415635650027,
      "loss": 6.1292,
      "mean_token_accuracy": 0.14809978529810905,
      "num_tokens": 13498175.0,
      "step": 6660
    },
    {
      "entropy": 6.373633527755738,
      "epoch": 0.5824521541553789,
      "grad_norm": 2.25,
      "learning_rate": 0.0004972366969942017,
      "loss": 6.0972,
      "mean_token_accuracy": 0.13438104838132858,
      "num_tokens": 13506864.0,
      "step": 6665
    },
    {
      "entropy": 6.240647315979004,
      "epoch": 0.5828891025080836,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004972318261607845,
      "loss": 6.0806,
      "mean_token_accuracy": 0.14570387601852416,
      "num_tokens": 13516208.0,
      "step": 6670
    },
    {
      "entropy": 6.246371078491211,
      "epoch": 0.5833260508607883,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004972269510648444,
      "loss": 6.1324,
      "mean_token_accuracy": 0.13572923019528388,
      "num_tokens": 13526902.0,
      "step": 6675
    },
    {
      "entropy": 6.275639009475708,
      "epoch": 0.583762999213493,
      "grad_norm": 1.7890625,
      "learning_rate": 0.000497222071706475,
      "loss": 6.0575,
      "mean_token_accuracy": 0.14732615202665328,
      "num_tokens": 13537001.0,
      "step": 6680
    },
    {
      "entropy": 6.204616689682007,
      "epoch": 0.5841999475661976,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004972171880857698,
      "loss": 6.0409,
      "mean_token_accuracy": 0.14583269581198693,
      "num_tokens": 13546971.0,
      "step": 6685
    },
    {
      "entropy": 6.217172479629516,
      "epoch": 0.5846368959189023,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004972123002028226,
      "loss": 6.0256,
      "mean_token_accuracy": 0.14713723808526993,
      "num_tokens": 13557202.0,
      "step": 6690
    },
    {
      "entropy": 6.122836589813232,
      "epoch": 0.5850738442716071,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004972074080577269,
      "loss": 6.0035,
      "mean_token_accuracy": 0.15193717703223228,
      "num_tokens": 13567291.0,
      "step": 6695
    },
    {
      "entropy": 6.241865825653076,
      "epoch": 0.5855107926243118,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004972025116505767,
      "loss": 6.0977,
      "mean_token_accuracy": 0.1464373327791691,
      "num_tokens": 13576864.0,
      "step": 6700
    },
    {
      "entropy": 6.275596284866333,
      "epoch": 0.5859477409770165,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004971976109814658,
      "loss": 6.0358,
      "mean_token_accuracy": 0.14644457399845123,
      "num_tokens": 13586728.0,
      "step": 6705
    },
    {
      "entropy": 6.211604118347168,
      "epoch": 0.5863846893297212,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004971927060504885,
      "loss": 6.092,
      "mean_token_accuracy": 0.1458321511745453,
      "num_tokens": 13596875.0,
      "step": 6710
    },
    {
      "entropy": 6.285997438430786,
      "epoch": 0.5868216376824259,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004971877968577386,
      "loss": 6.1062,
      "mean_token_accuracy": 0.13368350118398667,
      "num_tokens": 13606483.0,
      "step": 6715
    },
    {
      "entropy": 6.256731033325195,
      "epoch": 0.5872585860351307,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004971828834033102,
      "loss": 6.145,
      "mean_token_accuracy": 0.13932919576764108,
      "num_tokens": 13616780.0,
      "step": 6720
    },
    {
      "entropy": 6.237018632888794,
      "epoch": 0.5876955343878354,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004971779656872978,
      "loss": 6.1256,
      "mean_token_accuracy": 0.13926877230405807,
      "num_tokens": 13626418.0,
      "step": 6725
    },
    {
      "entropy": 6.289210176467895,
      "epoch": 0.5881324827405401,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004971730437097955,
      "loss": 6.1592,
      "mean_token_accuracy": 0.13757795691490174,
      "num_tokens": 13636956.0,
      "step": 6730
    },
    {
      "entropy": 6.232604026794434,
      "epoch": 0.5885694310932448,
      "grad_norm": 2.5,
      "learning_rate": 0.000497168117470898,
      "loss": 6.0869,
      "mean_token_accuracy": 0.1386633425951004,
      "num_tokens": 13647458.0,
      "step": 6735
    },
    {
      "entropy": 6.206624937057495,
      "epoch": 0.5890063794459495,
      "grad_norm": 1.875,
      "learning_rate": 0.0004971631869706995,
      "loss": 6.035,
      "mean_token_accuracy": 0.14804055988788606,
      "num_tokens": 13657655.0,
      "step": 6740
    },
    {
      "entropy": 6.239219284057617,
      "epoch": 0.5894433277986542,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004971582522092946,
      "loss": 6.1416,
      "mean_token_accuracy": 0.13588550686836243,
      "num_tokens": 13668472.0,
      "step": 6745
    },
    {
      "entropy": 6.236002111434937,
      "epoch": 0.589880276151359,
      "grad_norm": 2.25,
      "learning_rate": 0.000497153313186778,
      "loss": 5.9967,
      "mean_token_accuracy": 0.14495814442634583,
      "num_tokens": 13677319.0,
      "step": 6750
    },
    {
      "entropy": 6.204773950576782,
      "epoch": 0.5903172245040637,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004971483699032445,
      "loss": 6.0708,
      "mean_token_accuracy": 0.14527754485607147,
      "num_tokens": 13688037.0,
      "step": 6755
    },
    {
      "entropy": 6.345049619674683,
      "epoch": 0.5907541728567683,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004971434223587888,
      "loss": 6.1769,
      "mean_token_accuracy": 0.1332503601908684,
      "num_tokens": 13698441.0,
      "step": 6760
    },
    {
      "entropy": 6.232329988479615,
      "epoch": 0.591191121209473,
      "grad_norm": 2.0625,
      "learning_rate": 0.000497138470553506,
      "loss": 6.0971,
      "mean_token_accuracy": 0.1420059993863106,
      "num_tokens": 13708140.0,
      "step": 6765
    },
    {
      "entropy": 6.2557373046875,
      "epoch": 0.5916280695621777,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004971335144874909,
      "loss": 6.0795,
      "mean_token_accuracy": 0.14316331669688226,
      "num_tokens": 13718069.0,
      "step": 6770
    },
    {
      "entropy": 6.228231620788574,
      "epoch": 0.5920650179148824,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004971285541608385,
      "loss": 6.0625,
      "mean_token_accuracy": 0.15345999151468276,
      "num_tokens": 13727932.0,
      "step": 6775
    },
    {
      "entropy": 6.295587253570557,
      "epoch": 0.5925019662675872,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004971235895736441,
      "loss": 6.1376,
      "mean_token_accuracy": 0.13220032975077628,
      "num_tokens": 13738239.0,
      "step": 6780
    },
    {
      "entropy": 6.2889244079589846,
      "epoch": 0.5929389146202919,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004971186207260029,
      "loss": 6.0799,
      "mean_token_accuracy": 0.1396336004137993,
      "num_tokens": 13748622.0,
      "step": 6785
    },
    {
      "entropy": 6.160888195037842,
      "epoch": 0.5933758629729966,
      "grad_norm": 2.359375,
      "learning_rate": 0.00049711364761801,
      "loss": 6.0108,
      "mean_token_accuracy": 0.14156231358647348,
      "num_tokens": 13759221.0,
      "step": 6790
    },
    {
      "entropy": 6.237045097351074,
      "epoch": 0.5938128113257013,
      "grad_norm": 1.8125,
      "learning_rate": 0.000497108670249761,
      "loss": 6.0398,
      "mean_token_accuracy": 0.13876911327242852,
      "num_tokens": 13768343.0,
      "step": 6795
    },
    {
      "entropy": 6.276476001739502,
      "epoch": 0.594249759678406,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004971036886213515,
      "loss": 6.0394,
      "mean_token_accuracy": 0.14211647883057593,
      "num_tokens": 13778225.0,
      "step": 6800
    },
    {
      "entropy": 6.224419498443604,
      "epoch": 0.5946867080311107,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004970987027328767,
      "loss": 6.1142,
      "mean_token_accuracy": 0.13437844887375833,
      "num_tokens": 13788633.0,
      "step": 6805
    },
    {
      "entropy": 6.241243839263916,
      "epoch": 0.5951236563838155,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004970937125844325,
      "loss": 6.0122,
      "mean_token_accuracy": 0.14751050025224685,
      "num_tokens": 13798172.0,
      "step": 6810
    },
    {
      "entropy": 6.139668369293213,
      "epoch": 0.5955606047365202,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004970887181761144,
      "loss": 6.0118,
      "mean_token_accuracy": 0.15298185646533966,
      "num_tokens": 13807284.0,
      "step": 6815
    },
    {
      "entropy": 6.237003898620605,
      "epoch": 0.5959975530892249,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004970837195080184,
      "loss": 6.0931,
      "mean_token_accuracy": 0.13770748674869537,
      "num_tokens": 13816807.0,
      "step": 6820
    },
    {
      "entropy": 6.288899850845337,
      "epoch": 0.5964345014419296,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004970787165802403,
      "loss": 6.037,
      "mean_token_accuracy": 0.1411267563700676,
      "num_tokens": 13827635.0,
      "step": 6825
    },
    {
      "entropy": 6.234393358230591,
      "epoch": 0.5968714497946342,
      "grad_norm": 2.453125,
      "learning_rate": 0.000497073709392876,
      "loss": 6.1074,
      "mean_token_accuracy": 0.13737359121441842,
      "num_tokens": 13838060.0,
      "step": 6830
    },
    {
      "entropy": 6.285126733779907,
      "epoch": 0.5973083981473389,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004970686979460216,
      "loss": 6.1196,
      "mean_token_accuracy": 0.13926485776901246,
      "num_tokens": 13848669.0,
      "step": 6835
    },
    {
      "entropy": 6.260535764694214,
      "epoch": 0.5977453465000437,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004970636822397733,
      "loss": 6.0823,
      "mean_token_accuracy": 0.14194415062665938,
      "num_tokens": 13857949.0,
      "step": 6840
    },
    {
      "entropy": 6.241445779800415,
      "epoch": 0.5981822948527484,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004970586622742272,
      "loss": 6.0764,
      "mean_token_accuracy": 0.14263241216540337,
      "num_tokens": 13867726.0,
      "step": 6845
    },
    {
      "entropy": 6.324937200546264,
      "epoch": 0.5986192432054531,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004970536380494795,
      "loss": 6.1466,
      "mean_token_accuracy": 0.13447364792227745,
      "num_tokens": 13878072.0,
      "step": 6850
    },
    {
      "entropy": 6.193201541900635,
      "epoch": 0.5990561915581578,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004970486095656268,
      "loss": 6.0124,
      "mean_token_accuracy": 0.14746544361114503,
      "num_tokens": 13887771.0,
      "step": 6855
    },
    {
      "entropy": 6.231291103363037,
      "epoch": 0.5994931399108625,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004970435768227652,
      "loss": 6.0496,
      "mean_token_accuracy": 0.14115117266774177,
      "num_tokens": 13897867.0,
      "step": 6860
    },
    {
      "entropy": 6.28660569190979,
      "epoch": 0.5999300882635672,
      "grad_norm": 2.75,
      "learning_rate": 0.0004970385398209916,
      "loss": 6.128,
      "mean_token_accuracy": 0.14163656905293465,
      "num_tokens": 13909171.0,
      "step": 6865
    },
    {
      "entropy": 6.300090360641479,
      "epoch": 0.600367036616272,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004970334985604024,
      "loss": 6.1142,
      "mean_token_accuracy": 0.1396082989871502,
      "num_tokens": 13920374.0,
      "step": 6870
    },
    {
      "entropy": 6.193111419677734,
      "epoch": 0.6008039849689767,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004970284530410945,
      "loss": 6.1301,
      "mean_token_accuracy": 0.13556758910417557,
      "num_tokens": 13930150.0,
      "step": 6875
    },
    {
      "entropy": 6.279803276062012,
      "epoch": 0.6012409333216814,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004970234032631644,
      "loss": 6.0689,
      "mean_token_accuracy": 0.14032774344086646,
      "num_tokens": 13940553.0,
      "step": 6880
    },
    {
      "entropy": 6.191017389297485,
      "epoch": 0.6016778816743861,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004970183492267091,
      "loss": 6.1629,
      "mean_token_accuracy": 0.14311335012316703,
      "num_tokens": 13951086.0,
      "step": 6885
    },
    {
      "entropy": 6.236486196517944,
      "epoch": 0.6021148300270908,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004970132909318254,
      "loss": 6.0885,
      "mean_token_accuracy": 0.13655531257390977,
      "num_tokens": 13960684.0,
      "step": 6890
    },
    {
      "entropy": 6.116621685028076,
      "epoch": 0.6025517783797955,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004970082283786107,
      "loss": 5.9097,
      "mean_token_accuracy": 0.16360336542129517,
      "num_tokens": 13971409.0,
      "step": 6895
    },
    {
      "entropy": 6.257471227645874,
      "epoch": 0.6029887267325003,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004970031615671617,
      "loss": 6.094,
      "mean_token_accuracy": 0.14076051041483878,
      "num_tokens": 13981185.0,
      "step": 6900
    },
    {
      "entropy": 6.241898155212402,
      "epoch": 0.603425675085205,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004969980904975758,
      "loss": 6.1141,
      "mean_token_accuracy": 0.1358635351061821,
      "num_tokens": 13991670.0,
      "step": 6905
    },
    {
      "entropy": 6.14874758720398,
      "epoch": 0.6038626234379096,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004969930151699501,
      "loss": 6.067,
      "mean_token_accuracy": 0.14061281681060792,
      "num_tokens": 14002343.0,
      "step": 6910
    },
    {
      "entropy": 6.25538477897644,
      "epoch": 0.6042995717906143,
      "grad_norm": 1.953125,
      "learning_rate": 0.000496987935584382,
      "loss": 6.1392,
      "mean_token_accuracy": 0.13818174377083778,
      "num_tokens": 14013373.0,
      "step": 6915
    },
    {
      "entropy": 6.32241735458374,
      "epoch": 0.604736520143319,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004969828517409693,
      "loss": 6.0379,
      "mean_token_accuracy": 0.14847719222307204,
      "num_tokens": 14023233.0,
      "step": 6920
    },
    {
      "entropy": 6.1616730213165285,
      "epoch": 0.6051734684960238,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000496977763639809,
      "loss": 6.0843,
      "mean_token_accuracy": 0.1400049977004528,
      "num_tokens": 14033054.0,
      "step": 6925
    },
    {
      "entropy": 6.221465730667115,
      "epoch": 0.6056104168487285,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004969726712809988,
      "loss": 6.0258,
      "mean_token_accuracy": 0.1430351883172989,
      "num_tokens": 14042039.0,
      "step": 6930
    },
    {
      "entropy": 6.287037944793701,
      "epoch": 0.6060473652014332,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004969675746646364,
      "loss": 6.1629,
      "mean_token_accuracy": 0.13481699004769326,
      "num_tokens": 14051957.0,
      "step": 6935
    },
    {
      "entropy": 6.261262130737305,
      "epoch": 0.6064843135541379,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004969624737908197,
      "loss": 6.0563,
      "mean_token_accuracy": 0.1410438358783722,
      "num_tokens": 14062438.0,
      "step": 6940
    },
    {
      "entropy": 6.217139196395874,
      "epoch": 0.6069212619068426,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004969573686596464,
      "loss": 6.0822,
      "mean_token_accuracy": 0.13399507775902747,
      "num_tokens": 14073196.0,
      "step": 6945
    },
    {
      "entropy": 6.263551664352417,
      "epoch": 0.6073582102595473,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004969522592712146,
      "loss": 6.0711,
      "mean_token_accuracy": 0.14046932458877565,
      "num_tokens": 14083462.0,
      "step": 6950
    },
    {
      "entropy": 6.2073393821716305,
      "epoch": 0.6077951586122521,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004969471456256221,
      "loss": 6.1623,
      "mean_token_accuracy": 0.1367241285741329,
      "num_tokens": 14094194.0,
      "step": 6955
    },
    {
      "entropy": 6.240724849700928,
      "epoch": 0.6082321069649568,
      "grad_norm": 1.8984375,
      "learning_rate": 0.000496942027722967,
      "loss": 6.0842,
      "mean_token_accuracy": 0.14098508134484292,
      "num_tokens": 14104553.0,
      "step": 6960
    },
    {
      "entropy": 6.284235906600952,
      "epoch": 0.6086690553176615,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004969369055633476,
      "loss": 6.0435,
      "mean_token_accuracy": 0.13823190033435823,
      "num_tokens": 14114140.0,
      "step": 6965
    },
    {
      "entropy": 6.169156074523926,
      "epoch": 0.6091060036703662,
      "grad_norm": 2.234375,
      "learning_rate": 0.000496931779146862,
      "loss": 6.134,
      "mean_token_accuracy": 0.13523512482643127,
      "num_tokens": 14124723.0,
      "step": 6970
    },
    {
      "entropy": 6.283691024780273,
      "epoch": 0.6095429520230708,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004969266484736085,
      "loss": 6.0109,
      "mean_token_accuracy": 0.14279836937785148,
      "num_tokens": 14134030.0,
      "step": 6975
    },
    {
      "entropy": 6.196794176101685,
      "epoch": 0.6099799003757755,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004969215135436858,
      "loss": 5.9914,
      "mean_token_accuracy": 0.14249944314360619,
      "num_tokens": 14144691.0,
      "step": 6980
    },
    {
      "entropy": 6.203350162506103,
      "epoch": 0.6104168487284803,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004969163743571921,
      "loss": 6.162,
      "mean_token_accuracy": 0.13822323605418205,
      "num_tokens": 14155114.0,
      "step": 6985
    },
    {
      "entropy": 6.333395957946777,
      "epoch": 0.610853797081185,
      "grad_norm": 1.8671875,
      "learning_rate": 0.000496911230914226,
      "loss": 6.1543,
      "mean_token_accuracy": 0.13629033267498017,
      "num_tokens": 14166097.0,
      "step": 6990
    },
    {
      "entropy": 6.242578315734863,
      "epoch": 0.6112907454338897,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004969060832148862,
      "loss": 6.0253,
      "mean_token_accuracy": 0.13742341250181198,
      "num_tokens": 14176538.0,
      "step": 6995
    },
    {
      "entropy": 6.110963773727417,
      "epoch": 0.6117276937865944,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004969009312592715,
      "loss": 6.0637,
      "mean_token_accuracy": 0.1376984901726246,
      "num_tokens": 14187912.0,
      "step": 7000
    },
    {
      "entropy": 6.152584171295166,
      "epoch": 0.6121646421392991,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004968957750474806,
      "loss": 5.9536,
      "mean_token_accuracy": 0.15419106930494308,
      "num_tokens": 14198446.0,
      "step": 7005
    },
    {
      "entropy": 6.299173974990845,
      "epoch": 0.6126015904920038,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004968906145796126,
      "loss": 6.1685,
      "mean_token_accuracy": 0.13943691700696945,
      "num_tokens": 14207723.0,
      "step": 7010
    },
    {
      "entropy": 6.25934739112854,
      "epoch": 0.6130385388447086,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004968854498557662,
      "loss": 6.0521,
      "mean_token_accuracy": 0.1388497233390808,
      "num_tokens": 14217590.0,
      "step": 7015
    },
    {
      "entropy": 6.222770690917969,
      "epoch": 0.6134754871974133,
      "grad_norm": 3.125,
      "learning_rate": 0.0004968802808760407,
      "loss": 6.1184,
      "mean_token_accuracy": 0.1370713785290718,
      "num_tokens": 14226629.0,
      "step": 7020
    },
    {
      "entropy": 6.199009275436401,
      "epoch": 0.613912435550118,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004968751076405351,
      "loss": 6.1711,
      "mean_token_accuracy": 0.13404033258557319,
      "num_tokens": 14235782.0,
      "step": 7025
    },
    {
      "entropy": 6.162545156478882,
      "epoch": 0.6143493839028227,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004968699301493488,
      "loss": 6.043,
      "mean_token_accuracy": 0.14680892378091812,
      "num_tokens": 14246687.0,
      "step": 7030
    },
    {
      "entropy": 6.290499877929688,
      "epoch": 0.6147863322555274,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004968647484025808,
      "loss": 6.0835,
      "mean_token_accuracy": 0.1449754148721695,
      "num_tokens": 14255961.0,
      "step": 7035
    },
    {
      "entropy": 6.288722991943359,
      "epoch": 0.6152232806082321,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004968595624003307,
      "loss": 6.0929,
      "mean_token_accuracy": 0.1403207339346409,
      "num_tokens": 14266024.0,
      "step": 7040
    },
    {
      "entropy": 6.227539491653443,
      "epoch": 0.6156602289609369,
      "grad_norm": 4.6875,
      "learning_rate": 0.000496854372142698,
      "loss": 6.1157,
      "mean_token_accuracy": 0.1384843461215496,
      "num_tokens": 14276590.0,
      "step": 7045
    },
    {
      "entropy": 6.177028369903565,
      "epoch": 0.6160971773136416,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004968491776297821,
      "loss": 6.123,
      "mean_token_accuracy": 0.1399732105433941,
      "num_tokens": 14285992.0,
      "step": 7050
    },
    {
      "entropy": 6.218402957916259,
      "epoch": 0.6165341256663462,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004968439788616829,
      "loss": 6.0315,
      "mean_token_accuracy": 0.14302821904420854,
      "num_tokens": 14295843.0,
      "step": 7055
    },
    {
      "entropy": 6.2850103855133055,
      "epoch": 0.6169710740190509,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004968387758384999,
      "loss": 6.0785,
      "mean_token_accuracy": 0.14192734211683272,
      "num_tokens": 14305643.0,
      "step": 7060
    },
    {
      "entropy": 6.251075029373169,
      "epoch": 0.6174080223717556,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004968335685603328,
      "loss": 6.0443,
      "mean_token_accuracy": 0.14319950193166733,
      "num_tokens": 14315907.0,
      "step": 7065
    },
    {
      "entropy": 6.178477239608765,
      "epoch": 0.6178449707244603,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004968283570272817,
      "loss": 6.1085,
      "mean_token_accuracy": 0.13525563031435012,
      "num_tokens": 14325893.0,
      "step": 7070
    },
    {
      "entropy": 6.282196474075318,
      "epoch": 0.6182819190771651,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004968231412394464,
      "loss": 6.1274,
      "mean_token_accuracy": 0.1343388892710209,
      "num_tokens": 14337510.0,
      "step": 7075
    },
    {
      "entropy": 6.293983840942383,
      "epoch": 0.6187188674298698,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004968179211969272,
      "loss": 6.0476,
      "mean_token_accuracy": 0.14687205329537392,
      "num_tokens": 14346373.0,
      "step": 7080
    },
    {
      "entropy": 6.13792314529419,
      "epoch": 0.6191558157825745,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004968126968998239,
      "loss": 5.9217,
      "mean_token_accuracy": 0.16086165681481362,
      "num_tokens": 14356303.0,
      "step": 7085
    },
    {
      "entropy": 6.207563972473144,
      "epoch": 0.6195927641352792,
      "grad_norm": 7.0,
      "learning_rate": 0.0004968074683482369,
      "loss": 6.0758,
      "mean_token_accuracy": 0.14417827501893044,
      "num_tokens": 14366171.0,
      "step": 7090
    },
    {
      "entropy": 6.286942481994629,
      "epoch": 0.6200297124879839,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004968022355422663,
      "loss": 6.1578,
      "mean_token_accuracy": 0.1373531512916088,
      "num_tokens": 14376133.0,
      "step": 7095
    },
    {
      "entropy": 6.285527038574219,
      "epoch": 0.6204666608406886,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004967969984820128,
      "loss": 6.0821,
      "mean_token_accuracy": 0.13749169334769248,
      "num_tokens": 14385670.0,
      "step": 7100
    },
    {
      "entropy": 6.261067867279053,
      "epoch": 0.6209036091933934,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004967917571675765,
      "loss": 6.061,
      "mean_token_accuracy": 0.14238285794854164,
      "num_tokens": 14396002.0,
      "step": 7105
    },
    {
      "entropy": 6.194595718383789,
      "epoch": 0.6213405575460981,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004967865115990582,
      "loss": 6.1679,
      "mean_token_accuracy": 0.135838333517313,
      "num_tokens": 14406718.0,
      "step": 7110
    },
    {
      "entropy": 6.288463401794433,
      "epoch": 0.6217775058988028,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004967812617765583,
      "loss": 6.0948,
      "mean_token_accuracy": 0.13747066184878348,
      "num_tokens": 14417027.0,
      "step": 7115
    },
    {
      "entropy": 6.223663854598999,
      "epoch": 0.6222144542515075,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004967760077001775,
      "loss": 6.0515,
      "mean_token_accuracy": 0.14532688781619071,
      "num_tokens": 14426885.0,
      "step": 7120
    },
    {
      "entropy": 6.199916648864746,
      "epoch": 0.6226514026042121,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004967707493700169,
      "loss": 6.0422,
      "mean_token_accuracy": 0.14331899285316468,
      "num_tokens": 14437743.0,
      "step": 7125
    },
    {
      "entropy": 6.162434911727905,
      "epoch": 0.623088350956917,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004967654867861769,
      "loss": 6.0493,
      "mean_token_accuracy": 0.14364799186587335,
      "num_tokens": 14448808.0,
      "step": 7130
    },
    {
      "entropy": 6.2316086769104,
      "epoch": 0.6235252993096216,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004967602199487588,
      "loss": 6.0986,
      "mean_token_accuracy": 0.14064462780952453,
      "num_tokens": 14458560.0,
      "step": 7135
    },
    {
      "entropy": 6.288028764724731,
      "epoch": 0.6239622476623263,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004967549488578635,
      "loss": 6.1566,
      "mean_token_accuracy": 0.13784303665161132,
      "num_tokens": 14469375.0,
      "step": 7140
    },
    {
      "entropy": 6.292950677871704,
      "epoch": 0.624399196015031,
      "grad_norm": 1.78125,
      "learning_rate": 0.000496749673513592,
      "loss": 6.084,
      "mean_token_accuracy": 0.13916077464818954,
      "num_tokens": 14478444.0,
      "step": 7145
    },
    {
      "entropy": 6.194452714920044,
      "epoch": 0.6248361443677357,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004967443939160456,
      "loss": 5.9593,
      "mean_token_accuracy": 0.153594534099102,
      "num_tokens": 14487776.0,
      "step": 7150
    },
    {
      "entropy": 6.247198104858398,
      "epoch": 0.6252730927204404,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004967391100653255,
      "loss": 6.1037,
      "mean_token_accuracy": 0.14412544369697572,
      "num_tokens": 14497879.0,
      "step": 7155
    },
    {
      "entropy": 6.211316156387329,
      "epoch": 0.6257100410731452,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004967338219615331,
      "loss": 6.0354,
      "mean_token_accuracy": 0.14063015207648277,
      "num_tokens": 14507753.0,
      "step": 7160
    },
    {
      "entropy": 6.241280126571655,
      "epoch": 0.6261469894258499,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004967285296047699,
      "loss": 5.9936,
      "mean_token_accuracy": 0.14772187173366547,
      "num_tokens": 14517086.0,
      "step": 7165
    },
    {
      "entropy": 6.20732889175415,
      "epoch": 0.6265839377785546,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004967232329951372,
      "loss": 6.0801,
      "mean_token_accuracy": 0.1369892843067646,
      "num_tokens": 14527097.0,
      "step": 7170
    },
    {
      "entropy": 6.245126581192016,
      "epoch": 0.6270208861312593,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004967179321327367,
      "loss": 6.1008,
      "mean_token_accuracy": 0.14085085242986678,
      "num_tokens": 14537924.0,
      "step": 7175
    },
    {
      "entropy": 6.247621250152588,
      "epoch": 0.627457834483964,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004967126270176702,
      "loss": 6.1147,
      "mean_token_accuracy": 0.13855889439582825,
      "num_tokens": 14547542.0,
      "step": 7180
    },
    {
      "entropy": 6.145660495758056,
      "epoch": 0.6278947828366687,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004967073176500392,
      "loss": 6.04,
      "mean_token_accuracy": 0.1502846211194992,
      "num_tokens": 14557150.0,
      "step": 7185
    },
    {
      "entropy": 6.217853116989136,
      "epoch": 0.6283317311893735,
      "grad_norm": 5.75,
      "learning_rate": 0.0004967020040299458,
      "loss": 6.0505,
      "mean_token_accuracy": 0.14292264431715013,
      "num_tokens": 14567242.0,
      "step": 7190
    },
    {
      "entropy": 6.220586633682251,
      "epoch": 0.6287686795420782,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004966966861574917,
      "loss": 6.0302,
      "mean_token_accuracy": 0.14270174130797386,
      "num_tokens": 14576631.0,
      "step": 7195
    },
    {
      "entropy": 6.252045631408691,
      "epoch": 0.6292056278947828,
      "grad_norm": 3.203125,
      "learning_rate": 0.000496691364032779,
      "loss": 6.109,
      "mean_token_accuracy": 0.1352137267589569,
      "num_tokens": 14587731.0,
      "step": 7200
    },
    {
      "entropy": 6.341271495819091,
      "epoch": 0.6296425762474875,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004966860376559097,
      "loss": 6.1159,
      "mean_token_accuracy": 0.13765100985765458,
      "num_tokens": 14598136.0,
      "step": 7205
    },
    {
      "entropy": 6.183723735809326,
      "epoch": 0.6300795246001922,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004966807070269862,
      "loss": 6.076,
      "mean_token_accuracy": 0.1392249695956707,
      "num_tokens": 14607243.0,
      "step": 7210
    },
    {
      "entropy": 6.167708015441894,
      "epoch": 0.6305164729528969,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004966753721461105,
      "loss": 6.0973,
      "mean_token_accuracy": 0.1419666089117527,
      "num_tokens": 14617094.0,
      "step": 7215
    },
    {
      "entropy": 6.25998330116272,
      "epoch": 0.6309534213056017,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004966700330133849,
      "loss": 6.1021,
      "mean_token_accuracy": 0.1416136458516121,
      "num_tokens": 14626486.0,
      "step": 7220
    },
    {
      "entropy": 6.10648889541626,
      "epoch": 0.6313903696583064,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004966646896289121,
      "loss": 5.965,
      "mean_token_accuracy": 0.15503031462430955,
      "num_tokens": 14637170.0,
      "step": 7225
    },
    {
      "entropy": 6.209821176528931,
      "epoch": 0.6318273180110111,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004966593419927944,
      "loss": 6.0457,
      "mean_token_accuracy": 0.1434933751821518,
      "num_tokens": 14646396.0,
      "step": 7230
    },
    {
      "entropy": 6.252063035964966,
      "epoch": 0.6322642663637158,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004966539901051343,
      "loss": 6.095,
      "mean_token_accuracy": 0.1328526608645916,
      "num_tokens": 14656097.0,
      "step": 7235
    },
    {
      "entropy": 6.218412637710571,
      "epoch": 0.6327012147164205,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004966486339660345,
      "loss": 6.078,
      "mean_token_accuracy": 0.1426958732306957,
      "num_tokens": 14667124.0,
      "step": 7240
    },
    {
      "entropy": 6.180486488342285,
      "epoch": 0.6331381630691252,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004966432735755978,
      "loss": 6.1506,
      "mean_token_accuracy": 0.14178672879934312,
      "num_tokens": 14677427.0,
      "step": 7245
    },
    {
      "entropy": 6.19655032157898,
      "epoch": 0.63357511142183,
      "grad_norm": 1.9375,
      "learning_rate": 0.000496637908933927,
      "loss": 5.9975,
      "mean_token_accuracy": 0.1448795348405838,
      "num_tokens": 14686429.0,
      "step": 7250
    },
    {
      "entropy": 6.284691190719604,
      "epoch": 0.6340120597745347,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000496632540041125,
      "loss": 6.0608,
      "mean_token_accuracy": 0.14665302485227585,
      "num_tokens": 14696264.0,
      "step": 7255
    },
    {
      "entropy": 6.2796959400177,
      "epoch": 0.6344490081272394,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004966271668972946,
      "loss": 6.0781,
      "mean_token_accuracy": 0.14009574726223945,
      "num_tokens": 14705935.0,
      "step": 7260
    },
    {
      "entropy": 6.191766738891602,
      "epoch": 0.6348859564799441,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004966217895025391,
      "loss": 6.0473,
      "mean_token_accuracy": 0.14082057923078536,
      "num_tokens": 14715978.0,
      "step": 7265
    },
    {
      "entropy": 6.209609460830689,
      "epoch": 0.6353229048326487,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004966164078569616,
      "loss": 6.0643,
      "mean_token_accuracy": 0.13889703378081322,
      "num_tokens": 14726170.0,
      "step": 7270
    },
    {
      "entropy": 6.220421314239502,
      "epoch": 0.6357598531853534,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004966110219606652,
      "loss": 6.0417,
      "mean_token_accuracy": 0.14052743092179298,
      "num_tokens": 14735775.0,
      "step": 7275
    },
    {
      "entropy": 6.175203943252564,
      "epoch": 0.6361968015380582,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004966056318137532,
      "loss": 6.0409,
      "mean_token_accuracy": 0.1373811826109886,
      "num_tokens": 14746143.0,
      "step": 7280
    },
    {
      "entropy": 6.199851703643799,
      "epoch": 0.6366337498907629,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004966002374163293,
      "loss": 6.0865,
      "mean_token_accuracy": 0.14739727973937988,
      "num_tokens": 14756707.0,
      "step": 7285
    },
    {
      "entropy": 6.216392993927002,
      "epoch": 0.6370706982434676,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004965948387684964,
      "loss": 6.0948,
      "mean_token_accuracy": 0.14192008525133132,
      "num_tokens": 14765778.0,
      "step": 7290
    },
    {
      "entropy": 6.263886833190918,
      "epoch": 0.6375076465961723,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004965894358703587,
      "loss": 6.093,
      "mean_token_accuracy": 0.13505014330148696,
      "num_tokens": 14775264.0,
      "step": 7295
    },
    {
      "entropy": 6.247747802734375,
      "epoch": 0.637944594948877,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004965840287220193,
      "loss": 6.0689,
      "mean_token_accuracy": 0.1382345050573349,
      "num_tokens": 14785754.0,
      "step": 7300
    },
    {
      "entropy": 6.203045654296875,
      "epoch": 0.6383815433015817,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004965786173235821,
      "loss": 6.1191,
      "mean_token_accuracy": 0.13925366178154946,
      "num_tokens": 14795936.0,
      "step": 7305
    },
    {
      "entropy": 6.276138544082642,
      "epoch": 0.6388184916542865,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004965732016751511,
      "loss": 6.0951,
      "mean_token_accuracy": 0.14140693917870523,
      "num_tokens": 14805721.0,
      "step": 7310
    },
    {
      "entropy": 6.292239046096801,
      "epoch": 0.6392554400069912,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004965677817768299,
      "loss": 6.1287,
      "mean_token_accuracy": 0.13380056470632554,
      "num_tokens": 14816610.0,
      "step": 7315
    },
    {
      "entropy": 6.239180994033814,
      "epoch": 0.6396923883596959,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004965623576287225,
      "loss": 6.0956,
      "mean_token_accuracy": 0.1365938127040863,
      "num_tokens": 14826126.0,
      "step": 7320
    },
    {
      "entropy": 6.196110773086548,
      "epoch": 0.6401293367124006,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004965569292309331,
      "loss": 5.9961,
      "mean_token_accuracy": 0.14939529076218605,
      "num_tokens": 14835657.0,
      "step": 7325
    },
    {
      "entropy": 6.169877243041992,
      "epoch": 0.6405662850651053,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004965514965835655,
      "loss": 6.0636,
      "mean_token_accuracy": 0.1412982389330864,
      "num_tokens": 14845430.0,
      "step": 7330
    },
    {
      "entropy": 6.191928052902222,
      "epoch": 0.6410032334178101,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004965460596867243,
      "loss": 6.0251,
      "mean_token_accuracy": 0.1474901869893074,
      "num_tokens": 14855459.0,
      "step": 7335
    },
    {
      "entropy": 6.267540836334229,
      "epoch": 0.6414401817705148,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004965406185405135,
      "loss": 6.0863,
      "mean_token_accuracy": 0.14111222848296165,
      "num_tokens": 14865375.0,
      "step": 7340
    },
    {
      "entropy": 6.233385133743286,
      "epoch": 0.6418771301232195,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004965351731450376,
      "loss": 6.048,
      "mean_token_accuracy": 0.14531854838132857,
      "num_tokens": 14875211.0,
      "step": 7345
    },
    {
      "entropy": 6.214669418334961,
      "epoch": 0.6423140784759241,
      "grad_norm": 1.875,
      "learning_rate": 0.000496529723500401,
      "loss": 6.0973,
      "mean_token_accuracy": 0.13480584993958472,
      "num_tokens": 14886221.0,
      "step": 7350
    },
    {
      "entropy": 6.259183168411255,
      "epoch": 0.6427510268286288,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004965242696067081,
      "loss": 6.0566,
      "mean_token_accuracy": 0.14169811755418776,
      "num_tokens": 14896002.0,
      "step": 7355
    },
    {
      "entropy": 6.207713556289673,
      "epoch": 0.6431879751813335,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004965188114640637,
      "loss": 6.1089,
      "mean_token_accuracy": 0.1374245673418045,
      "num_tokens": 14906311.0,
      "step": 7360
    },
    {
      "entropy": 6.266039562225342,
      "epoch": 0.6436249235340383,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004965133490725725,
      "loss": 6.091,
      "mean_token_accuracy": 0.14037528187036513,
      "num_tokens": 14915890.0,
      "step": 7365
    },
    {
      "entropy": 6.244173765182495,
      "epoch": 0.644061871886743,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004965078824323391,
      "loss": 6.041,
      "mean_token_accuracy": 0.14121145233511925,
      "num_tokens": 14925891.0,
      "step": 7370
    },
    {
      "entropy": 6.1994599342346195,
      "epoch": 0.6444988202394477,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004965024115434685,
      "loss": 6.1013,
      "mean_token_accuracy": 0.13864176496863365,
      "num_tokens": 14937272.0,
      "step": 7375
    },
    {
      "entropy": 6.257472133636474,
      "epoch": 0.6449357685921524,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004964969364060656,
      "loss": 6.1105,
      "mean_token_accuracy": 0.13872194141149521,
      "num_tokens": 14948135.0,
      "step": 7380
    },
    {
      "entropy": 6.358830785751342,
      "epoch": 0.6453727169448571,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004964914570202353,
      "loss": 6.1556,
      "mean_token_accuracy": 0.13665004670619965,
      "num_tokens": 14959254.0,
      "step": 7385
    },
    {
      "entropy": 6.236004257202149,
      "epoch": 0.6458096652975618,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004964859733860829,
      "loss": 6.074,
      "mean_token_accuracy": 0.1428222142159939,
      "num_tokens": 14969313.0,
      "step": 7390
    },
    {
      "entropy": 6.188684606552124,
      "epoch": 0.6462466136502666,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004964804855037134,
      "loss": 6.072,
      "mean_token_accuracy": 0.14139073044061662,
      "num_tokens": 14978963.0,
      "step": 7395
    },
    {
      "entropy": 6.290366458892822,
      "epoch": 0.6466835620029713,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004964749933732322,
      "loss": 6.1085,
      "mean_token_accuracy": 0.14476230666041373,
      "num_tokens": 14989355.0,
      "step": 7400
    },
    {
      "entropy": 6.203551769256592,
      "epoch": 0.647120510355676,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004964694969947445,
      "loss": 6.0458,
      "mean_token_accuracy": 0.14362569898366928,
      "num_tokens": 14998995.0,
      "step": 7405
    },
    {
      "entropy": 6.224933671951294,
      "epoch": 0.6475574587083807,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004964639963683558,
      "loss": 6.0905,
      "mean_token_accuracy": 0.13989042565226556,
      "num_tokens": 15008762.0,
      "step": 7410
    },
    {
      "entropy": 6.311505746841431,
      "epoch": 0.6479944070610854,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004964584914941716,
      "loss": 6.0972,
      "mean_token_accuracy": 0.13879491463303567,
      "num_tokens": 15020062.0,
      "step": 7415
    },
    {
      "entropy": 6.3743222713470455,
      "epoch": 0.64843135541379,
      "grad_norm": 1.875,
      "learning_rate": 0.0004964529823722975,
      "loss": 6.2193,
      "mean_token_accuracy": 0.12883895188570021,
      "num_tokens": 15030440.0,
      "step": 7420
    },
    {
      "entropy": 6.287896394729614,
      "epoch": 0.6488683037664948,
      "grad_norm": 2.5,
      "learning_rate": 0.0004964474690028393,
      "loss": 6.0729,
      "mean_token_accuracy": 0.14202678576111794,
      "num_tokens": 15041709.0,
      "step": 7425
    },
    {
      "entropy": 6.2483141899108885,
      "epoch": 0.6493052521191995,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004964419513859025,
      "loss": 6.1909,
      "mean_token_accuracy": 0.13687760010361671,
      "num_tokens": 15052277.0,
      "step": 7430
    },
    {
      "entropy": 6.188128232955933,
      "epoch": 0.6497422004719042,
      "grad_norm": 2.25,
      "learning_rate": 0.000496436429521593,
      "loss": 6.0676,
      "mean_token_accuracy": 0.14018882662057877,
      "num_tokens": 15061870.0,
      "step": 7435
    },
    {
      "entropy": 6.123696947097779,
      "epoch": 0.6501791488246089,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004964309034100168,
      "loss": 5.944,
      "mean_token_accuracy": 0.1480805829167366,
      "num_tokens": 15071403.0,
      "step": 7440
    },
    {
      "entropy": 6.193117904663086,
      "epoch": 0.6506160971773136,
      "grad_norm": 2.25,
      "learning_rate": 0.0004964253730512799,
      "loss": 6.041,
      "mean_token_accuracy": 0.1450770802795887,
      "num_tokens": 15081651.0,
      "step": 7445
    },
    {
      "entropy": 6.257673120498657,
      "epoch": 0.6510530455300183,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004964198384454882,
      "loss": 6.1119,
      "mean_token_accuracy": 0.1403050921857357,
      "num_tokens": 15093403.0,
      "step": 7450
    },
    {
      "entropy": 6.285808038711548,
      "epoch": 0.6514899938827231,
      "grad_norm": 1.75,
      "learning_rate": 0.000496414299592748,
      "loss": 6.1717,
      "mean_token_accuracy": 0.13339342251420022,
      "num_tokens": 15103618.0,
      "step": 7455
    },
    {
      "entropy": 6.264203071594238,
      "epoch": 0.6519269422354278,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004964087564931656,
      "loss": 6.1001,
      "mean_token_accuracy": 0.1440037965774536,
      "num_tokens": 15113926.0,
      "step": 7460
    },
    {
      "entropy": 6.311422634124756,
      "epoch": 0.6523638905881325,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004964032091468472,
      "loss": 6.1483,
      "mean_token_accuracy": 0.13408324718475342,
      "num_tokens": 15125302.0,
      "step": 7465
    },
    {
      "entropy": 6.240755319595337,
      "epoch": 0.6528008389408372,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004963976575538991,
      "loss": 6.0381,
      "mean_token_accuracy": 0.14353784397244454,
      "num_tokens": 15135943.0,
      "step": 7470
    },
    {
      "entropy": 6.187025737762451,
      "epoch": 0.6532377872935419,
      "grad_norm": 2.25,
      "learning_rate": 0.000496392101714428,
      "loss": 6.0834,
      "mean_token_accuracy": 0.14651306569576264,
      "num_tokens": 15145001.0,
      "step": 7475
    },
    {
      "entropy": 6.224766111373901,
      "epoch": 0.6536747356462466,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004963865416285405,
      "loss": 5.9831,
      "mean_token_accuracy": 0.13959733992815018,
      "num_tokens": 15154567.0,
      "step": 7480
    },
    {
      "entropy": 6.292343664169311,
      "epoch": 0.6541116839989514,
      "grad_norm": 2.0,
      "learning_rate": 0.0004963809772963431,
      "loss": 6.1736,
      "mean_token_accuracy": 0.14008966982364654,
      "num_tokens": 15164311.0,
      "step": 7485
    },
    {
      "entropy": 6.2469398021698,
      "epoch": 0.6545486323516561,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004963754087179425,
      "loss": 6.0559,
      "mean_token_accuracy": 0.13879222944378852,
      "num_tokens": 15175217.0,
      "step": 7490
    },
    {
      "entropy": 6.2084550857543945,
      "epoch": 0.6549855807043607,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004963698358934455,
      "loss": 6.0604,
      "mean_token_accuracy": 0.14221766889095305,
      "num_tokens": 15184895.0,
      "step": 7495
    },
    {
      "entropy": 6.173346376419067,
      "epoch": 0.6554225290570654,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004963642588229591,
      "loss": 6.0367,
      "mean_token_accuracy": 0.14261841848492623,
      "num_tokens": 15194380.0,
      "step": 7500
    },
    {
      "entropy": 6.267757225036621,
      "epoch": 0.6558594774097701,
      "grad_norm": 6.3125,
      "learning_rate": 0.0004963586775065902,
      "loss": 6.0882,
      "mean_token_accuracy": 0.137705647200346,
      "num_tokens": 15204364.0,
      "step": 7505
    },
    {
      "entropy": 6.271279907226562,
      "epoch": 0.6562964257624748,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004963530919444459,
      "loss": 6.0421,
      "mean_token_accuracy": 0.1366719014942646,
      "num_tokens": 15214485.0,
      "step": 7510
    },
    {
      "entropy": 6.287254810333252,
      "epoch": 0.6567333741151796,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004963475021366333,
      "loss": 6.2012,
      "mean_token_accuracy": 0.12970216646790506,
      "num_tokens": 15224764.0,
      "step": 7515
    },
    {
      "entropy": 6.31780252456665,
      "epoch": 0.6571703224678843,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004963419080832597,
      "loss": 6.1879,
      "mean_token_accuracy": 0.13702881038188935,
      "num_tokens": 15235481.0,
      "step": 7520
    },
    {
      "entropy": 6.2787665843963625,
      "epoch": 0.657607270820589,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004963363097844323,
      "loss": 6.1296,
      "mean_token_accuracy": 0.13650887086987495,
      "num_tokens": 15245332.0,
      "step": 7525
    },
    {
      "entropy": 6.2709825992584225,
      "epoch": 0.6580442191732937,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004963307072402584,
      "loss": 6.1492,
      "mean_token_accuracy": 0.13837921172380446,
      "num_tokens": 15255562.0,
      "step": 7530
    },
    {
      "entropy": 6.2294000625610355,
      "epoch": 0.6584811675259984,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004963251004508456,
      "loss": 6.0622,
      "mean_token_accuracy": 0.1389525756239891,
      "num_tokens": 15265655.0,
      "step": 7535
    },
    {
      "entropy": 6.200943660736084,
      "epoch": 0.6589181158787032,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004963194894163015,
      "loss": 6.008,
      "mean_token_accuracy": 0.14637767225503923,
      "num_tokens": 15276311.0,
      "step": 7540
    },
    {
      "entropy": 6.233555889129638,
      "epoch": 0.6593550642314079,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004963138741367334,
      "loss": 6.1558,
      "mean_token_accuracy": 0.13676377534866332,
      "num_tokens": 15287142.0,
      "step": 7545
    },
    {
      "entropy": 6.2695239067077635,
      "epoch": 0.6597920125841126,
      "grad_norm": 3.875,
      "learning_rate": 0.0004963082546122495,
      "loss": 6.1526,
      "mean_token_accuracy": 0.135238941013813,
      "num_tokens": 15297484.0,
      "step": 7550
    },
    {
      "entropy": 6.293298292160034,
      "epoch": 0.6602289609368173,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004963026308429571,
      "loss": 6.0999,
      "mean_token_accuracy": 0.1388798713684082,
      "num_tokens": 15307468.0,
      "step": 7555
    },
    {
      "entropy": 6.204461097717285,
      "epoch": 0.660665909289522,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004962970028289644,
      "loss": 6.0408,
      "mean_token_accuracy": 0.14689301177859307,
      "num_tokens": 15318487.0,
      "step": 7560
    },
    {
      "entropy": 6.193296766281128,
      "epoch": 0.6611028576422266,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004962913705703791,
      "loss": 6.0832,
      "mean_token_accuracy": 0.1438783273100853,
      "num_tokens": 15329193.0,
      "step": 7565
    },
    {
      "entropy": 6.2236976146698,
      "epoch": 0.6615398059949315,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004962857340673095,
      "loss": 6.0539,
      "mean_token_accuracy": 0.1355200909078121,
      "num_tokens": 15338886.0,
      "step": 7570
    },
    {
      "entropy": 6.209378719329834,
      "epoch": 0.6619767543476361,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004962800933198635,
      "loss": 5.9951,
      "mean_token_accuracy": 0.15128109306097032,
      "num_tokens": 15348629.0,
      "step": 7575
    },
    {
      "entropy": 6.210313034057617,
      "epoch": 0.6624137027003408,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004962744483281495,
      "loss": 6.0883,
      "mean_token_accuracy": 0.14175621643662453,
      "num_tokens": 15358004.0,
      "step": 7580
    },
    {
      "entropy": 6.222506761550903,
      "epoch": 0.6628506510530455,
      "grad_norm": 2.625,
      "learning_rate": 0.0004962687990922754,
      "loss": 6.1186,
      "mean_token_accuracy": 0.1353178545832634,
      "num_tokens": 15368018.0,
      "step": 7585
    },
    {
      "entropy": 6.31260313987732,
      "epoch": 0.6632875994057502,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00049626314561235,
      "loss": 6.0669,
      "mean_token_accuracy": 0.13854215443134307,
      "num_tokens": 15377443.0,
      "step": 7590
    },
    {
      "entropy": 6.222037744522095,
      "epoch": 0.6637245477584549,
      "grad_norm": 3.125,
      "learning_rate": 0.0004962574878884814,
      "loss": 6.0873,
      "mean_token_accuracy": 0.14046914577484132,
      "num_tokens": 15388278.0,
      "step": 7595
    },
    {
      "entropy": 6.175523900985718,
      "epoch": 0.6641614961111597,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004962518259207782,
      "loss": 6.0909,
      "mean_token_accuracy": 0.14351003468036652,
      "num_tokens": 15398619.0,
      "step": 7600
    },
    {
      "entropy": 6.205916500091552,
      "epoch": 0.6645984444638644,
      "grad_norm": 2.125,
      "learning_rate": 0.0004962461597093491,
      "loss": 5.9868,
      "mean_token_accuracy": 0.1472811594605446,
      "num_tokens": 15408635.0,
      "step": 7605
    },
    {
      "entropy": 6.270863628387451,
      "epoch": 0.6650353928165691,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004962404892543028,
      "loss": 6.0929,
      "mean_token_accuracy": 0.13835323452949524,
      "num_tokens": 15419113.0,
      "step": 7610
    },
    {
      "entropy": 6.3479838371276855,
      "epoch": 0.6654723411692738,
      "grad_norm": 2.203125,
      "learning_rate": 0.000496234814555748,
      "loss": 6.092,
      "mean_token_accuracy": 0.13462133705615997,
      "num_tokens": 15428959.0,
      "step": 7615
    },
    {
      "entropy": 6.219027853012085,
      "epoch": 0.6659092895219785,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004962291356137934,
      "loss": 6.1336,
      "mean_token_accuracy": 0.13156894743442535,
      "num_tokens": 15439116.0,
      "step": 7620
    },
    {
      "entropy": 6.240478801727295,
      "epoch": 0.6663462378746832,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004962234524285481,
      "loss": 6.1185,
      "mean_token_accuracy": 0.14176129549741745,
      "num_tokens": 15449715.0,
      "step": 7625
    },
    {
      "entropy": 6.279317903518677,
      "epoch": 0.666783186227388,
      "grad_norm": 4.625,
      "learning_rate": 0.0004962177650001212,
      "loss": 5.926,
      "mean_token_accuracy": 0.15156869888305663,
      "num_tokens": 15459170.0,
      "step": 7630
    },
    {
      "entropy": 6.206975221633911,
      "epoch": 0.6672201345800927,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004962120733286215,
      "loss": 6.0362,
      "mean_token_accuracy": 0.1426810458302498,
      "num_tokens": 15469497.0,
      "step": 7635
    },
    {
      "entropy": 6.160815477371216,
      "epoch": 0.6676570829327974,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004962063774141583,
      "loss": 6.156,
      "mean_token_accuracy": 0.13761526569724084,
      "num_tokens": 15481298.0,
      "step": 7640
    },
    {
      "entropy": 6.1859899997711185,
      "epoch": 0.668094031285502,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004962006772568409,
      "loss": 6.1148,
      "mean_token_accuracy": 0.1422424003481865,
      "num_tokens": 15491696.0,
      "step": 7645
    },
    {
      "entropy": 6.275316429138184,
      "epoch": 0.6685309796382067,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004961949728567787,
      "loss": 6.1222,
      "mean_token_accuracy": 0.14227507188916205,
      "num_tokens": 15502412.0,
      "step": 7650
    },
    {
      "entropy": 6.297860479354858,
      "epoch": 0.6689679279909114,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004961892642140809,
      "loss": 6.1011,
      "mean_token_accuracy": 0.13902661204338074,
      "num_tokens": 15513436.0,
      "step": 7655
    },
    {
      "entropy": 6.19545259475708,
      "epoch": 0.6694048763436162,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004961835513288572,
      "loss": 6.0945,
      "mean_token_accuracy": 0.14622755721211433,
      "num_tokens": 15524564.0,
      "step": 7660
    },
    {
      "entropy": 6.2803600311279295,
      "epoch": 0.6698418246963209,
      "grad_norm": 3.359375,
      "learning_rate": 0.000496177834201217,
      "loss": 6.0663,
      "mean_token_accuracy": 0.13727382943034172,
      "num_tokens": 15534539.0,
      "step": 7665
    },
    {
      "entropy": 6.324079895019532,
      "epoch": 0.6702787730490256,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004961721128312702,
      "loss": 6.1479,
      "mean_token_accuracy": 0.1315089724957943,
      "num_tokens": 15544571.0,
      "step": 7670
    },
    {
      "entropy": 6.228364181518555,
      "epoch": 0.6707157214017303,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004961663872191263,
      "loss": 6.0835,
      "mean_token_accuracy": 0.1470337465405464,
      "num_tokens": 15555066.0,
      "step": 7675
    },
    {
      "entropy": 6.252587223052979,
      "epoch": 0.671152669754435,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004961606573648951,
      "loss": 6.116,
      "mean_token_accuracy": 0.13753531277179717,
      "num_tokens": 15565128.0,
      "step": 7680
    },
    {
      "entropy": 6.247122573852539,
      "epoch": 0.6715896181071397,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004961549232686867,
      "loss": 6.1793,
      "mean_token_accuracy": 0.1353545904159546,
      "num_tokens": 15576394.0,
      "step": 7685
    },
    {
      "entropy": 6.27998194694519,
      "epoch": 0.6720265664598445,
      "grad_norm": 2.90625,
      "learning_rate": 0.000496149184930611,
      "loss": 6.0889,
      "mean_token_accuracy": 0.1401011899113655,
      "num_tokens": 15585806.0,
      "step": 7690
    },
    {
      "entropy": 6.261417484283447,
      "epoch": 0.6724635148125492,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004961434423507779,
      "loss": 6.0754,
      "mean_token_accuracy": 0.14093467295169831,
      "num_tokens": 15596471.0,
      "step": 7695
    },
    {
      "entropy": 6.231219911575318,
      "epoch": 0.6729004631652539,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004961376955292979,
      "loss": 5.9898,
      "mean_token_accuracy": 0.15060002207756043,
      "num_tokens": 15605720.0,
      "step": 7700
    },
    {
      "entropy": 6.23662543296814,
      "epoch": 0.6733374115179586,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004961319444662809,
      "loss": 6.0816,
      "mean_token_accuracy": 0.14469746202230455,
      "num_tokens": 15616167.0,
      "step": 7705
    },
    {
      "entropy": 6.315139818191528,
      "epoch": 0.6737743598706633,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004961261891618374,
      "loss": 6.1516,
      "mean_token_accuracy": 0.13433588072657585,
      "num_tokens": 15627033.0,
      "step": 7710
    },
    {
      "entropy": 6.232497930526733,
      "epoch": 0.6742113082233679,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004961204296160777,
      "loss": 6.1109,
      "mean_token_accuracy": 0.13909128829836845,
      "num_tokens": 15637460.0,
      "step": 7715
    },
    {
      "entropy": 6.21066632270813,
      "epoch": 0.6746482565760727,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004961146658291123,
      "loss": 6.0259,
      "mean_token_accuracy": 0.14700669944286346,
      "num_tokens": 15646027.0,
      "step": 7720
    },
    {
      "entropy": 6.202454948425293,
      "epoch": 0.6750852049287774,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004961088978010518,
      "loss": 6.0718,
      "mean_token_accuracy": 0.1390362374484539,
      "num_tokens": 15655241.0,
      "step": 7725
    },
    {
      "entropy": 6.1901716709136965,
      "epoch": 0.6755221532814821,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004961031255320067,
      "loss": 6.0247,
      "mean_token_accuracy": 0.13531052097678184,
      "num_tokens": 15666213.0,
      "step": 7730
    },
    {
      "entropy": 6.2788177013397215,
      "epoch": 0.6759591016341868,
      "grad_norm": 2.0,
      "learning_rate": 0.0004960973490220878,
      "loss": 6.0702,
      "mean_token_accuracy": 0.13685087263584136,
      "num_tokens": 15676985.0,
      "step": 7735
    },
    {
      "entropy": 6.3100268840789795,
      "epoch": 0.6763960499868915,
      "grad_norm": 2.625,
      "learning_rate": 0.0004960915682714059,
      "loss": 6.1415,
      "mean_token_accuracy": 0.13514363691210746,
      "num_tokens": 15687519.0,
      "step": 7740
    },
    {
      "entropy": 6.270909023284912,
      "epoch": 0.6768329983395963,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004960857832800719,
      "loss": 6.1146,
      "mean_token_accuracy": 0.1358247369527817,
      "num_tokens": 15697401.0,
      "step": 7745
    },
    {
      "entropy": 6.2234748840332035,
      "epoch": 0.677269946692301,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004960799940481968,
      "loss": 6.1316,
      "mean_token_accuracy": 0.13038893416523933,
      "num_tokens": 15708159.0,
      "step": 7750
    },
    {
      "entropy": 6.244372081756592,
      "epoch": 0.6777068950450057,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004960742005758914,
      "loss": 6.0919,
      "mean_token_accuracy": 0.13842951506376266,
      "num_tokens": 15717749.0,
      "step": 7755
    },
    {
      "entropy": 6.1883094787597654,
      "epoch": 0.6781438433977104,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004960684028632671,
      "loss": 6.0478,
      "mean_token_accuracy": 0.14212872162461282,
      "num_tokens": 15726868.0,
      "step": 7760
    },
    {
      "entropy": 6.226647138595581,
      "epoch": 0.6785807917504151,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004960626009104349,
      "loss": 6.0395,
      "mean_token_accuracy": 0.14510313421487808,
      "num_tokens": 15736038.0,
      "step": 7765
    },
    {
      "entropy": 6.2614130020141605,
      "epoch": 0.6790177401031198,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004960567947175064,
      "loss": 6.0844,
      "mean_token_accuracy": 0.1364120401442051,
      "num_tokens": 15746676.0,
      "step": 7770
    },
    {
      "entropy": 6.291885280609131,
      "epoch": 0.6794546884558246,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004960509842845926,
      "loss": 6.0756,
      "mean_token_accuracy": 0.13798356875777246,
      "num_tokens": 15756064.0,
      "step": 7775
    },
    {
      "entropy": 6.259097528457642,
      "epoch": 0.6798916368085293,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004960451696118049,
      "loss": 6.0776,
      "mean_token_accuracy": 0.14294374734163284,
      "num_tokens": 15767414.0,
      "step": 7780
    },
    {
      "entropy": 6.291348648071289,
      "epoch": 0.680328585161234,
      "grad_norm": 2.0,
      "learning_rate": 0.0004960393506992553,
      "loss": 6.0673,
      "mean_token_accuracy": 0.14387979209423066,
      "num_tokens": 15778276.0,
      "step": 7785
    },
    {
      "entropy": 6.193063640594483,
      "epoch": 0.6807655335139386,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004960335275470551,
      "loss": 5.9891,
      "mean_token_accuracy": 0.14073141291737556,
      "num_tokens": 15788365.0,
      "step": 7790
    },
    {
      "entropy": 6.228651285171509,
      "epoch": 0.6812024818666433,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004960277001553159,
      "loss": 6.0115,
      "mean_token_accuracy": 0.14451875910162926,
      "num_tokens": 15799021.0,
      "step": 7795
    },
    {
      "entropy": 6.253414106369019,
      "epoch": 0.681639430219348,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004960218685241498,
      "loss": 6.0968,
      "mean_token_accuracy": 0.13949331343173982,
      "num_tokens": 15808779.0,
      "step": 7800
    },
    {
      "entropy": 6.259351921081543,
      "epoch": 0.6820763785720528,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004960160326536681,
      "loss": 6.0875,
      "mean_token_accuracy": 0.14508993923664093,
      "num_tokens": 15818515.0,
      "step": 7805
    },
    {
      "entropy": 6.226183128356934,
      "epoch": 0.6825133269247575,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004960101925439834,
      "loss": 6.0612,
      "mean_token_accuracy": 0.13916484117507935,
      "num_tokens": 15828843.0,
      "step": 7810
    },
    {
      "entropy": 6.291645860671997,
      "epoch": 0.6829502752774622,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004960043481952072,
      "loss": 6.0883,
      "mean_token_accuracy": 0.13636393025517463,
      "num_tokens": 15839323.0,
      "step": 7815
    },
    {
      "entropy": 6.273739051818848,
      "epoch": 0.6833872236301669,
      "grad_norm": 2.1875,
      "learning_rate": 0.000495998499607452,
      "loss": 6.1042,
      "mean_token_accuracy": 0.1382362000644207,
      "num_tokens": 15848399.0,
      "step": 7820
    },
    {
      "entropy": 6.231923580169678,
      "epoch": 0.6838241719828716,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004959926467808296,
      "loss": 5.9833,
      "mean_token_accuracy": 0.14106935560703276,
      "num_tokens": 15858137.0,
      "step": 7825
    },
    {
      "entropy": 6.147455167770386,
      "epoch": 0.6842611203355763,
      "grad_norm": 1.75,
      "learning_rate": 0.0004959867897154526,
      "loss": 6.0652,
      "mean_token_accuracy": 0.13927642107009888,
      "num_tokens": 15867643.0,
      "step": 7830
    },
    {
      "entropy": 6.20361294746399,
      "epoch": 0.6846980686882811,
      "grad_norm": 2.453125,
      "learning_rate": 0.000495980928411433,
      "loss": 5.9594,
      "mean_token_accuracy": 0.14536722749471664,
      "num_tokens": 15877523.0,
      "step": 7835
    },
    {
      "entropy": 6.281510686874389,
      "epoch": 0.6851350170409858,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004959750628688834,
      "loss": 6.1689,
      "mean_token_accuracy": 0.13679477348923683,
      "num_tokens": 15886946.0,
      "step": 7840
    },
    {
      "entropy": 6.176962566375733,
      "epoch": 0.6855719653936905,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004959691930879164,
      "loss": 5.9974,
      "mean_token_accuracy": 0.14489075690507888,
      "num_tokens": 15896646.0,
      "step": 7845
    },
    {
      "entropy": 6.213808679580689,
      "epoch": 0.6860089137463952,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004959633190686445,
      "loss": 6.0565,
      "mean_token_accuracy": 0.14461535140872,
      "num_tokens": 15907204.0,
      "step": 7850
    },
    {
      "entropy": 6.218833875656128,
      "epoch": 0.6864458620990999,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004959574408111804,
      "loss": 6.0808,
      "mean_token_accuracy": 0.14740002974867822,
      "num_tokens": 15917634.0,
      "step": 7855
    },
    {
      "entropy": 6.2356006622314455,
      "epoch": 0.6868828104518045,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004959515583156367,
      "loss": 6.0524,
      "mean_token_accuracy": 0.14325984716415405,
      "num_tokens": 15926584.0,
      "step": 7860
    },
    {
      "entropy": 6.226416015625,
      "epoch": 0.6873197588045094,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004959456715821264,
      "loss": 6.0337,
      "mean_token_accuracy": 0.14065923020243645,
      "num_tokens": 15936636.0,
      "step": 7865
    },
    {
      "entropy": 6.235525274276734,
      "epoch": 0.687756707157214,
      "grad_norm": 2.25,
      "learning_rate": 0.0004959397806107624,
      "loss": 6.0298,
      "mean_token_accuracy": 0.14805042818188668,
      "num_tokens": 15946113.0,
      "step": 7870
    },
    {
      "entropy": 6.264631223678589,
      "epoch": 0.6881936555099187,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004959338854016574,
      "loss": 6.0688,
      "mean_token_accuracy": 0.13688136637210846,
      "num_tokens": 15956630.0,
      "step": 7875
    },
    {
      "entropy": 6.21695032119751,
      "epoch": 0.6886306038626234,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004959279859549249,
      "loss": 6.0031,
      "mean_token_accuracy": 0.14520470499992372,
      "num_tokens": 15967129.0,
      "step": 7880
    },
    {
      "entropy": 6.236218690872192,
      "epoch": 0.6890675522153281,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004959220822706779,
      "loss": 6.0094,
      "mean_token_accuracy": 0.1407023474574089,
      "num_tokens": 15976663.0,
      "step": 7885
    },
    {
      "entropy": 6.247277879714966,
      "epoch": 0.6895045005680328,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004959161743490295,
      "loss": 6.0194,
      "mean_token_accuracy": 0.1417624481022358,
      "num_tokens": 15986666.0,
      "step": 7890
    },
    {
      "entropy": 6.123452425003052,
      "epoch": 0.6899414489207376,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004959102621900931,
      "loss": 5.9587,
      "mean_token_accuracy": 0.1560263492166996,
      "num_tokens": 15996233.0,
      "step": 7895
    },
    {
      "entropy": 6.247806692123413,
      "epoch": 0.6903783972734423,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004959043457939822,
      "loss": 6.1629,
      "mean_token_accuracy": 0.13539744317531585,
      "num_tokens": 16008188.0,
      "step": 7900
    },
    {
      "entropy": 6.2406538963317875,
      "epoch": 0.690815345626147,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004958984251608101,
      "loss": 6.0559,
      "mean_token_accuracy": 0.1421897292137146,
      "num_tokens": 16018494.0,
      "step": 7905
    },
    {
      "entropy": 6.234479522705078,
      "epoch": 0.6912522939788517,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004958925002906905,
      "loss": 6.0358,
      "mean_token_accuracy": 0.15123399272561072,
      "num_tokens": 16028531.0,
      "step": 7910
    },
    {
      "entropy": 6.252134323120117,
      "epoch": 0.6916892423315564,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000495886571183737,
      "loss": 6.0394,
      "mean_token_accuracy": 0.1430049479007721,
      "num_tokens": 16039050.0,
      "step": 7915
    },
    {
      "entropy": 6.2517712116241455,
      "epoch": 0.6921261906842611,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004958806378400634,
      "loss": 6.1397,
      "mean_token_accuracy": 0.1348439186811447,
      "num_tokens": 16050751.0,
      "step": 7920
    },
    {
      "entropy": 6.271159839630127,
      "epoch": 0.6925631390369659,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004958747002597834,
      "loss": 6.0794,
      "mean_token_accuracy": 0.13974782675504685,
      "num_tokens": 16060758.0,
      "step": 7925
    },
    {
      "entropy": 6.229480361938476,
      "epoch": 0.6930000873896706,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004958687584430109,
      "loss": 6.0786,
      "mean_token_accuracy": 0.1384492725133896,
      "num_tokens": 16070852.0,
      "step": 7930
    },
    {
      "entropy": 6.184289503097534,
      "epoch": 0.6934370357423753,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004958628123898597,
      "loss": 6.002,
      "mean_token_accuracy": 0.13939088210463524,
      "num_tokens": 16080081.0,
      "step": 7935
    },
    {
      "entropy": 6.234544944763184,
      "epoch": 0.6938739840950799,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004958568621004442,
      "loss": 6.0387,
      "mean_token_accuracy": 0.14234753847122192,
      "num_tokens": 16090483.0,
      "step": 7940
    },
    {
      "entropy": 6.2850720405578615,
      "epoch": 0.6943109324477846,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004958509075748785,
      "loss": 6.1128,
      "mean_token_accuracy": 0.13589274138212204,
      "num_tokens": 16101136.0,
      "step": 7945
    },
    {
      "entropy": 6.244620084762573,
      "epoch": 0.6947478808004894,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004958449488132764,
      "loss": 6.0454,
      "mean_token_accuracy": 0.1499411940574646,
      "num_tokens": 16110124.0,
      "step": 7950
    },
    {
      "entropy": 6.179692363739013,
      "epoch": 0.6951848291531941,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004958389858157526,
      "loss": 5.9704,
      "mean_token_accuracy": 0.14949270188808442,
      "num_tokens": 16119581.0,
      "step": 7955
    },
    {
      "entropy": 6.2103077411651615,
      "epoch": 0.6956217775058988,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004958330185824213,
      "loss": 6.0651,
      "mean_token_accuracy": 0.1400443397462368,
      "num_tokens": 16130119.0,
      "step": 7960
    },
    {
      "entropy": 6.248662233352661,
      "epoch": 0.6960587258586035,
      "grad_norm": 5.0625,
      "learning_rate": 0.000495827047113397,
      "loss": 6.0809,
      "mean_token_accuracy": 0.14603182747960092,
      "num_tokens": 16140702.0,
      "step": 7965
    },
    {
      "entropy": 6.2235160827636715,
      "epoch": 0.6964956742113082,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004958210714087942,
      "loss": 6.0596,
      "mean_token_accuracy": 0.14232370778918266,
      "num_tokens": 16151107.0,
      "step": 7970
    },
    {
      "entropy": 6.261234617233276,
      "epoch": 0.6969326225640129,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004958150914687276,
      "loss": 6.0158,
      "mean_token_accuracy": 0.14557732939720153,
      "num_tokens": 16160779.0,
      "step": 7975
    },
    {
      "entropy": 6.222548103332519,
      "epoch": 0.6973695709167177,
      "grad_norm": 2.25,
      "learning_rate": 0.0004958091072933118,
      "loss": 5.9529,
      "mean_token_accuracy": 0.1484435260295868,
      "num_tokens": 16169624.0,
      "step": 7980
    },
    {
      "entropy": 6.0988383293151855,
      "epoch": 0.6978065192694224,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004958031188826616,
      "loss": 5.9475,
      "mean_token_accuracy": 0.15440112948417664,
      "num_tokens": 16179192.0,
      "step": 7985
    },
    {
      "entropy": 6.12196102142334,
      "epoch": 0.6982434676221271,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004957971262368919,
      "loss": 6.0292,
      "mean_token_accuracy": 0.14307815954089165,
      "num_tokens": 16188847.0,
      "step": 7990
    },
    {
      "entropy": 6.280730962753296,
      "epoch": 0.6986804159748318,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004957911293561176,
      "loss": 6.1556,
      "mean_token_accuracy": 0.1361527405679226,
      "num_tokens": 16199480.0,
      "step": 7995
    },
    {
      "entropy": 6.313855600357056,
      "epoch": 0.6991173643275365,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004957851282404538,
      "loss": 6.0374,
      "mean_token_accuracy": 0.14604747369885446,
      "num_tokens": 16208818.0,
      "step": 8000
    },
    {
      "entropy": 6.254707622528076,
      "epoch": 0.6995543126802412,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004957791228900154,
      "loss": 6.0069,
      "mean_token_accuracy": 0.14893947616219522,
      "num_tokens": 16219298.0,
      "step": 8005
    },
    {
      "entropy": 6.158475303649903,
      "epoch": 0.699991261032946,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004957731133049179,
      "loss": 6.0139,
      "mean_token_accuracy": 0.1494004338979721,
      "num_tokens": 16228170.0,
      "step": 8010
    },
    {
      "entropy": 6.210049057006836,
      "epoch": 0.7004282093856506,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004957670994852764,
      "loss": 6.0901,
      "mean_token_accuracy": 0.13958207219839097,
      "num_tokens": 16237552.0,
      "step": 8015
    },
    {
      "entropy": 6.274960088729858,
      "epoch": 0.7008651577383553,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004957610814312062,
      "loss": 6.1267,
      "mean_token_accuracy": 0.13958656713366507,
      "num_tokens": 16248157.0,
      "step": 8020
    },
    {
      "entropy": 6.231983327865601,
      "epoch": 0.70130210609106,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004957550591428229,
      "loss": 6.0187,
      "mean_token_accuracy": 0.14684519171714783,
      "num_tokens": 16258266.0,
      "step": 8025
    },
    {
      "entropy": 6.173481273651123,
      "epoch": 0.7017390544437647,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004957490326202418,
      "loss": 6.1012,
      "mean_token_accuracy": 0.14092958196997643,
      "num_tokens": 16268769.0,
      "step": 8030
    },
    {
      "entropy": 6.266806983947754,
      "epoch": 0.7021760027964694,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004957430018635785,
      "loss": 6.0834,
      "mean_token_accuracy": 0.14316139072179795,
      "num_tokens": 16279688.0,
      "step": 8035
    },
    {
      "entropy": 6.2848986148834225,
      "epoch": 0.7026129511491742,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004957369668729488,
      "loss": 6.0307,
      "mean_token_accuracy": 0.14487592950463296,
      "num_tokens": 16289441.0,
      "step": 8040
    },
    {
      "entropy": 6.2462770462036135,
      "epoch": 0.7030498995018789,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004957309276484685,
      "loss": 6.0428,
      "mean_token_accuracy": 0.14013879671692847,
      "num_tokens": 16299802.0,
      "step": 8045
    },
    {
      "entropy": 6.190154266357422,
      "epoch": 0.7034868478545836,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004957248841902533,
      "loss": 5.9858,
      "mean_token_accuracy": 0.1485394701361656,
      "num_tokens": 16309577.0,
      "step": 8050
    },
    {
      "entropy": 6.148804140090943,
      "epoch": 0.7039237962072883,
      "grad_norm": 7.90625,
      "learning_rate": 0.0004957188364984191,
      "loss": 5.9687,
      "mean_token_accuracy": 0.14889626950025558,
      "num_tokens": 16319205.0,
      "step": 8055
    },
    {
      "entropy": 6.155892086029053,
      "epoch": 0.704360744559993,
      "grad_norm": 2.265625,
      "learning_rate": 0.000495712784573082,
      "loss": 6.0385,
      "mean_token_accuracy": 0.14606433138251304,
      "num_tokens": 16329650.0,
      "step": 8060
    },
    {
      "entropy": 6.1896062850952145,
      "epoch": 0.7047976929126977,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004957067284143581,
      "loss": 6.0358,
      "mean_token_accuracy": 0.14281625300645828,
      "num_tokens": 16339242.0,
      "step": 8065
    },
    {
      "entropy": 6.206028747558594,
      "epoch": 0.7052346412654025,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004957006680223634,
      "loss": 5.9999,
      "mean_token_accuracy": 0.14916329383850097,
      "num_tokens": 16348441.0,
      "step": 8070
    },
    {
      "entropy": 6.207070684432983,
      "epoch": 0.7056715896181072,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004956946033972143,
      "loss": 6.0092,
      "mean_token_accuracy": 0.14172426983714104,
      "num_tokens": 16357425.0,
      "step": 8075
    },
    {
      "entropy": 6.220908689498901,
      "epoch": 0.7061085379708119,
      "grad_norm": 1.8359375,
      "learning_rate": 0.000495688534539027,
      "loss": 6.0056,
      "mean_token_accuracy": 0.14150218665599823,
      "num_tokens": 16366180.0,
      "step": 8080
    },
    {
      "entropy": 6.196384572982788,
      "epoch": 0.7065454863235165,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004956824614479181,
      "loss": 6.0748,
      "mean_token_accuracy": 0.1426027871668339,
      "num_tokens": 16375714.0,
      "step": 8085
    },
    {
      "entropy": 6.260721111297608,
      "epoch": 0.7069824346762212,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004956763841240038,
      "loss": 6.0586,
      "mean_token_accuracy": 0.1394146427512169,
      "num_tokens": 16386051.0,
      "step": 8090
    },
    {
      "entropy": 6.267225837707519,
      "epoch": 0.7074193830289259,
      "grad_norm": 2.546875,
      "learning_rate": 0.000495670302567401,
      "loss": 6.0939,
      "mean_token_accuracy": 0.13717012405395507,
      "num_tokens": 16396616.0,
      "step": 8095
    },
    {
      "entropy": 6.190759563446045,
      "epoch": 0.7078563313816307,
      "grad_norm": 2.5,
      "learning_rate": 0.000495664216778226,
      "loss": 6.0039,
      "mean_token_accuracy": 0.15091822892427445,
      "num_tokens": 16406301.0,
      "step": 8100
    },
    {
      "entropy": 6.247671604156494,
      "epoch": 0.7082932797343354,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004956581267565957,
      "loss": 6.0896,
      "mean_token_accuracy": 0.136067433655262,
      "num_tokens": 16417107.0,
      "step": 8105
    },
    {
      "entropy": 6.2221972942352295,
      "epoch": 0.7087302280870401,
      "grad_norm": 2.328125,
      "learning_rate": 0.000495652032502627,
      "loss": 6.0103,
      "mean_token_accuracy": 0.1416815422475338,
      "num_tokens": 16427292.0,
      "step": 8110
    },
    {
      "entropy": 6.1572459697723385,
      "epoch": 0.7091671764397448,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004956459340164366,
      "loss": 6.0285,
      "mean_token_accuracy": 0.14646508246660234,
      "num_tokens": 16438191.0,
      "step": 8115
    },
    {
      "entropy": 6.198068523406983,
      "epoch": 0.7096041247924495,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004956398312981417,
      "loss": 6.0083,
      "mean_token_accuracy": 0.14299816489219666,
      "num_tokens": 16448237.0,
      "step": 8120
    },
    {
      "entropy": 6.161847925186157,
      "epoch": 0.7100410731451542,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004956337243478592,
      "loss": 6.1038,
      "mean_token_accuracy": 0.14033075869083406,
      "num_tokens": 16458947.0,
      "step": 8125
    },
    {
      "entropy": 6.261043071746826,
      "epoch": 0.710478021497859,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004956276131657061,
      "loss": 5.9975,
      "mean_token_accuracy": 0.1499764770269394,
      "num_tokens": 16467464.0,
      "step": 8130
    },
    {
      "entropy": 6.234200191497803,
      "epoch": 0.7109149698505637,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004956214977518,
      "loss": 6.0217,
      "mean_token_accuracy": 0.15601760447025298,
      "num_tokens": 16477747.0,
      "step": 8135
    },
    {
      "entropy": 6.255400466918945,
      "epoch": 0.7113519182032684,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004956153781062578,
      "loss": 6.1566,
      "mean_token_accuracy": 0.1387567326426506,
      "num_tokens": 16488111.0,
      "step": 8140
    },
    {
      "entropy": 6.221999979019165,
      "epoch": 0.7117888665559731,
      "grad_norm": 1.875,
      "learning_rate": 0.0004956092542291971,
      "loss": 6.025,
      "mean_token_accuracy": 0.14879337698221207,
      "num_tokens": 16497689.0,
      "step": 8145
    },
    {
      "entropy": 6.253456401824951,
      "epoch": 0.7122258149086778,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004956031261207353,
      "loss": 6.1228,
      "mean_token_accuracy": 0.13987948149442672,
      "num_tokens": 16508515.0,
      "step": 8150
    },
    {
      "entropy": 6.206867694854736,
      "epoch": 0.7126627632613826,
      "grad_norm": 2.0625,
      "learning_rate": 0.00049559699378099,
      "loss": 6.0309,
      "mean_token_accuracy": 0.15083636566996575,
      "num_tokens": 16519184.0,
      "step": 8155
    },
    {
      "entropy": 6.259245777130127,
      "epoch": 0.7130997116140873,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004955908572100787,
      "loss": 6.1133,
      "mean_token_accuracy": 0.13577307909727096,
      "num_tokens": 16529917.0,
      "step": 8160
    },
    {
      "entropy": 6.258401250839233,
      "epoch": 0.7135366599667919,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004955847164081192,
      "loss": 6.0413,
      "mean_token_accuracy": 0.1360163688659668,
      "num_tokens": 16539687.0,
      "step": 8165
    },
    {
      "entropy": 6.217234992980957,
      "epoch": 0.7139736083194966,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004955785713752292,
      "loss": 6.1034,
      "mean_token_accuracy": 0.14174101278185844,
      "num_tokens": 16550273.0,
      "step": 8170
    },
    {
      "entropy": 6.310365152359009,
      "epoch": 0.7144105566722013,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004955724221115267,
      "loss": 6.1547,
      "mean_token_accuracy": 0.13744727075099944,
      "num_tokens": 16562215.0,
      "step": 8175
    },
    {
      "entropy": 6.33936333656311,
      "epoch": 0.714847505024906,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004955662686171295,
      "loss": 6.1254,
      "mean_token_accuracy": 0.14110765382647514,
      "num_tokens": 16573218.0,
      "step": 8180
    },
    {
      "entropy": 6.30700569152832,
      "epoch": 0.7152844533776108,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004955601108921557,
      "loss": 6.1039,
      "mean_token_accuracy": 0.1380538746714592,
      "num_tokens": 16582740.0,
      "step": 8185
    },
    {
      "entropy": 6.177458715438843,
      "epoch": 0.7157214017303155,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004955539489367235,
      "loss": 6.0794,
      "mean_token_accuracy": 0.14526512026786803,
      "num_tokens": 16592169.0,
      "step": 8190
    },
    {
      "entropy": 6.230506229400635,
      "epoch": 0.7161583500830202,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004955477827509509,
      "loss": 6.0438,
      "mean_token_accuracy": 0.1382740132510662,
      "num_tokens": 16602706.0,
      "step": 8195
    },
    {
      "entropy": 6.224744129180908,
      "epoch": 0.7165952984357249,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004955416123349562,
      "loss": 6.0184,
      "mean_token_accuracy": 0.1419399008154869,
      "num_tokens": 16612012.0,
      "step": 8200
    },
    {
      "entropy": 6.212010669708252,
      "epoch": 0.7170322467884296,
      "grad_norm": 2.4375,
      "learning_rate": 0.000495535437688858,
      "loss": 6.1355,
      "mean_token_accuracy": 0.1388662949204445,
      "num_tokens": 16621798.0,
      "step": 8205
    },
    {
      "entropy": 6.242937612533569,
      "epoch": 0.7174691951411343,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004955292588127744,
      "loss": 6.0544,
      "mean_token_accuracy": 0.1456645667552948,
      "num_tokens": 16631792.0,
      "step": 8210
    },
    {
      "entropy": 6.3231405258178714,
      "epoch": 0.7179061434938391,
      "grad_norm": 1.859375,
      "learning_rate": 0.000495523075706824,
      "loss": 6.0022,
      "mean_token_accuracy": 0.1423792839050293,
      "num_tokens": 16641655.0,
      "step": 8215
    },
    {
      "entropy": 6.209512710571289,
      "epoch": 0.7183430918465438,
      "grad_norm": 2.125,
      "learning_rate": 0.0004955168883711257,
      "loss": 6.0916,
      "mean_token_accuracy": 0.1400701344013214,
      "num_tokens": 16651210.0,
      "step": 8220
    },
    {
      "entropy": 6.239299964904785,
      "epoch": 0.7187800401992485,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004955106968057977,
      "loss": 6.055,
      "mean_token_accuracy": 0.13928594663739205,
      "num_tokens": 16660078.0,
      "step": 8225
    },
    {
      "entropy": 6.282232236862183,
      "epoch": 0.7192169885519532,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004955045010109592,
      "loss": 6.0895,
      "mean_token_accuracy": 0.13874608874320984,
      "num_tokens": 16671211.0,
      "step": 8230
    },
    {
      "entropy": 6.26195707321167,
      "epoch": 0.7196539369046578,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004954983009867286,
      "loss": 6.0128,
      "mean_token_accuracy": 0.1426847517490387,
      "num_tokens": 16680850.0,
      "step": 8235
    },
    {
      "entropy": 6.222715616226196,
      "epoch": 0.7200908852573625,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004954920967332252,
      "loss": 5.9936,
      "mean_token_accuracy": 0.14225781932473183,
      "num_tokens": 16690447.0,
      "step": 8240
    },
    {
      "entropy": 6.199004411697388,
      "epoch": 0.7205278336100673,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004954858882505678,
      "loss": 6.0999,
      "mean_token_accuracy": 0.13997026309370994,
      "num_tokens": 16701061.0,
      "step": 8245
    },
    {
      "entropy": 6.212587594985962,
      "epoch": 0.720964781962772,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004954796755388755,
      "loss": 6.001,
      "mean_token_accuracy": 0.14577565342187881,
      "num_tokens": 16711139.0,
      "step": 8250
    },
    {
      "entropy": 6.244093561172486,
      "epoch": 0.7214017303154767,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004954734585982675,
      "loss": 6.1077,
      "mean_token_accuracy": 0.13529232516884804,
      "num_tokens": 16722343.0,
      "step": 8255
    },
    {
      "entropy": 6.283843612670898,
      "epoch": 0.7218386786681814,
      "grad_norm": 2.25,
      "learning_rate": 0.0004954672374288631,
      "loss": 6.1331,
      "mean_token_accuracy": 0.13186091929674149,
      "num_tokens": 16733475.0,
      "step": 8260
    },
    {
      "entropy": 6.264413690567016,
      "epoch": 0.7222756270208861,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004954610120307815,
      "loss": 6.0815,
      "mean_token_accuracy": 0.13933211341500282,
      "num_tokens": 16744356.0,
      "step": 8265
    },
    {
      "entropy": 6.231073570251465,
      "epoch": 0.7227125753735908,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004954547824041422,
      "loss": 6.101,
      "mean_token_accuracy": 0.13735386803746225,
      "num_tokens": 16755275.0,
      "step": 8270
    },
    {
      "entropy": 6.183216857910156,
      "epoch": 0.7231495237262956,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004954485485490647,
      "loss": 6.0098,
      "mean_token_accuracy": 0.14503678679466248,
      "num_tokens": 16766114.0,
      "step": 8275
    },
    {
      "entropy": 6.222272157669067,
      "epoch": 0.7235864720790003,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004954423104656685,
      "loss": 6.0066,
      "mean_token_accuracy": 0.15063316375017166,
      "num_tokens": 16776391.0,
      "step": 8280
    },
    {
      "entropy": 6.177164506912232,
      "epoch": 0.724023420431705,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004954360681540732,
      "loss": 5.9861,
      "mean_token_accuracy": 0.14972576797008513,
      "num_tokens": 16784845.0,
      "step": 8285
    },
    {
      "entropy": 6.165531492233276,
      "epoch": 0.7244603687844097,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004954298216143986,
      "loss": 6.0193,
      "mean_token_accuracy": 0.14583383798599242,
      "num_tokens": 16794735.0,
      "step": 8290
    },
    {
      "entropy": 6.314350938796997,
      "epoch": 0.7248973171371144,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004954235708467647,
      "loss": 6.101,
      "mean_token_accuracy": 0.14092472940683365,
      "num_tokens": 16803643.0,
      "step": 8295
    },
    {
      "entropy": 6.289931726455689,
      "epoch": 0.725334265489819,
      "grad_norm": 2.09375,
      "learning_rate": 0.000495417315851291,
      "loss": 6.0928,
      "mean_token_accuracy": 0.14327989518642426,
      "num_tokens": 16813064.0,
      "step": 8300
    },
    {
      "entropy": 6.163081979751587,
      "epoch": 0.7257712138425239,
      "grad_norm": 5.84375,
      "learning_rate": 0.0004954110566280978,
      "loss": 6.118,
      "mean_token_accuracy": 0.13695041239261627,
      "num_tokens": 16823166.0,
      "step": 8305
    },
    {
      "entropy": 6.233119249343872,
      "epoch": 0.7262081621952285,
      "grad_norm": 2.8125,
      "learning_rate": 0.000495404793177305,
      "loss": 6.1137,
      "mean_token_accuracy": 0.14075919687747956,
      "num_tokens": 16833546.0,
      "step": 8310
    },
    {
      "entropy": 6.289675045013428,
      "epoch": 0.7266451105479332,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004953985254990328,
      "loss": 5.977,
      "mean_token_accuracy": 0.1457753524184227,
      "num_tokens": 16843794.0,
      "step": 8315
    },
    {
      "entropy": 6.305609607696534,
      "epoch": 0.7270820589006379,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004953922535934014,
      "loss": 6.1238,
      "mean_token_accuracy": 0.1379474401473999,
      "num_tokens": 16855677.0,
      "step": 8320
    },
    {
      "entropy": 6.25561408996582,
      "epoch": 0.7275190072533426,
      "grad_norm": 3.546875,
      "learning_rate": 0.000495385977460531,
      "loss": 6.0786,
      "mean_token_accuracy": 0.13430355340242386,
      "num_tokens": 16866137.0,
      "step": 8325
    },
    {
      "entropy": 6.246871757507324,
      "epoch": 0.7279559556060473,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004953796971005422,
      "loss": 5.9903,
      "mean_token_accuracy": 0.14452763125300408,
      "num_tokens": 16875997.0,
      "step": 8330
    },
    {
      "entropy": 6.188179063796997,
      "epoch": 0.7283929039587521,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004953734125135552,
      "loss": 6.0429,
      "mean_token_accuracy": 0.1476132944226265,
      "num_tokens": 16885838.0,
      "step": 8335
    },
    {
      "entropy": 6.1498676300048825,
      "epoch": 0.7288298523114568,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004953671236996909,
      "loss": 5.9903,
      "mean_token_accuracy": 0.14977214187383653,
      "num_tokens": 16895533.0,
      "step": 8340
    },
    {
      "entropy": 6.210428714752197,
      "epoch": 0.7292668006641615,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004953608306590696,
      "loss": 6.1168,
      "mean_token_accuracy": 0.13882470726966858,
      "num_tokens": 16906258.0,
      "step": 8345
    },
    {
      "entropy": 6.277710390090943,
      "epoch": 0.7297037490168662,
      "grad_norm": 7.25,
      "learning_rate": 0.000495354533391812,
      "loss": 6.0658,
      "mean_token_accuracy": 0.1460177093744278,
      "num_tokens": 16917123.0,
      "step": 8350
    },
    {
      "entropy": 6.2469895362854,
      "epoch": 0.7301406973695709,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004953482318980389,
      "loss": 6.0819,
      "mean_token_accuracy": 0.146676953881979,
      "num_tokens": 16927533.0,
      "step": 8355
    },
    {
      "entropy": 6.306425762176514,
      "epoch": 0.7305776457222757,
      "grad_norm": 2.75,
      "learning_rate": 0.0004953419261778714,
      "loss": 6.0353,
      "mean_token_accuracy": 0.14144542664289475,
      "num_tokens": 16937763.0,
      "step": 8360
    },
    {
      "entropy": 6.167293548583984,
      "epoch": 0.7310145940749804,
      "grad_norm": 3.75,
      "learning_rate": 0.0004953356162314304,
      "loss": 5.9904,
      "mean_token_accuracy": 0.1453294850885868,
      "num_tokens": 16947173.0,
      "step": 8365
    },
    {
      "entropy": 6.2046772003173825,
      "epoch": 0.7314515424276851,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004953293020588368,
      "loss": 6.0751,
      "mean_token_accuracy": 0.1462822139263153,
      "num_tokens": 16956856.0,
      "step": 8370
    },
    {
      "entropy": 6.280859804153442,
      "epoch": 0.7318884907803898,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004953229836602115,
      "loss": 6.0658,
      "mean_token_accuracy": 0.14148651659488679,
      "num_tokens": 16966705.0,
      "step": 8375
    },
    {
      "entropy": 6.276563119888306,
      "epoch": 0.7323254391330944,
      "grad_norm": 2.15625,
      "learning_rate": 0.000495316661035676,
      "loss": 6.0953,
      "mean_token_accuracy": 0.13892802894115447,
      "num_tokens": 16977026.0,
      "step": 8380
    },
    {
      "entropy": 6.253737878799439,
      "epoch": 0.7327623874857991,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004953103341853517,
      "loss": 6.0175,
      "mean_token_accuracy": 0.14437578320503236,
      "num_tokens": 16986800.0,
      "step": 8385
    },
    {
      "entropy": 6.184905242919922,
      "epoch": 0.7331993358385039,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004953040031093596,
      "loss": 6.0095,
      "mean_token_accuracy": 0.14561595544219016,
      "num_tokens": 16996392.0,
      "step": 8390
    },
    {
      "entropy": 6.1834025382995605,
      "epoch": 0.7336362841912086,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004952976678078213,
      "loss": 6.0258,
      "mean_token_accuracy": 0.13908783942461014,
      "num_tokens": 17006430.0,
      "step": 8395
    },
    {
      "entropy": 6.277691650390625,
      "epoch": 0.7340732325439133,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004952913282808584,
      "loss": 6.0718,
      "mean_token_accuracy": 0.13975248634815216,
      "num_tokens": 17017168.0,
      "step": 8400
    },
    {
      "entropy": 6.195901918411255,
      "epoch": 0.734510180896618,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004952849845285922,
      "loss": 5.9861,
      "mean_token_accuracy": 0.1482916310429573,
      "num_tokens": 17027337.0,
      "step": 8405
    },
    {
      "entropy": 6.192442750930786,
      "epoch": 0.7349471292493227,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004952786365511447,
      "loss": 5.9609,
      "mean_token_accuracy": 0.1502939209342003,
      "num_tokens": 17036839.0,
      "step": 8410
    },
    {
      "entropy": 6.207208728790283,
      "epoch": 0.7353840776020274,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004952722843486375,
      "loss": 6.1754,
      "mean_token_accuracy": 0.13479250371456147,
      "num_tokens": 17048221.0,
      "step": 8415
    },
    {
      "entropy": 6.183388614654541,
      "epoch": 0.7358210259547322,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004952659279211924,
      "loss": 5.9936,
      "mean_token_accuracy": 0.14208328127861022,
      "num_tokens": 17057694.0,
      "step": 8420
    },
    {
      "entropy": 6.24350118637085,
      "epoch": 0.7362579743074369,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004952595672689315,
      "loss": 6.0821,
      "mean_token_accuracy": 0.1346749857068062,
      "num_tokens": 17066492.0,
      "step": 8425
    },
    {
      "entropy": 6.296481704711914,
      "epoch": 0.7366949226601416,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004952532023919766,
      "loss": 6.0853,
      "mean_token_accuracy": 0.13946571722626686,
      "num_tokens": 17077211.0,
      "step": 8430
    },
    {
      "entropy": 6.272731971740723,
      "epoch": 0.7371318710128463,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004952468332904498,
      "loss": 6.0663,
      "mean_token_accuracy": 0.13983459323644637,
      "num_tokens": 17088665.0,
      "step": 8435
    },
    {
      "entropy": 6.221385145187378,
      "epoch": 0.737568819365551,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004952404599644735,
      "loss": 6.0671,
      "mean_token_accuracy": 0.1424330674111843,
      "num_tokens": 17099767.0,
      "step": 8440
    },
    {
      "entropy": 6.212916564941406,
      "epoch": 0.7380057677182557,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004952340824141698,
      "loss": 5.9632,
      "mean_token_accuracy": 0.14879536181688308,
      "num_tokens": 17109414.0,
      "step": 8445
    },
    {
      "entropy": 6.184593915939331,
      "epoch": 0.7384427160709605,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004952277006396609,
      "loss": 6.0458,
      "mean_token_accuracy": 0.13909671306610108,
      "num_tokens": 17119107.0,
      "step": 8450
    },
    {
      "entropy": 6.2288893699646,
      "epoch": 0.7388796644236652,
      "grad_norm": 2.0,
      "learning_rate": 0.0004952213146410693,
      "loss": 6.0838,
      "mean_token_accuracy": 0.13814739882946014,
      "num_tokens": 17129010.0,
      "step": 8455
    },
    {
      "entropy": 6.261946964263916,
      "epoch": 0.7393166127763698,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004952149244185174,
      "loss": 6.0964,
      "mean_token_accuracy": 0.13839999288320542,
      "num_tokens": 17138951.0,
      "step": 8460
    },
    {
      "entropy": 6.250641870498657,
      "epoch": 0.7397535611290745,
      "grad_norm": 3.75,
      "learning_rate": 0.0004952085299721279,
      "loss": 6.0539,
      "mean_token_accuracy": 0.14264668449759482,
      "num_tokens": 17149416.0,
      "step": 8465
    },
    {
      "entropy": 6.261624002456665,
      "epoch": 0.7401905094817792,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004952021313020235,
      "loss": 6.0094,
      "mean_token_accuracy": 0.14614282846450805,
      "num_tokens": 17160068.0,
      "step": 8470
    },
    {
      "entropy": 6.160351133346557,
      "epoch": 0.7406274578344839,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004951957284083269,
      "loss": 6.0394,
      "mean_token_accuracy": 0.14056818932294846,
      "num_tokens": 17168972.0,
      "step": 8475
    },
    {
      "entropy": 6.171355581283569,
      "epoch": 0.7410644061871887,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004951893212911607,
      "loss": 6.0005,
      "mean_token_accuracy": 0.1472700320184231,
      "num_tokens": 17178239.0,
      "step": 8480
    },
    {
      "entropy": 6.282161617279053,
      "epoch": 0.7415013545398934,
      "grad_norm": 3.0,
      "learning_rate": 0.000495182909950648,
      "loss": 6.1736,
      "mean_token_accuracy": 0.1356131374835968,
      "num_tokens": 17190619.0,
      "step": 8485
    },
    {
      "entropy": 6.201739168167114,
      "epoch": 0.7419383028925981,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004951764943869117,
      "loss": 5.9605,
      "mean_token_accuracy": 0.14864636063575745,
      "num_tokens": 17201041.0,
      "step": 8490
    },
    {
      "entropy": 6.141462278366089,
      "epoch": 0.7423752512453028,
      "grad_norm": 2.25,
      "learning_rate": 0.0004951700746000749,
      "loss": 6.0711,
      "mean_token_accuracy": 0.14122827351093292,
      "num_tokens": 17211779.0,
      "step": 8495
    },
    {
      "entropy": 6.1928120136260985,
      "epoch": 0.7428121995980075,
      "grad_norm": 3.125,
      "learning_rate": 0.0004951636505902606,
      "loss": 6.0281,
      "mean_token_accuracy": 0.1515021190047264,
      "num_tokens": 17221249.0,
      "step": 8500
    },
    {
      "entropy": 6.260330533981323,
      "epoch": 0.7432491479507122,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004951572223575922,
      "loss": 6.0783,
      "mean_token_accuracy": 0.13569876849651336,
      "num_tokens": 17231634.0,
      "step": 8505
    },
    {
      "entropy": 6.245717906951905,
      "epoch": 0.743686096303417,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004951507899021929,
      "loss": 6.0935,
      "mean_token_accuracy": 0.14073003754019736,
      "num_tokens": 17241751.0,
      "step": 8510
    },
    {
      "entropy": 6.228300333023071,
      "epoch": 0.7441230446561217,
      "grad_norm": 2.65625,
      "learning_rate": 0.000495144353224186,
      "loss": 6.0844,
      "mean_token_accuracy": 0.14138879254460335,
      "num_tokens": 17251962.0,
      "step": 8515
    },
    {
      "entropy": 6.234739637374878,
      "epoch": 0.7445599930088264,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004951379123236952,
      "loss": 5.9775,
      "mean_token_accuracy": 0.14257938861846925,
      "num_tokens": 17261895.0,
      "step": 8520
    },
    {
      "entropy": 6.225541877746582,
      "epoch": 0.744996941361531,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004951314672008439,
      "loss": 6.0575,
      "mean_token_accuracy": 0.1443983182311058,
      "num_tokens": 17272025.0,
      "step": 8525
    },
    {
      "entropy": 6.255196285247803,
      "epoch": 0.7454338897142357,
      "grad_norm": 2.25,
      "learning_rate": 0.0004951250178557556,
      "loss": 6.0426,
      "mean_token_accuracy": 0.15038821697235108,
      "num_tokens": 17282705.0,
      "step": 8530
    },
    {
      "entropy": 6.1981690406799315,
      "epoch": 0.7458708380669404,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004951185642885542,
      "loss": 5.9895,
      "mean_token_accuracy": 0.1476744569838047,
      "num_tokens": 17293045.0,
      "step": 8535
    },
    {
      "entropy": 6.228192090988159,
      "epoch": 0.7463077864196452,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004951121064993633,
      "loss": 6.0772,
      "mean_token_accuracy": 0.13934364169836044,
      "num_tokens": 17303893.0,
      "step": 8540
    },
    {
      "entropy": 6.282638168334961,
      "epoch": 0.7467447347723499,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004951056444883069,
      "loss": 6.0682,
      "mean_token_accuracy": 0.13697793558239937,
      "num_tokens": 17314250.0,
      "step": 8545
    },
    {
      "entropy": 6.217985582351685,
      "epoch": 0.7471816831250546,
      "grad_norm": 2.59375,
      "learning_rate": 0.000495099178255509,
      "loss": 5.9987,
      "mean_token_accuracy": 0.1416642740368843,
      "num_tokens": 17324268.0,
      "step": 8550
    },
    {
      "entropy": 6.198323345184326,
      "epoch": 0.7476186314777593,
      "grad_norm": 2.0,
      "learning_rate": 0.0004950927078010934,
      "loss": 6.0926,
      "mean_token_accuracy": 0.13649949058890343,
      "num_tokens": 17334512.0,
      "step": 8555
    },
    {
      "entropy": 6.267451524734497,
      "epoch": 0.748055579830464,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004950862331251845,
      "loss": 6.0463,
      "mean_token_accuracy": 0.14405351355671883,
      "num_tokens": 17346244.0,
      "step": 8560
    },
    {
      "entropy": 6.216365003585816,
      "epoch": 0.7484925281831688,
      "grad_norm": 2.375,
      "learning_rate": 0.0004950797542279061,
      "loss": 6.0453,
      "mean_token_accuracy": 0.14253236427903176,
      "num_tokens": 17355832.0,
      "step": 8565
    },
    {
      "entropy": 6.192827749252319,
      "epoch": 0.7489294765358735,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004950732711093828,
      "loss": 6.0718,
      "mean_token_accuracy": 0.14349730908870698,
      "num_tokens": 17366871.0,
      "step": 8570
    },
    {
      "entropy": 6.305753374099732,
      "epoch": 0.7493664248885782,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004950667837697388,
      "loss": 6.1303,
      "mean_token_accuracy": 0.13418168425559998,
      "num_tokens": 17377242.0,
      "step": 8575
    },
    {
      "entropy": 6.3307852268219,
      "epoch": 0.7498033732412829,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004950602922090988,
      "loss": 6.0811,
      "mean_token_accuracy": 0.13639613315463067,
      "num_tokens": 17387942.0,
      "step": 8580
    },
    {
      "entropy": 6.26843581199646,
      "epoch": 0.7502403215939876,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004950537964275868,
      "loss": 6.0543,
      "mean_token_accuracy": 0.13597817122936248,
      "num_tokens": 17397413.0,
      "step": 8585
    },
    {
      "entropy": 6.206974363327026,
      "epoch": 0.7506772699466923,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004950472964253278,
      "loss": 6.0139,
      "mean_token_accuracy": 0.1416654609143734,
      "num_tokens": 17408456.0,
      "step": 8590
    },
    {
      "entropy": 6.206663322448731,
      "epoch": 0.7511142182993971,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004950407922024463,
      "loss": 6.0102,
      "mean_token_accuracy": 0.14704025387763978,
      "num_tokens": 17417805.0,
      "step": 8595
    },
    {
      "entropy": 6.196809720993042,
      "epoch": 0.7515511666521018,
      "grad_norm": 2.75,
      "learning_rate": 0.0004950342837590671,
      "loss": 5.9645,
      "mean_token_accuracy": 0.15146640837192535,
      "num_tokens": 17426932.0,
      "step": 8600
    },
    {
      "entropy": 6.19592547416687,
      "epoch": 0.7519881150048064,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004950277710953152,
      "loss": 6.0623,
      "mean_token_accuracy": 0.14430806338787078,
      "num_tokens": 17436770.0,
      "step": 8605
    },
    {
      "entropy": 6.190313053131104,
      "epoch": 0.7524250633575111,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004950212542113153,
      "loss": 5.961,
      "mean_token_accuracy": 0.14544311463832854,
      "num_tokens": 17447852.0,
      "step": 8610
    },
    {
      "entropy": 6.149398517608643,
      "epoch": 0.7528620117102158,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004950147331071924,
      "loss": 5.99,
      "mean_token_accuracy": 0.1404561571776867,
      "num_tokens": 17458020.0,
      "step": 8615
    },
    {
      "entropy": 6.211113929748535,
      "epoch": 0.7532989600629205,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004950082077830717,
      "loss": 6.0906,
      "mean_token_accuracy": 0.13863237872719764,
      "num_tokens": 17468604.0,
      "step": 8620
    },
    {
      "entropy": 6.26945686340332,
      "epoch": 0.7537359084156253,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004950016782390782,
      "loss": 6.0325,
      "mean_token_accuracy": 0.14416363313794137,
      "num_tokens": 17478374.0,
      "step": 8625
    },
    {
      "entropy": 6.269142770767212,
      "epoch": 0.75417285676833,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004949951444753372,
      "loss": 6.0359,
      "mean_token_accuracy": 0.14331215620040894,
      "num_tokens": 17487894.0,
      "step": 8630
    },
    {
      "entropy": 6.221175622940064,
      "epoch": 0.7546098051210347,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004949886064919741,
      "loss": 5.996,
      "mean_token_accuracy": 0.14074719920754433,
      "num_tokens": 17498473.0,
      "step": 8635
    },
    {
      "entropy": 6.211667490005493,
      "epoch": 0.7550467534737394,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004949820642891142,
      "loss": 6.0298,
      "mean_token_accuracy": 0.14195635169744492,
      "num_tokens": 17507796.0,
      "step": 8640
    },
    {
      "entropy": 6.171490859985352,
      "epoch": 0.7554837018264441,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004949755178668832,
      "loss": 5.9967,
      "mean_token_accuracy": 0.1477346181869507,
      "num_tokens": 17517629.0,
      "step": 8645
    },
    {
      "entropy": 6.260648393630982,
      "epoch": 0.7559206501791488,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004949689672254064,
      "loss": 6.1036,
      "mean_token_accuracy": 0.1329352766275406,
      "num_tokens": 17528726.0,
      "step": 8650
    },
    {
      "entropy": 6.275094127655029,
      "epoch": 0.7563575985318536,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004949624123648096,
      "loss": 6.0014,
      "mean_token_accuracy": 0.1422254428267479,
      "num_tokens": 17539022.0,
      "step": 8655
    },
    {
      "entropy": 6.223832321166992,
      "epoch": 0.7567945468845583,
      "grad_norm": 2.375,
      "learning_rate": 0.0004949558532852185,
      "loss": 6.0461,
      "mean_token_accuracy": 0.1439749851822853,
      "num_tokens": 17549293.0,
      "step": 8660
    },
    {
      "entropy": 6.183422327041626,
      "epoch": 0.757231495237263,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004949492899867587,
      "loss": 6.0326,
      "mean_token_accuracy": 0.13842322826385497,
      "num_tokens": 17559072.0,
      "step": 8665
    },
    {
      "entropy": 6.190749454498291,
      "epoch": 0.7576684435899677,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004949427224695564,
      "loss": 6.0226,
      "mean_token_accuracy": 0.14318446591496467,
      "num_tokens": 17568757.0,
      "step": 8670
    },
    {
      "entropy": 6.226632213592529,
      "epoch": 0.7581053919426723,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004949361507337374,
      "loss": 6.0193,
      "mean_token_accuracy": 0.14537923634052277,
      "num_tokens": 17578288.0,
      "step": 8675
    },
    {
      "entropy": 6.2610218048095705,
      "epoch": 0.758542340295377,
      "grad_norm": 2.375,
      "learning_rate": 0.0004949295747794277,
      "loss": 6.116,
      "mean_token_accuracy": 0.1382712870836258,
      "num_tokens": 17588864.0,
      "step": 8680
    },
    {
      "entropy": 6.254505634307861,
      "epoch": 0.7589792886480818,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004949229946067537,
      "loss": 6.0921,
      "mean_token_accuracy": 0.14055313169956207,
      "num_tokens": 17599103.0,
      "step": 8685
    },
    {
      "entropy": 6.184361553192138,
      "epoch": 0.7594162370007865,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004949164102158413,
      "loss": 6.0025,
      "mean_token_accuracy": 0.1467547371983528,
      "num_tokens": 17608608.0,
      "step": 8690
    },
    {
      "entropy": 6.228552436828613,
      "epoch": 0.7598531853534912,
      "grad_norm": 2.78125,
      "learning_rate": 0.000494909821606817,
      "loss": 6.0745,
      "mean_token_accuracy": 0.14077405631542206,
      "num_tokens": 17618604.0,
      "step": 8695
    },
    {
      "entropy": 6.319025087356567,
      "epoch": 0.7602901337061959,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004949032287798069,
      "loss": 6.0387,
      "mean_token_accuracy": 0.14739960730075835,
      "num_tokens": 17628479.0,
      "step": 8700
    },
    {
      "entropy": 6.231346940994262,
      "epoch": 0.7607270820589006,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004948966317349379,
      "loss": 6.0412,
      "mean_token_accuracy": 0.14681292027235032,
      "num_tokens": 17637519.0,
      "step": 8705
    },
    {
      "entropy": 6.216566514968872,
      "epoch": 0.7611640304116053,
      "grad_norm": 3.375,
      "learning_rate": 0.0004948900304723361,
      "loss": 6.0174,
      "mean_token_accuracy": 0.1492878243327141,
      "num_tokens": 17647902.0,
      "step": 8710
    },
    {
      "entropy": 6.274780988693237,
      "epoch": 0.7616009787643101,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004948834249921285,
      "loss": 6.0401,
      "mean_token_accuracy": 0.1431396059691906,
      "num_tokens": 17658576.0,
      "step": 8715
    },
    {
      "entropy": 6.21819658279419,
      "epoch": 0.7620379271170148,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004948768152944414,
      "loss": 6.0763,
      "mean_token_accuracy": 0.14016674533486367,
      "num_tokens": 17667059.0,
      "step": 8720
    },
    {
      "entropy": 6.277122974395752,
      "epoch": 0.7624748754697195,
      "grad_norm": 4.125,
      "learning_rate": 0.0004948702013794018,
      "loss": 6.1627,
      "mean_token_accuracy": 0.1352974832057953,
      "num_tokens": 17677447.0,
      "step": 8725
    },
    {
      "entropy": 6.284607887268066,
      "epoch": 0.7629118238224242,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004948635832471366,
      "loss": 6.0174,
      "mean_token_accuracy": 0.15216316804289817,
      "num_tokens": 17687893.0,
      "step": 8730
    },
    {
      "entropy": 6.226282358169556,
      "epoch": 0.7633487721751289,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004948569608977726,
      "loss": 6.081,
      "mean_token_accuracy": 0.1410996735095978,
      "num_tokens": 17698599.0,
      "step": 8735
    },
    {
      "entropy": 6.234055852890014,
      "epoch": 0.7637857205278336,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004948503343314369,
      "loss": 6.0832,
      "mean_token_accuracy": 0.13510883152484893,
      "num_tokens": 17709711.0,
      "step": 8740
    },
    {
      "entropy": 6.240414333343506,
      "epoch": 0.7642226688805384,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004948437035482566,
      "loss": 6.0425,
      "mean_token_accuracy": 0.14039023146033286,
      "num_tokens": 17719613.0,
      "step": 8745
    },
    {
      "entropy": 6.2663328647613525,
      "epoch": 0.764659617233243,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004948370685483588,
      "loss": 6.0177,
      "mean_token_accuracy": 0.15150065273046492,
      "num_tokens": 17729829.0,
      "step": 8750
    },
    {
      "entropy": 6.268952989578247,
      "epoch": 0.7650965655859477,
      "grad_norm": 2.0,
      "learning_rate": 0.0004948304293318709,
      "loss": 6.0165,
      "mean_token_accuracy": 0.1446221023797989,
      "num_tokens": 17739254.0,
      "step": 8755
    },
    {
      "entropy": 6.228422737121582,
      "epoch": 0.7655335139386524,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004948237858989201,
      "loss": 6.044,
      "mean_token_accuracy": 0.13771776854991913,
      "num_tokens": 17749294.0,
      "step": 8760
    },
    {
      "entropy": 6.206667804718018,
      "epoch": 0.7659704622913571,
      "grad_norm": 2.40625,
      "learning_rate": 0.000494817138249634,
      "loss": 5.9668,
      "mean_token_accuracy": 0.150910571962595,
      "num_tokens": 17758866.0,
      "step": 8765
    },
    {
      "entropy": 6.358490991592407,
      "epoch": 0.7664074106440619,
      "grad_norm": 3.546875,
      "learning_rate": 0.00049481048638414,
      "loss": 6.1511,
      "mean_token_accuracy": 0.1335010588169098,
      "num_tokens": 17769202.0,
      "step": 8770
    },
    {
      "entropy": 6.226506757736206,
      "epoch": 0.7668443589967666,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004948038303025657,
      "loss": 6.0997,
      "mean_token_accuracy": 0.14109473377466203,
      "num_tokens": 17779125.0,
      "step": 8775
    },
    {
      "entropy": 6.270475721359253,
      "epoch": 0.7672813073494713,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004947971700050387,
      "loss": 6.0489,
      "mean_token_accuracy": 0.13899814411997796,
      "num_tokens": 17788696.0,
      "step": 8780
    },
    {
      "entropy": 6.223231029510498,
      "epoch": 0.767718255702176,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004947905054916869,
      "loss": 6.073,
      "mean_token_accuracy": 0.1383361414074898,
      "num_tokens": 17799867.0,
      "step": 8785
    },
    {
      "entropy": 6.293824672698975,
      "epoch": 0.7681552040548807,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004947838367626379,
      "loss": 6.0456,
      "mean_token_accuracy": 0.13762528523802758,
      "num_tokens": 17809418.0,
      "step": 8790
    },
    {
      "entropy": 6.258719491958618,
      "epoch": 0.7685921524075854,
      "grad_norm": 5.625,
      "learning_rate": 0.0004947771638180198,
      "loss": 5.9662,
      "mean_token_accuracy": 0.15071939826011657,
      "num_tokens": 17818554.0,
      "step": 8795
    },
    {
      "entropy": 6.220383358001709,
      "epoch": 0.7690291007602902,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004947704866579605,
      "loss": 6.0176,
      "mean_token_accuracy": 0.14543282389640808,
      "num_tokens": 17828918.0,
      "step": 8800
    },
    {
      "entropy": 6.254160785675049,
      "epoch": 0.7694660491129949,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004947638052825883,
      "loss": 6.1642,
      "mean_token_accuracy": 0.13365367203950881,
      "num_tokens": 17840791.0,
      "step": 8805
    },
    {
      "entropy": 6.312784910202026,
      "epoch": 0.7699029974656996,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004947571196920309,
      "loss": 6.0491,
      "mean_token_accuracy": 0.14061523079872132,
      "num_tokens": 17850675.0,
      "step": 8810
    },
    {
      "entropy": 6.22687726020813,
      "epoch": 0.7703399458184043,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004947504298864169,
      "loss": 6.0204,
      "mean_token_accuracy": 0.14554693624377252,
      "num_tokens": 17860684.0,
      "step": 8815
    },
    {
      "entropy": 6.176032638549804,
      "epoch": 0.770776894171109,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004947437358658746,
      "loss": 6.0348,
      "mean_token_accuracy": 0.14274812787771224,
      "num_tokens": 17870394.0,
      "step": 8820
    },
    {
      "entropy": 6.246598482131958,
      "epoch": 0.7712138425238136,
      "grad_norm": 3.5,
      "learning_rate": 0.0004947370376305321,
      "loss": 6.0102,
      "mean_token_accuracy": 0.14171857833862306,
      "num_tokens": 17881256.0,
      "step": 8825
    },
    {
      "entropy": 6.213253498077393,
      "epoch": 0.7716507908765184,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004947303351805183,
      "loss": 6.0616,
      "mean_token_accuracy": 0.14157484471797943,
      "num_tokens": 17890888.0,
      "step": 8830
    },
    {
      "entropy": 6.2633696556091305,
      "epoch": 0.7720877392292231,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004947236285159615,
      "loss": 6.0172,
      "mean_token_accuracy": 0.13945037126541138,
      "num_tokens": 17900507.0,
      "step": 8835
    },
    {
      "entropy": 6.311097908020019,
      "epoch": 0.7725246875819278,
      "grad_norm": 2.75,
      "learning_rate": 0.0004947169176369902,
      "loss": 6.0912,
      "mean_token_accuracy": 0.14449197351932525,
      "num_tokens": 17911333.0,
      "step": 8840
    },
    {
      "entropy": 6.277305555343628,
      "epoch": 0.7729616359346325,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004947102025437334,
      "loss": 6.0376,
      "mean_token_accuracy": 0.14333002939820289,
      "num_tokens": 17921540.0,
      "step": 8845
    },
    {
      "entropy": 6.267017030715943,
      "epoch": 0.7733985842873372,
      "grad_norm": 3.125,
      "learning_rate": 0.0004947034832363199,
      "loss": 6.1127,
      "mean_token_accuracy": 0.14070233479142188,
      "num_tokens": 17931820.0,
      "step": 8850
    },
    {
      "entropy": 6.212412738800049,
      "epoch": 0.7738355326400419,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004946967597148782,
      "loss": 6.041,
      "mean_token_accuracy": 0.13900020495057105,
      "num_tokens": 17942717.0,
      "step": 8855
    },
    {
      "entropy": 6.234614562988281,
      "epoch": 0.7742724809927467,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004946900319795377,
      "loss": 6.0407,
      "mean_token_accuracy": 0.14958684593439103,
      "num_tokens": 17953775.0,
      "step": 8860
    },
    {
      "entropy": 6.263529682159424,
      "epoch": 0.7747094293454514,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004946833000304274,
      "loss": 6.0137,
      "mean_token_accuracy": 0.1431668758392334,
      "num_tokens": 17963877.0,
      "step": 8865
    },
    {
      "entropy": 6.2535178661346436,
      "epoch": 0.7751463776981561,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004946765638676761,
      "loss": 6.029,
      "mean_token_accuracy": 0.14579931572079657,
      "num_tokens": 17974015.0,
      "step": 8870
    },
    {
      "entropy": 6.155847454071045,
      "epoch": 0.7755833260508608,
      "grad_norm": 2.5,
      "learning_rate": 0.0004946698234914132,
      "loss": 6.0453,
      "mean_token_accuracy": 0.1468222126364708,
      "num_tokens": 17984189.0,
      "step": 8875
    },
    {
      "entropy": 6.211731052398681,
      "epoch": 0.7760202744035655,
      "grad_norm": 4.8125,
      "learning_rate": 0.000494663078901768,
      "loss": 5.9818,
      "mean_token_accuracy": 0.15032537877559662,
      "num_tokens": 17994615.0,
      "step": 8880
    },
    {
      "entropy": 6.22483377456665,
      "epoch": 0.7764572227562702,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004946563300988699,
      "loss": 6.0309,
      "mean_token_accuracy": 0.1406807705760002,
      "num_tokens": 18004727.0,
      "step": 8885
    },
    {
      "entropy": 6.251943349838257,
      "epoch": 0.776894171108975,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004946495770828482,
      "loss": 6.0829,
      "mean_token_accuracy": 0.13911537006497382,
      "num_tokens": 18014576.0,
      "step": 8890
    },
    {
      "entropy": 6.256908798217774,
      "epoch": 0.7773311194616797,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004946428198538326,
      "loss": 6.0464,
      "mean_token_accuracy": 0.14529479444026946,
      "num_tokens": 18024709.0,
      "step": 8895
    },
    {
      "entropy": 6.234633207321167,
      "epoch": 0.7777680678143843,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004946360584119526,
      "loss": 6.0003,
      "mean_token_accuracy": 0.14170599430799485,
      "num_tokens": 18034673.0,
      "step": 8900
    },
    {
      "entropy": 6.235233211517334,
      "epoch": 0.778205016167089,
      "grad_norm": 2.34375,
      "learning_rate": 0.000494629292757338,
      "loss": 6.0189,
      "mean_token_accuracy": 0.14689105078577996,
      "num_tokens": 18044774.0,
      "step": 8905
    },
    {
      "entropy": 6.176680564880371,
      "epoch": 0.7786419645197937,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004946225228901184,
      "loss": 6.0145,
      "mean_token_accuracy": 0.14234003871679307,
      "num_tokens": 18055244.0,
      "step": 8910
    },
    {
      "entropy": 6.275831079483032,
      "epoch": 0.7790789128724984,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004946157488104236,
      "loss": 6.1403,
      "mean_token_accuracy": 0.13471165522933007,
      "num_tokens": 18066051.0,
      "step": 8915
    },
    {
      "entropy": 6.3291338920593265,
      "epoch": 0.7795158612252032,
      "grad_norm": 4.875,
      "learning_rate": 0.0004946089705183839,
      "loss": 6.133,
      "mean_token_accuracy": 0.1323649190366268,
      "num_tokens": 18077693.0,
      "step": 8920
    },
    {
      "entropy": 6.2520617008209225,
      "epoch": 0.7799528095779079,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004946021880141289,
      "loss": 6.0715,
      "mean_token_accuracy": 0.1427716851234436,
      "num_tokens": 18088032.0,
      "step": 8925
    },
    {
      "entropy": 6.197879409790039,
      "epoch": 0.7803897579306126,
      "grad_norm": 1.796875,
      "learning_rate": 0.000494595401297789,
      "loss": 6.0263,
      "mean_token_accuracy": 0.1420186921954155,
      "num_tokens": 18097998.0,
      "step": 8930
    },
    {
      "entropy": 6.204164600372314,
      "epoch": 0.7808267062833173,
      "grad_norm": 2.375,
      "learning_rate": 0.0004945886103694941,
      "loss": 5.9896,
      "mean_token_accuracy": 0.1460868537425995,
      "num_tokens": 18107538.0,
      "step": 8935
    },
    {
      "entropy": 6.201864337921142,
      "epoch": 0.781263654636022,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004945818152293747,
      "loss": 6.0463,
      "mean_token_accuracy": 0.1413077510893345,
      "num_tokens": 18117792.0,
      "step": 8940
    },
    {
      "entropy": 6.282734966278076,
      "epoch": 0.7817006029887267,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004945750158775609,
      "loss": 6.085,
      "mean_token_accuracy": 0.14083878919482232,
      "num_tokens": 18128005.0,
      "step": 8945
    },
    {
      "entropy": 6.255158472061157,
      "epoch": 0.7821375513414315,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004945682123141833,
      "loss": 6.0458,
      "mean_token_accuracy": 0.15105608999729156,
      "num_tokens": 18138398.0,
      "step": 8950
    },
    {
      "entropy": 6.29579725265503,
      "epoch": 0.7825744996941362,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004945614045393724,
      "loss": 6.0194,
      "mean_token_accuracy": 0.1422487013041973,
      "num_tokens": 18147676.0,
      "step": 8955
    },
    {
      "entropy": 6.122381496429443,
      "epoch": 0.7830114480468409,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004945545925532587,
      "loss": 5.9191,
      "mean_token_accuracy": 0.15262250900268554,
      "num_tokens": 18156842.0,
      "step": 8960
    },
    {
      "entropy": 6.136346292495728,
      "epoch": 0.7834483963995456,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004945477763559728,
      "loss": 6.046,
      "mean_token_accuracy": 0.14279976412653922,
      "num_tokens": 18166902.0,
      "step": 8965
    },
    {
      "entropy": 6.245600891113281,
      "epoch": 0.7838853447522502,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004945409559476455,
      "loss": 6.033,
      "mean_token_accuracy": 0.146183517575264,
      "num_tokens": 18177561.0,
      "step": 8970
    },
    {
      "entropy": 6.1954508304595945,
      "epoch": 0.784322293104955,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004945341313284077,
      "loss": 6.0179,
      "mean_token_accuracy": 0.14810565561056138,
      "num_tokens": 18187739.0,
      "step": 8975
    },
    {
      "entropy": 6.176522874832154,
      "epoch": 0.7847592414576597,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004945273024983902,
      "loss": 5.9692,
      "mean_token_accuracy": 0.15180836766958236,
      "num_tokens": 18198435.0,
      "step": 8980
    },
    {
      "entropy": 6.29358811378479,
      "epoch": 0.7851961898103644,
      "grad_norm": 3.3125,
      "learning_rate": 0.000494520469457724,
      "loss": 5.9838,
      "mean_token_accuracy": 0.14602330476045608,
      "num_tokens": 18208374.0,
      "step": 8985
    },
    {
      "entropy": 6.118836355209351,
      "epoch": 0.7856331381630691,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004945136322065402,
      "loss": 6.0569,
      "mean_token_accuracy": 0.15092428624629975,
      "num_tokens": 18219229.0,
      "step": 8990
    },
    {
      "entropy": 6.1668847560882565,
      "epoch": 0.7860700865157738,
      "grad_norm": 2.75,
      "learning_rate": 0.0004945067907449699,
      "loss": 6.0319,
      "mean_token_accuracy": 0.14282664731144906,
      "num_tokens": 18229208.0,
      "step": 8995
    },
    {
      "entropy": 6.2895505905151365,
      "epoch": 0.7865070348684785,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004944999450731442,
      "loss": 6.1199,
      "mean_token_accuracy": 0.13042119517922401,
      "num_tokens": 18238454.0,
      "step": 9000
    },
    {
      "epoch": 0.7865070348684785,
      "eval_entropy": 6.187632340912894,
      "eval_loss": 6.064093112945557,
      "eval_mean_token_accuracy": 0.14895293781437627,
      "eval_num_tokens": 18238454.0,
      "eval_runtime": 28.2519,
      "eval_samples_per_second": 1310.142,
      "eval_steps_per_second": 163.777,
      "step": 9000
    },
    {
      "entropy": 6.350997018814087,
      "epoch": 0.7869439832211833,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004944930951911947,
      "loss": 6.0218,
      "mean_token_accuracy": 0.14062996804714203,
      "num_tokens": 18248845.0,
      "step": 9005
    },
    {
      "entropy": 6.235470581054687,
      "epoch": 0.787380931573888,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004944862410992526,
      "loss": 6.0184,
      "mean_token_accuracy": 0.14461514875292777,
      "num_tokens": 18258912.0,
      "step": 9010
    },
    {
      "entropy": 6.23782057762146,
      "epoch": 0.7878178799265927,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004944793827974494,
      "loss": 6.0697,
      "mean_token_accuracy": 0.13982785046100615,
      "num_tokens": 18269330.0,
      "step": 9015
    },
    {
      "entropy": 6.270591497421265,
      "epoch": 0.7882548282792974,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004944725202859165,
      "loss": 6.0564,
      "mean_token_accuracy": 0.13964123651385307,
      "num_tokens": 18279721.0,
      "step": 9020
    },
    {
      "entropy": 6.158119153976441,
      "epoch": 0.7886917766320021,
      "grad_norm": 1.75,
      "learning_rate": 0.0004944656535647858,
      "loss": 5.9632,
      "mean_token_accuracy": 0.1489733189344406,
      "num_tokens": 18288581.0,
      "step": 9025
    },
    {
      "entropy": 6.190168857574463,
      "epoch": 0.7891287249847068,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004944587826341888,
      "loss": 6.0052,
      "mean_token_accuracy": 0.1436466746032238,
      "num_tokens": 18298179.0,
      "step": 9030
    },
    {
      "entropy": 6.253784799575806,
      "epoch": 0.7895656733374116,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004944519074942574,
      "loss": 6.0209,
      "mean_token_accuracy": 0.14425861164927484,
      "num_tokens": 18307957.0,
      "step": 9035
    },
    {
      "entropy": 6.126337194442749,
      "epoch": 0.7900026216901163,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004944450281451234,
      "loss": 6.0161,
      "mean_token_accuracy": 0.13894797191023828,
      "num_tokens": 18317799.0,
      "step": 9040
    },
    {
      "entropy": 6.242519569396973,
      "epoch": 0.790439570042821,
      "grad_norm": 3.125,
      "learning_rate": 0.0004944381445869188,
      "loss": 6.065,
      "mean_token_accuracy": 0.14522837102413177,
      "num_tokens": 18327889.0,
      "step": 9045
    },
    {
      "entropy": 6.252655553817749,
      "epoch": 0.7908765183955256,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004944312568197756,
      "loss": 6.0534,
      "mean_token_accuracy": 0.1454584777355194,
      "num_tokens": 18338288.0,
      "step": 9050
    },
    {
      "entropy": 6.218295621871948,
      "epoch": 0.7913134667482303,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004944243648438258,
      "loss": 6.1079,
      "mean_token_accuracy": 0.13389210775494576,
      "num_tokens": 18349070.0,
      "step": 9055
    },
    {
      "entropy": 6.296699571609497,
      "epoch": 0.791750415100935,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004944174686592019,
      "loss": 6.0069,
      "mean_token_accuracy": 0.14725763127207755,
      "num_tokens": 18359683.0,
      "step": 9060
    },
    {
      "entropy": 6.2603308200836185,
      "epoch": 0.7921873634536398,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004944105682660358,
      "loss": 6.0214,
      "mean_token_accuracy": 0.14442202001810073,
      "num_tokens": 18371355.0,
      "step": 9065
    },
    {
      "entropy": 6.131388235092163,
      "epoch": 0.7926243118063445,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004944036636644601,
      "loss": 6.0293,
      "mean_token_accuracy": 0.1461861990392208,
      "num_tokens": 18381591.0,
      "step": 9070
    },
    {
      "entropy": 6.1675091743469235,
      "epoch": 0.7930612601590492,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004943967548546071,
      "loss": 6.006,
      "mean_token_accuracy": 0.1439528949558735,
      "num_tokens": 18390760.0,
      "step": 9075
    },
    {
      "entropy": 6.254599571228027,
      "epoch": 0.7934982085117539,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004943898418366096,
      "loss": 6.0541,
      "mean_token_accuracy": 0.14123787581920624,
      "num_tokens": 18400643.0,
      "step": 9080
    },
    {
      "entropy": 6.274891138076782,
      "epoch": 0.7939351568644586,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004943829246105997,
      "loss": 6.0899,
      "mean_token_accuracy": 0.1439736530184746,
      "num_tokens": 18411559.0,
      "step": 9085
    },
    {
      "entropy": 6.235226821899414,
      "epoch": 0.7943721052171633,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004943760031767106,
      "loss": 6.0753,
      "mean_token_accuracy": 0.1427392303943634,
      "num_tokens": 18422645.0,
      "step": 9090
    },
    {
      "entropy": 6.25205397605896,
      "epoch": 0.7948090535698681,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004943690775350746,
      "loss": 6.0323,
      "mean_token_accuracy": 0.14323223605751992,
      "num_tokens": 18432369.0,
      "step": 9095
    },
    {
      "entropy": 6.2375733852386475,
      "epoch": 0.7952460019225728,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004943621476858249,
      "loss": 6.0085,
      "mean_token_accuracy": 0.14254568815231322,
      "num_tokens": 18442143.0,
      "step": 9100
    },
    {
      "entropy": 6.159643507003784,
      "epoch": 0.7956829502752775,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004943552136290942,
      "loss": 6.0468,
      "mean_token_accuracy": 0.14481618478894234,
      "num_tokens": 18452161.0,
      "step": 9105
    },
    {
      "entropy": 6.264166975021363,
      "epoch": 0.7961198986279822,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004943482753650154,
      "loss": 6.1266,
      "mean_token_accuracy": 0.13852701485157012,
      "num_tokens": 18461873.0,
      "step": 9110
    },
    {
      "entropy": 6.3209999084472654,
      "epoch": 0.7965568469806869,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004943413328937221,
      "loss": 6.0409,
      "mean_token_accuracy": 0.14282170534133912,
      "num_tokens": 18472584.0,
      "step": 9115
    },
    {
      "entropy": 6.239671850204468,
      "epoch": 0.7969937953333915,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004943343862153467,
      "loss": 5.9867,
      "mean_token_accuracy": 0.14517282769083978,
      "num_tokens": 18482174.0,
      "step": 9120
    },
    {
      "entropy": 6.217202758789062,
      "epoch": 0.7974307436860963,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004943274353300231,
      "loss": 6.1233,
      "mean_token_accuracy": 0.1405440501868725,
      "num_tokens": 18492861.0,
      "step": 9125
    },
    {
      "entropy": 6.25796537399292,
      "epoch": 0.797867692038801,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004943204802378843,
      "loss": 6.048,
      "mean_token_accuracy": 0.1437128081917763,
      "num_tokens": 18502830.0,
      "step": 9130
    },
    {
      "entropy": 6.285251092910767,
      "epoch": 0.7983046403915057,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004943135209390638,
      "loss": 6.0325,
      "mean_token_accuracy": 0.14276544004678726,
      "num_tokens": 18513063.0,
      "step": 9135
    },
    {
      "entropy": 6.174069166183472,
      "epoch": 0.7987415887442104,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004943065574336949,
      "loss": 6.0217,
      "mean_token_accuracy": 0.14521622508764268,
      "num_tokens": 18523773.0,
      "step": 9140
    },
    {
      "entropy": 6.205516242980957,
      "epoch": 0.7991785370969151,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004942995897219115,
      "loss": 6.0379,
      "mean_token_accuracy": 0.14667071476578714,
      "num_tokens": 18533101.0,
      "step": 9145
    },
    {
      "entropy": 6.2843996524810795,
      "epoch": 0.7996154854496198,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004942926178038469,
      "loss": 6.0592,
      "mean_token_accuracy": 0.14422493949532508,
      "num_tokens": 18543863.0,
      "step": 9150
    },
    {
      "entropy": 6.172523546218872,
      "epoch": 0.8000524338023246,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004942856416796352,
      "loss": 6.0283,
      "mean_token_accuracy": 0.1453754872083664,
      "num_tokens": 18554249.0,
      "step": 9155
    },
    {
      "entropy": 6.226995134353638,
      "epoch": 0.8004893821550293,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004942786613494098,
      "loss": 6.0348,
      "mean_token_accuracy": 0.13995733261108398,
      "num_tokens": 18563525.0,
      "step": 9160
    },
    {
      "entropy": 6.261930751800537,
      "epoch": 0.800926330507734,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004942716768133048,
      "loss": 6.0216,
      "mean_token_accuracy": 0.14677432626485826,
      "num_tokens": 18572734.0,
      "step": 9165
    },
    {
      "entropy": 6.1590970993042,
      "epoch": 0.8013632788604387,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004942646880714543,
      "loss": 6.0708,
      "mean_token_accuracy": 0.14118833616375923,
      "num_tokens": 18581939.0,
      "step": 9170
    },
    {
      "entropy": 6.216191101074219,
      "epoch": 0.8018002272131434,
      "grad_norm": 2.015625,
      "learning_rate": 0.000494257695123992,
      "loss": 6.0573,
      "mean_token_accuracy": 0.1449749894440174,
      "num_tokens": 18591635.0,
      "step": 9175
    },
    {
      "entropy": 6.228406238555908,
      "epoch": 0.8022371755658482,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004942506979710524,
      "loss": 5.9754,
      "mean_token_accuracy": 0.14958217293024062,
      "num_tokens": 18602060.0,
      "step": 9180
    },
    {
      "entropy": 6.229957485198975,
      "epoch": 0.8026741239185529,
      "grad_norm": 2.375,
      "learning_rate": 0.0004942436966127693,
      "loss": 6.0288,
      "mean_token_accuracy": 0.14447982758283615,
      "num_tokens": 18611187.0,
      "step": 9185
    },
    {
      "entropy": 6.2823692798614506,
      "epoch": 0.8031110722712576,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004942366910492775,
      "loss": 5.9782,
      "mean_token_accuracy": 0.1483898475766182,
      "num_tokens": 18620329.0,
      "step": 9190
    },
    {
      "entropy": 6.275428104400635,
      "epoch": 0.8035480206239622,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004942296812807109,
      "loss": 6.0349,
      "mean_token_accuracy": 0.14326571002602578,
      "num_tokens": 18629600.0,
      "step": 9195
    },
    {
      "entropy": 6.196285676956177,
      "epoch": 0.8039849689766669,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004942226673072041,
      "loss": 6.0865,
      "mean_token_accuracy": 0.14189935475587845,
      "num_tokens": 18640771.0,
      "step": 9200
    },
    {
      "entropy": 6.146600818634033,
      "epoch": 0.8044219173293716,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004942156491288917,
      "loss": 5.9999,
      "mean_token_accuracy": 0.1470084510743618,
      "num_tokens": 18650606.0,
      "step": 9205
    },
    {
      "entropy": 6.216950178146362,
      "epoch": 0.8048588656820764,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004942086267459083,
      "loss": 6.0328,
      "mean_token_accuracy": 0.14324128851294518,
      "num_tokens": 18660487.0,
      "step": 9210
    },
    {
      "entropy": 6.258160352706909,
      "epoch": 0.8052958140347811,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004942016001583886,
      "loss": 6.0367,
      "mean_token_accuracy": 0.14204690083861352,
      "num_tokens": 18670987.0,
      "step": 9215
    },
    {
      "entropy": 6.230315351486206,
      "epoch": 0.8057327623874858,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004941945693664674,
      "loss": 6.103,
      "mean_token_accuracy": 0.13973608016967773,
      "num_tokens": 18682560.0,
      "step": 9220
    },
    {
      "entropy": 6.223834896087647,
      "epoch": 0.8061697107401905,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004941875343702795,
      "loss": 6.0555,
      "mean_token_accuracy": 0.14226875007152556,
      "num_tokens": 18692743.0,
      "step": 9225
    },
    {
      "entropy": 6.241143035888672,
      "epoch": 0.8066066590928952,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004941804951699597,
      "loss": 5.9943,
      "mean_token_accuracy": 0.1508474126458168,
      "num_tokens": 18703039.0,
      "step": 9230
    },
    {
      "entropy": 6.186641025543213,
      "epoch": 0.8070436074455999,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004941734517656432,
      "loss": 6.0682,
      "mean_token_accuracy": 0.14369699433445932,
      "num_tokens": 18713777.0,
      "step": 9235
    },
    {
      "entropy": 6.199865484237671,
      "epoch": 0.8074805557983047,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004941664041574651,
      "loss": 6.065,
      "mean_token_accuracy": 0.14116380661725997,
      "num_tokens": 18723422.0,
      "step": 9240
    },
    {
      "entropy": 6.224121236801148,
      "epoch": 0.8079175041510094,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004941593523455605,
      "loss": 6.0183,
      "mean_token_accuracy": 0.14372921362519264,
      "num_tokens": 18734369.0,
      "step": 9245
    },
    {
      "entropy": 6.283028888702392,
      "epoch": 0.8083544525037141,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004941522963300647,
      "loss": 6.0748,
      "mean_token_accuracy": 0.1408497080206871,
      "num_tokens": 18744647.0,
      "step": 9250
    },
    {
      "entropy": 6.2740415096282955,
      "epoch": 0.8087914008564188,
      "grad_norm": 2.453125,
      "learning_rate": 0.000494145236111113,
      "loss": 6.0477,
      "mean_token_accuracy": 0.1435216970741749,
      "num_tokens": 18753929.0,
      "step": 9255
    },
    {
      "entropy": 6.268417692184448,
      "epoch": 0.8092283492091235,
      "grad_norm": 2.25,
      "learning_rate": 0.0004941381716888409,
      "loss": 6.0357,
      "mean_token_accuracy": 0.14020807147026063,
      "num_tokens": 18764103.0,
      "step": 9260
    },
    {
      "entropy": 6.180013179779053,
      "epoch": 0.8096652975618281,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004941311030633838,
      "loss": 6.0622,
      "mean_token_accuracy": 0.13822172060608864,
      "num_tokens": 18773756.0,
      "step": 9265
    },
    {
      "entropy": 6.218051147460938,
      "epoch": 0.810102245914533,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004941240302348773,
      "loss": 5.9285,
      "mean_token_accuracy": 0.15631069988012314,
      "num_tokens": 18783778.0,
      "step": 9270
    },
    {
      "entropy": 6.228174448013306,
      "epoch": 0.8105391942672376,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004941169532034571,
      "loss": 6.0769,
      "mean_token_accuracy": 0.14282265678048134,
      "num_tokens": 18794664.0,
      "step": 9275
    },
    {
      "entropy": 6.157448863983154,
      "epoch": 0.8109761426199423,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004941098719692589,
      "loss": 6.015,
      "mean_token_accuracy": 0.14017972275614737,
      "num_tokens": 18803773.0,
      "step": 9280
    },
    {
      "entropy": 6.194372272491455,
      "epoch": 0.811413090972647,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004941027865324186,
      "loss": 6.0872,
      "mean_token_accuracy": 0.13763938322663308,
      "num_tokens": 18814419.0,
      "step": 9285
    },
    {
      "entropy": 6.257919836044311,
      "epoch": 0.8118500393253517,
      "grad_norm": 2.828125,
      "learning_rate": 0.000494095696893072,
      "loss": 6.0297,
      "mean_token_accuracy": 0.13992701172828675,
      "num_tokens": 18825011.0,
      "step": 9290
    },
    {
      "entropy": 6.247425651550293,
      "epoch": 0.8122869876780564,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004940886030513551,
      "loss": 5.9624,
      "mean_token_accuracy": 0.14393307715654374,
      "num_tokens": 18835003.0,
      "step": 9295
    },
    {
      "entropy": 6.23016505241394,
      "epoch": 0.8127239360307612,
      "grad_norm": 3.125,
      "learning_rate": 0.0004940815050074041,
      "loss": 6.0432,
      "mean_token_accuracy": 0.1394846647977829,
      "num_tokens": 18845072.0,
      "step": 9300
    },
    {
      "entropy": 6.238551139831543,
      "epoch": 0.8131608843834659,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004940744027613549,
      "loss": 6.0018,
      "mean_token_accuracy": 0.14630599319934845,
      "num_tokens": 18854583.0,
      "step": 9305
    },
    {
      "entropy": 6.161306428909302,
      "epoch": 0.8135978327361706,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004940672963133439,
      "loss": 6.07,
      "mean_token_accuracy": 0.14326853901147843,
      "num_tokens": 18865460.0,
      "step": 9310
    },
    {
      "entropy": 6.258480882644653,
      "epoch": 0.8140347810888753,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004940601856635074,
      "loss": 6.0263,
      "mean_token_accuracy": 0.14527803659439087,
      "num_tokens": 18875568.0,
      "step": 9315
    },
    {
      "entropy": 6.162207841873169,
      "epoch": 0.81447172944158,
      "grad_norm": 2.5,
      "learning_rate": 0.0004940530708119818,
      "loss": 5.9447,
      "mean_token_accuracy": 0.1445034012198448,
      "num_tokens": 18884894.0,
      "step": 9320
    },
    {
      "entropy": 6.137970972061157,
      "epoch": 0.8149086777942847,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004940459517589034,
      "loss": 6.0279,
      "mean_token_accuracy": 0.1465800143778324,
      "num_tokens": 18895353.0,
      "step": 9325
    },
    {
      "entropy": 6.2135459899902346,
      "epoch": 0.8153456261469895,
      "grad_norm": 3.25,
      "learning_rate": 0.000494038828504409,
      "loss": 6.0399,
      "mean_token_accuracy": 0.1466113120317459,
      "num_tokens": 18905474.0,
      "step": 9330
    },
    {
      "entropy": 6.271397686004638,
      "epoch": 0.8157825744996942,
      "grad_norm": 5.1875,
      "learning_rate": 0.0004940317010486351,
      "loss": 6.0089,
      "mean_token_accuracy": 0.1458955377340317,
      "num_tokens": 18914797.0,
      "step": 9335
    },
    {
      "entropy": 6.237304162979126,
      "epoch": 0.8162195228523988,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004940245693917183,
      "loss": 6.0694,
      "mean_token_accuracy": 0.13753879815340042,
      "num_tokens": 18924642.0,
      "step": 9340
    },
    {
      "entropy": 6.144796895980835,
      "epoch": 0.8166564712051035,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004940174335337955,
      "loss": 5.9325,
      "mean_token_accuracy": 0.1523873269557953,
      "num_tokens": 18934026.0,
      "step": 9345
    },
    {
      "entropy": 6.215007543563843,
      "epoch": 0.8170934195578082,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004940102934750037,
      "loss": 5.971,
      "mean_token_accuracy": 0.1510114438831806,
      "num_tokens": 18943262.0,
      "step": 9350
    },
    {
      "entropy": 6.272810459136963,
      "epoch": 0.8175303679105129,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004940031492154795,
      "loss": 5.9926,
      "mean_token_accuracy": 0.13940275311470032,
      "num_tokens": 18953082.0,
      "step": 9355
    },
    {
      "entropy": 6.2077549457550045,
      "epoch": 0.8179673162632177,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004939960007553603,
      "loss": 6.0333,
      "mean_token_accuracy": 0.14470342099666594,
      "num_tokens": 18962408.0,
      "step": 9360
    },
    {
      "entropy": 6.201269245147705,
      "epoch": 0.8184042646159224,
      "grad_norm": 2.171875,
      "learning_rate": 0.000493988848094783,
      "loss": 6.0797,
      "mean_token_accuracy": 0.1364889159798622,
      "num_tokens": 18972918.0,
      "step": 9365
    },
    {
      "entropy": 6.190881776809692,
      "epoch": 0.8188412129686271,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004939816912338848,
      "loss": 6.0251,
      "mean_token_accuracy": 0.14143316373229026,
      "num_tokens": 18983152.0,
      "step": 9370
    },
    {
      "entropy": 6.224991512298584,
      "epoch": 0.8192781613213318,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004939745301728031,
      "loss": 6.0413,
      "mean_token_accuracy": 0.14381223320960998,
      "num_tokens": 18993613.0,
      "step": 9375
    },
    {
      "entropy": 6.210654783248901,
      "epoch": 0.8197151096740365,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004939673649116752,
      "loss": 5.958,
      "mean_token_accuracy": 0.14649604260921478,
      "num_tokens": 19004079.0,
      "step": 9380
    },
    {
      "entropy": 6.203503894805908,
      "epoch": 0.8201520580267413,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004939601954506383,
      "loss": 6.12,
      "mean_token_accuracy": 0.1403985969722271,
      "num_tokens": 19013994.0,
      "step": 9385
    },
    {
      "entropy": 6.2642467498779295,
      "epoch": 0.820589006379446,
      "grad_norm": 2.375,
      "learning_rate": 0.0004939530217898303,
      "loss": 6.0843,
      "mean_token_accuracy": 0.1376613236963749,
      "num_tokens": 19024674.0,
      "step": 9390
    },
    {
      "entropy": 6.233977746963501,
      "epoch": 0.8210259547321507,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004939458439293886,
      "loss": 6.0544,
      "mean_token_accuracy": 0.13939779102802277,
      "num_tokens": 19035333.0,
      "step": 9395
    },
    {
      "entropy": 6.3187164783477785,
      "epoch": 0.8214629030848554,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004939386618694508,
      "loss": 6.0907,
      "mean_token_accuracy": 0.13914038836956025,
      "num_tokens": 19046279.0,
      "step": 9400
    },
    {
      "entropy": 6.2559949398040775,
      "epoch": 0.8218998514375601,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004939314756101547,
      "loss": 6.0455,
      "mean_token_accuracy": 0.13177771270275115,
      "num_tokens": 19056506.0,
      "step": 9405
    },
    {
      "entropy": 6.215059089660644,
      "epoch": 0.8223367997902647,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004939242851516384,
      "loss": 6.0352,
      "mean_token_accuracy": 0.14404356777667998,
      "num_tokens": 19067217.0,
      "step": 9410
    },
    {
      "entropy": 6.203421640396118,
      "epoch": 0.8227737481429696,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004939170904940395,
      "loss": 6.0983,
      "mean_token_accuracy": 0.13966502249240875,
      "num_tokens": 19079028.0,
      "step": 9415
    },
    {
      "entropy": 6.177074670791626,
      "epoch": 0.8232106964956742,
      "grad_norm": 2.171875,
      "learning_rate": 0.000493909891637496,
      "loss": 6.0491,
      "mean_token_accuracy": 0.14438274502754211,
      "num_tokens": 19088236.0,
      "step": 9420
    },
    {
      "entropy": 6.2156205654144285,
      "epoch": 0.8236476448483789,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004939026885821461,
      "loss": 6.0242,
      "mean_token_accuracy": 0.15105926394462585,
      "num_tokens": 19099598.0,
      "step": 9425
    },
    {
      "entropy": 6.225408506393433,
      "epoch": 0.8240845932010836,
      "grad_norm": 2.3125,
      "learning_rate": 0.000493895481328128,
      "loss": 6.0781,
      "mean_token_accuracy": 0.14559734389185905,
      "num_tokens": 19110347.0,
      "step": 9430
    },
    {
      "entropy": 6.331104803085327,
      "epoch": 0.8245215415537883,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004938882698755798,
      "loss": 6.0484,
      "mean_token_accuracy": 0.14259000048041343,
      "num_tokens": 19120643.0,
      "step": 9435
    },
    {
      "entropy": 6.1864618301391605,
      "epoch": 0.824958489906493,
      "grad_norm": 3.390625,
      "learning_rate": 0.00049388105422464,
      "loss": 6.0175,
      "mean_token_accuracy": 0.14517145678400994,
      "num_tokens": 19130209.0,
      "step": 9440
    },
    {
      "entropy": 6.147251749038697,
      "epoch": 0.8253954382591978,
      "grad_norm": 2.125,
      "learning_rate": 0.0004938738343754467,
      "loss": 5.9695,
      "mean_token_accuracy": 0.1517564684152603,
      "num_tokens": 19139196.0,
      "step": 9445
    },
    {
      "entropy": 6.221241903305054,
      "epoch": 0.8258323866119025,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004938666103281389,
      "loss": 6.0091,
      "mean_token_accuracy": 0.14541823118925096,
      "num_tokens": 19149618.0,
      "step": 9450
    },
    {
      "entropy": 6.258189678192139,
      "epoch": 0.8262693349646072,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004938593820828545,
      "loss": 6.0046,
      "mean_token_accuracy": 0.13990169763565063,
      "num_tokens": 19159875.0,
      "step": 9455
    },
    {
      "entropy": 6.221347332000732,
      "epoch": 0.8267062833173119,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004938521496397326,
      "loss": 6.0374,
      "mean_token_accuracy": 0.14358133673667908,
      "num_tokens": 19170403.0,
      "step": 9460
    },
    {
      "entropy": 6.250455045700074,
      "epoch": 0.8271432316700166,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004938449129989117,
      "loss": 6.0021,
      "mean_token_accuracy": 0.15078073218464852,
      "num_tokens": 19181187.0,
      "step": 9465
    },
    {
      "entropy": 6.222884702682495,
      "epoch": 0.8275801800227213,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004938376721605308,
      "loss": 6.0492,
      "mean_token_accuracy": 0.1423380881547928,
      "num_tokens": 19191328.0,
      "step": 9470
    },
    {
      "entropy": 6.144410610198975,
      "epoch": 0.8280171283754261,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004938304271247286,
      "loss": 6.0199,
      "mean_token_accuracy": 0.14642967134714127,
      "num_tokens": 19202041.0,
      "step": 9475
    },
    {
      "entropy": 6.253562498092651,
      "epoch": 0.8284540767281308,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004938231778916442,
      "loss": 6.0017,
      "mean_token_accuracy": 0.14358637407422065,
      "num_tokens": 19212016.0,
      "step": 9480
    },
    {
      "entropy": 6.250165224075317,
      "epoch": 0.8288910250808355,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004938159244614167,
      "loss": 6.0103,
      "mean_token_accuracy": 0.14502643197774887,
      "num_tokens": 19222111.0,
      "step": 9485
    },
    {
      "entropy": 6.282345104217529,
      "epoch": 0.8293279734335401,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004938086668341849,
      "loss": 6.0019,
      "mean_token_accuracy": 0.1438644751906395,
      "num_tokens": 19231865.0,
      "step": 9490
    },
    {
      "entropy": 6.206636810302735,
      "epoch": 0.8297649217862448,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004938014050100884,
      "loss": 5.9792,
      "mean_token_accuracy": 0.1493251107633114,
      "num_tokens": 19242993.0,
      "step": 9495
    },
    {
      "entropy": 6.232339715957641,
      "epoch": 0.8302018701389495,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004937941389892662,
      "loss": 6.0414,
      "mean_token_accuracy": 0.14360441714525224,
      "num_tokens": 19252776.0,
      "step": 9500
    },
    {
      "entropy": 6.186702013015747,
      "epoch": 0.8306388184916543,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004937868687718577,
      "loss": 5.9806,
      "mean_token_accuracy": 0.1452575981616974,
      "num_tokens": 19263022.0,
      "step": 9505
    },
    {
      "entropy": 6.231074666976928,
      "epoch": 0.831075766844359,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004937795943580026,
      "loss": 6.0257,
      "mean_token_accuracy": 0.14902030676603317,
      "num_tokens": 19271859.0,
      "step": 9510
    },
    {
      "entropy": 6.201964712142944,
      "epoch": 0.8315127151970637,
      "grad_norm": 2.453125,
      "learning_rate": 0.00049377231574784,
      "loss": 5.9893,
      "mean_token_accuracy": 0.14588339105248452,
      "num_tokens": 19282771.0,
      "step": 9515
    },
    {
      "entropy": 6.124718809127808,
      "epoch": 0.8319496635497684,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004937650329415099,
      "loss": 6.0148,
      "mean_token_accuracy": 0.13867617696523665,
      "num_tokens": 19292985.0,
      "step": 9520
    },
    {
      "entropy": 6.162050247192383,
      "epoch": 0.8323866119024731,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004937577459391519,
      "loss": 6.0447,
      "mean_token_accuracy": 0.14593885689973832,
      "num_tokens": 19303409.0,
      "step": 9525
    },
    {
      "entropy": 6.232680320739746,
      "epoch": 0.8328235602551778,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004937504547409056,
      "loss": 6.0148,
      "mean_token_accuracy": 0.1468958795070648,
      "num_tokens": 19312632.0,
      "step": 9530
    },
    {
      "entropy": 6.1946220874786375,
      "epoch": 0.8332605086078826,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004937431593469109,
      "loss": 5.9211,
      "mean_token_accuracy": 0.1472806453704834,
      "num_tokens": 19322759.0,
      "step": 9535
    },
    {
      "entropy": 6.196166229248047,
      "epoch": 0.8336974569605873,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004937358597573079,
      "loss": 5.953,
      "mean_token_accuracy": 0.15301361829042434,
      "num_tokens": 19333307.0,
      "step": 9540
    },
    {
      "entropy": 6.186828327178955,
      "epoch": 0.834134405313292,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004937285559722364,
      "loss": 6.055,
      "mean_token_accuracy": 0.14334636330604553,
      "num_tokens": 19342528.0,
      "step": 9545
    },
    {
      "entropy": 6.26044807434082,
      "epoch": 0.8345713536659967,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004937212479918367,
      "loss": 6.0122,
      "mean_token_accuracy": 0.14196290001273154,
      "num_tokens": 19352640.0,
      "step": 9550
    },
    {
      "entropy": 6.267553424835205,
      "epoch": 0.8350083020187014,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004937139358162488,
      "loss": 6.0214,
      "mean_token_accuracy": 0.14564134776592255,
      "num_tokens": 19363285.0,
      "step": 9555
    },
    {
      "entropy": 6.223315048217773,
      "epoch": 0.835445250371406,
      "grad_norm": 2.4375,
      "learning_rate": 0.000493706619445613,
      "loss": 6.0149,
      "mean_token_accuracy": 0.1506032481789589,
      "num_tokens": 19373141.0,
      "step": 9560
    },
    {
      "entropy": 6.200173187255859,
      "epoch": 0.8358821987241108,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004936992988800697,
      "loss": 6.0553,
      "mean_token_accuracy": 0.14346984699368476,
      "num_tokens": 19383091.0,
      "step": 9565
    },
    {
      "entropy": 6.221242427825928,
      "epoch": 0.8363191470768155,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004936919741197591,
      "loss": 5.9847,
      "mean_token_accuracy": 0.14986307099461554,
      "num_tokens": 19393163.0,
      "step": 9570
    },
    {
      "entropy": 6.148035335540771,
      "epoch": 0.8367560954295202,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004936846451648219,
      "loss": 5.9673,
      "mean_token_accuracy": 0.15079953372478486,
      "num_tokens": 19403475.0,
      "step": 9575
    },
    {
      "entropy": 6.1440918922424315,
      "epoch": 0.8371930437822249,
      "grad_norm": 2.125,
      "learning_rate": 0.0004936773120153987,
      "loss": 5.9805,
      "mean_token_accuracy": 0.14774260222911834,
      "num_tokens": 19413302.0,
      "step": 9580
    },
    {
      "entropy": 6.21998872756958,
      "epoch": 0.8376299921349296,
      "grad_norm": 2.390625,
      "learning_rate": 0.00049366997467163,
      "loss": 5.9797,
      "mean_token_accuracy": 0.14915448427200317,
      "num_tokens": 19423473.0,
      "step": 9585
    },
    {
      "entropy": 6.251493215560913,
      "epoch": 0.8380669404876344,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004936626331336567,
      "loss": 6.0773,
      "mean_token_accuracy": 0.1349373109638691,
      "num_tokens": 19433289.0,
      "step": 9590
    },
    {
      "entropy": 6.26294207572937,
      "epoch": 0.8385038888403391,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004936552874016194,
      "loss": 6.0684,
      "mean_token_accuracy": 0.14192093834280967,
      "num_tokens": 19442188.0,
      "step": 9595
    },
    {
      "entropy": 6.2142900943756105,
      "epoch": 0.8389408371930438,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004936479374756592,
      "loss": 6.0664,
      "mean_token_accuracy": 0.14174751862883567,
      "num_tokens": 19452971.0,
      "step": 9600
    },
    {
      "entropy": 6.275484418869018,
      "epoch": 0.8393777855457485,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004936405833559169,
      "loss": 6.0651,
      "mean_token_accuracy": 0.1378170482814312,
      "num_tokens": 19463263.0,
      "step": 9605
    },
    {
      "entropy": 6.3095964908599855,
      "epoch": 0.8398147338984532,
      "grad_norm": 7.28125,
      "learning_rate": 0.0004936332250425337,
      "loss": 6.0571,
      "mean_token_accuracy": 0.13926204591989516,
      "num_tokens": 19473234.0,
      "step": 9610
    },
    {
      "entropy": 6.218443155288696,
      "epoch": 0.8402516822511579,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004936258625356507,
      "loss": 6.0417,
      "mean_token_accuracy": 0.140377489477396,
      "num_tokens": 19483714.0,
      "step": 9615
    },
    {
      "entropy": 6.155925655364991,
      "epoch": 0.8406886306038627,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004936184958354091,
      "loss": 5.9497,
      "mean_token_accuracy": 0.15539525002241134,
      "num_tokens": 19493128.0,
      "step": 9620
    },
    {
      "entropy": 6.25476565361023,
      "epoch": 0.8411255789565674,
      "grad_norm": 5.8125,
      "learning_rate": 0.00049361112494195,
      "loss": 6.0713,
      "mean_token_accuracy": 0.13749251365661622,
      "num_tokens": 19503401.0,
      "step": 9625
    },
    {
      "entropy": 6.254153394699097,
      "epoch": 0.8415625273092721,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004936037498554153,
      "loss": 6.1115,
      "mean_token_accuracy": 0.14547596126794815,
      "num_tokens": 19513050.0,
      "step": 9630
    },
    {
      "entropy": 6.232684087753296,
      "epoch": 0.8419994756619767,
      "grad_norm": 5.90625,
      "learning_rate": 0.000493596370575946,
      "loss": 5.9896,
      "mean_token_accuracy": 0.14650416597723961,
      "num_tokens": 19524534.0,
      "step": 9635
    },
    {
      "entropy": 6.22262659072876,
      "epoch": 0.8424364240146814,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004935889871036837,
      "loss": 5.9747,
      "mean_token_accuracy": 0.14624593481421472,
      "num_tokens": 19534863.0,
      "step": 9640
    },
    {
      "entropy": 6.256587171554566,
      "epoch": 0.8428733723673861,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004935815994387701,
      "loss": 6.0446,
      "mean_token_accuracy": 0.13790009617805482,
      "num_tokens": 19545037.0,
      "step": 9645
    },
    {
      "entropy": 6.228785419464112,
      "epoch": 0.8433103207200909,
      "grad_norm": 2.421875,
      "learning_rate": 0.000493574207581347,
      "loss": 6.038,
      "mean_token_accuracy": 0.1383309245109558,
      "num_tokens": 19555377.0,
      "step": 9650
    },
    {
      "entropy": 6.237904119491577,
      "epoch": 0.8437472690727956,
      "grad_norm": 7.59375,
      "learning_rate": 0.0004935668115315559,
      "loss": 6.054,
      "mean_token_accuracy": 0.14487089663743974,
      "num_tokens": 19565874.0,
      "step": 9655
    },
    {
      "entropy": 6.134960556030274,
      "epoch": 0.8441842174255003,
      "grad_norm": 6.0625,
      "learning_rate": 0.000493559411289539,
      "loss": 5.8676,
      "mean_token_accuracy": 0.16119038313627243,
      "num_tokens": 19575002.0,
      "step": 9660
    },
    {
      "entropy": 6.1698222160339355,
      "epoch": 0.844621165778205,
      "grad_norm": 2.34375,
      "learning_rate": 0.000493552006855438,
      "loss": 6.0118,
      "mean_token_accuracy": 0.14253657013177873,
      "num_tokens": 19585863.0,
      "step": 9665
    },
    {
      "entropy": 6.21726803779602,
      "epoch": 0.8450581141309097,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004935445982293949,
      "loss": 5.9706,
      "mean_token_accuracy": 0.14346363991498948,
      "num_tokens": 19596787.0,
      "step": 9670
    },
    {
      "entropy": 6.225195360183716,
      "epoch": 0.8454950624836144,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004935371854115519,
      "loss": 6.071,
      "mean_token_accuracy": 0.1384153299033642,
      "num_tokens": 19607114.0,
      "step": 9675
    },
    {
      "entropy": 6.18015570640564,
      "epoch": 0.8459320108363192,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004935297684020513,
      "loss": 5.9584,
      "mean_token_accuracy": 0.15262675136327744,
      "num_tokens": 19617334.0,
      "step": 9680
    },
    {
      "entropy": 6.248830986022949,
      "epoch": 0.8463689591890239,
      "grad_norm": 2.25,
      "learning_rate": 0.0004935223472010352,
      "loss": 6.0347,
      "mean_token_accuracy": 0.1399633213877678,
      "num_tokens": 19626844.0,
      "step": 9685
    },
    {
      "entropy": 6.182112073898315,
      "epoch": 0.8468059075417286,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004935149218086459,
      "loss": 6.0293,
      "mean_token_accuracy": 0.14446952790021897,
      "num_tokens": 19637897.0,
      "step": 9690
    },
    {
      "entropy": 6.254135274887085,
      "epoch": 0.8472428558944333,
      "grad_norm": 1.9765625,
      "learning_rate": 0.000493507492225026,
      "loss": 6.1051,
      "mean_token_accuracy": 0.14066995829343795,
      "num_tokens": 19649161.0,
      "step": 9695
    },
    {
      "entropy": 6.2464367866516115,
      "epoch": 0.847679804247138,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004935000584503178,
      "loss": 6.0189,
      "mean_token_accuracy": 0.14391484782099723,
      "num_tokens": 19659328.0,
      "step": 9700
    },
    {
      "entropy": 6.223616981506348,
      "epoch": 0.8481167525998426,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004934926204846641,
      "loss": 6.1117,
      "mean_token_accuracy": 0.13933897837996484,
      "num_tokens": 19669196.0,
      "step": 9705
    },
    {
      "entropy": 6.22830605506897,
      "epoch": 0.8485537009525475,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004934851783282073,
      "loss": 5.9864,
      "mean_token_accuracy": 0.14183563366532326,
      "num_tokens": 19680019.0,
      "step": 9710
    },
    {
      "entropy": 6.22342882156372,
      "epoch": 0.8489906493052521,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004934777319810904,
      "loss": 6.0892,
      "mean_token_accuracy": 0.14166997671127318,
      "num_tokens": 19689391.0,
      "step": 9715
    },
    {
      "entropy": 6.262296199798584,
      "epoch": 0.8494275976579568,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004934702814434561,
      "loss": 6.0843,
      "mean_token_accuracy": 0.13788991197943687,
      "num_tokens": 19699683.0,
      "step": 9720
    },
    {
      "entropy": 6.2552093982696535,
      "epoch": 0.8498645460106615,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004934628267154474,
      "loss": 6.1293,
      "mean_token_accuracy": 0.13570168763399124,
      "num_tokens": 19709844.0,
      "step": 9725
    },
    {
      "entropy": 6.20659909248352,
      "epoch": 0.8503014943633662,
      "grad_norm": 3.109375,
      "learning_rate": 0.000493455367797207,
      "loss": 6.0774,
      "mean_token_accuracy": 0.13346961587667466,
      "num_tokens": 19719926.0,
      "step": 9730
    },
    {
      "entropy": 6.262333250045776,
      "epoch": 0.8507384427160709,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004934479046888784,
      "loss": 6.0088,
      "mean_token_accuracy": 0.1410963289439678,
      "num_tokens": 19730548.0,
      "step": 9735
    },
    {
      "entropy": 6.256011199951172,
      "epoch": 0.8511753910687757,
      "grad_norm": 1.5,
      "learning_rate": 0.0004934404373906045,
      "loss": 6.0637,
      "mean_token_accuracy": 0.14503197818994523,
      "num_tokens": 19740471.0,
      "step": 9740
    },
    {
      "entropy": 6.264379692077637,
      "epoch": 0.8516123394214804,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004934329659025283,
      "loss": 5.9762,
      "mean_token_accuracy": 0.14722604900598527,
      "num_tokens": 19749395.0,
      "step": 9745
    },
    {
      "entropy": 6.226193284988403,
      "epoch": 0.8520492877741851,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004934254902247936,
      "loss": 6.0616,
      "mean_token_accuracy": 0.1382533073425293,
      "num_tokens": 19759203.0,
      "step": 9750
    },
    {
      "entropy": 6.239722824096679,
      "epoch": 0.8524862361268898,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004934180103575436,
      "loss": 5.961,
      "mean_token_accuracy": 0.15295470356941224,
      "num_tokens": 19768977.0,
      "step": 9755
    },
    {
      "entropy": 6.192998743057251,
      "epoch": 0.8529231844795945,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004934105263009216,
      "loss": 5.9377,
      "mean_token_accuracy": 0.15048784539103507,
      "num_tokens": 19778569.0,
      "step": 9760
    },
    {
      "entropy": 6.207692241668701,
      "epoch": 0.8533601328322992,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004934030380550713,
      "loss": 6.0533,
      "mean_token_accuracy": 0.14660714119672774,
      "num_tokens": 19787939.0,
      "step": 9765
    },
    {
      "entropy": 6.21990385055542,
      "epoch": 0.853797081185004,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004933955456201362,
      "loss": 6.0355,
      "mean_token_accuracy": 0.14408317506313323,
      "num_tokens": 19798407.0,
      "step": 9770
    },
    {
      "entropy": 6.271046352386475,
      "epoch": 0.8542340295377087,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004933880489962602,
      "loss": 6.0545,
      "mean_token_accuracy": 0.1449744649231434,
      "num_tokens": 19808495.0,
      "step": 9775
    },
    {
      "entropy": 6.225653457641601,
      "epoch": 0.8546709778904134,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004933805481835868,
      "loss": 5.9924,
      "mean_token_accuracy": 0.14210559278726578,
      "num_tokens": 19818709.0,
      "step": 9780
    },
    {
      "entropy": 6.251949834823608,
      "epoch": 0.855107926243118,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004933730431822603,
      "loss": 6.0571,
      "mean_token_accuracy": 0.14727669656276704,
      "num_tokens": 19828865.0,
      "step": 9785
    },
    {
      "entropy": 6.201934719085694,
      "epoch": 0.8555448745958227,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004933655339924243,
      "loss": 5.9821,
      "mean_token_accuracy": 0.1444874882698059,
      "num_tokens": 19838757.0,
      "step": 9790
    },
    {
      "entropy": 6.1520508289337155,
      "epoch": 0.8559818229485275,
      "grad_norm": 2.25,
      "learning_rate": 0.000493358020614223,
      "loss": 6.0247,
      "mean_token_accuracy": 0.144384902715683,
      "num_tokens": 19848967.0,
      "step": 9795
    },
    {
      "entropy": 6.252551889419555,
      "epoch": 0.8564187713012322,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004933505030478003,
      "loss": 6.0357,
      "mean_token_accuracy": 0.1498258039355278,
      "num_tokens": 19859487.0,
      "step": 9800
    },
    {
      "entropy": 6.249651098251343,
      "epoch": 0.8568557196539369,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004933429812933005,
      "loss": 6.1027,
      "mean_token_accuracy": 0.14054232686758042,
      "num_tokens": 19870170.0,
      "step": 9805
    },
    {
      "entropy": 6.206701469421387,
      "epoch": 0.8572926680066416,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004933354553508681,
      "loss": 5.9842,
      "mean_token_accuracy": 0.14744419753551483,
      "num_tokens": 19881031.0,
      "step": 9810
    },
    {
      "entropy": 6.246293687820435,
      "epoch": 0.8577296163593463,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004933279252206472,
      "loss": 6.0379,
      "mean_token_accuracy": 0.14254553094506264,
      "num_tokens": 19891082.0,
      "step": 9815
    },
    {
      "entropy": 6.248610782623291,
      "epoch": 0.858166564712051,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004933203909027823,
      "loss": 6.0212,
      "mean_token_accuracy": 0.13794225379824637,
      "num_tokens": 19900626.0,
      "step": 9820
    },
    {
      "entropy": 6.167263460159302,
      "epoch": 0.8586035130647558,
      "grad_norm": 3.75,
      "learning_rate": 0.0004933128523974179,
      "loss": 6.0139,
      "mean_token_accuracy": 0.15036819577217103,
      "num_tokens": 19910660.0,
      "step": 9825
    },
    {
      "entropy": 6.169804334640503,
      "epoch": 0.8590404614174605,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004933053097046986,
      "loss": 5.9964,
      "mean_token_accuracy": 0.1429131418466568,
      "num_tokens": 19920476.0,
      "step": 9830
    },
    {
      "entropy": 6.323930168151856,
      "epoch": 0.8594774097701652,
      "grad_norm": 1.8125,
      "learning_rate": 0.000493297762824769,
      "loss": 6.0961,
      "mean_token_accuracy": 0.1377591110765934,
      "num_tokens": 19931151.0,
      "step": 9835
    },
    {
      "entropy": 6.28590202331543,
      "epoch": 0.8599143581228699,
      "grad_norm": 1.8203125,
      "learning_rate": 0.000493290211757774,
      "loss": 5.9989,
      "mean_token_accuracy": 0.13884489238262177,
      "num_tokens": 19942469.0,
      "step": 9840
    },
    {
      "entropy": 6.139708995819092,
      "epoch": 0.8603513064755746,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004932826565038584,
      "loss": 6.0772,
      "mean_token_accuracy": 0.1426509328186512,
      "num_tokens": 19953055.0,
      "step": 9845
    },
    {
      "entropy": 6.0895037174224855,
      "epoch": 0.8607882548282793,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004932750970631669,
      "loss": 6.031,
      "mean_token_accuracy": 0.14375355914235116,
      "num_tokens": 19963944.0,
      "step": 9850
    },
    {
      "entropy": 6.2525804996490475,
      "epoch": 0.8612252031809841,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004932675334358448,
      "loss": 5.9634,
      "mean_token_accuracy": 0.15010778605937958,
      "num_tokens": 19973913.0,
      "step": 9855
    },
    {
      "entropy": 6.233993339538574,
      "epoch": 0.8616621515336887,
      "grad_norm": 2.375,
      "learning_rate": 0.000493259965622037,
      "loss": 6.0215,
      "mean_token_accuracy": 0.1483843058347702,
      "num_tokens": 19984130.0,
      "step": 9860
    },
    {
      "entropy": 6.161510467529297,
      "epoch": 0.8620990998863934,
      "grad_norm": 2.0,
      "learning_rate": 0.0004932523936218887,
      "loss": 5.9519,
      "mean_token_accuracy": 0.14606448411941528,
      "num_tokens": 19993840.0,
      "step": 9865
    },
    {
      "entropy": 6.261183786392212,
      "epoch": 0.8625360482390981,
      "grad_norm": 10.375,
      "learning_rate": 0.000493244817435545,
      "loss": 6.0413,
      "mean_token_accuracy": 0.137623693048954,
      "num_tokens": 20004228.0,
      "step": 9870
    },
    {
      "entropy": 6.2544409275054935,
      "epoch": 0.8629729965918028,
      "grad_norm": 1.75,
      "learning_rate": 0.0004932372370631515,
      "loss": 6.0329,
      "mean_token_accuracy": 0.1442359358072281,
      "num_tokens": 20012620.0,
      "step": 9875
    },
    {
      "entropy": 6.1343893051147464,
      "epoch": 0.8634099449445075,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004932296525048535,
      "loss": 6.1106,
      "mean_token_accuracy": 0.1413384832441807,
      "num_tokens": 20022892.0,
      "step": 9880
    },
    {
      "entropy": 6.248836421966553,
      "epoch": 0.8638468932972123,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004932220637607962,
      "loss": 5.9863,
      "mean_token_accuracy": 0.1510588899254799,
      "num_tokens": 20032788.0,
      "step": 9885
    },
    {
      "entropy": 6.199680423736572,
      "epoch": 0.864283841649917,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004932144708311256,
      "loss": 6.0213,
      "mean_token_accuracy": 0.14951885342597962,
      "num_tokens": 20043176.0,
      "step": 9890
    },
    {
      "entropy": 6.144923543930053,
      "epoch": 0.8647207900026217,
      "grad_norm": 1.9921875,
      "learning_rate": 0.000493206873715987,
      "loss": 6.0261,
      "mean_token_accuracy": 0.1444010391831398,
      "num_tokens": 20052298.0,
      "step": 9895
    },
    {
      "entropy": 6.224962568283081,
      "epoch": 0.8651577383553264,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004931992724155262,
      "loss": 6.008,
      "mean_token_accuracy": 0.143085827678442,
      "num_tokens": 20063163.0,
      "step": 9900
    },
    {
      "entropy": 6.309250116348267,
      "epoch": 0.8655946867080311,
      "grad_norm": 1.90625,
      "learning_rate": 0.000493191666929889,
      "loss": 6.0523,
      "mean_token_accuracy": 0.1428293138742447,
      "num_tokens": 20073644.0,
      "step": 9905
    },
    {
      "entropy": 6.244245195388794,
      "epoch": 0.8660316350607358,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004931840572592214,
      "loss": 5.9877,
      "mean_token_accuracy": 0.14725518748164176,
      "num_tokens": 20084457.0,
      "step": 9910
    },
    {
      "entropy": 6.120822286605835,
      "epoch": 0.8664685834134406,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004931764434036692,
      "loss": 5.9392,
      "mean_token_accuracy": 0.14560910165309907,
      "num_tokens": 20094153.0,
      "step": 9915
    },
    {
      "entropy": 6.150817060470581,
      "epoch": 0.8669055317661453,
      "grad_norm": 2.375,
      "learning_rate": 0.0004931688253633784,
      "loss": 6.0929,
      "mean_token_accuracy": 0.14018966257572174,
      "num_tokens": 20104031.0,
      "step": 9920
    },
    {
      "entropy": 6.226406574249268,
      "epoch": 0.86734248011885,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004931612031384953,
      "loss": 6.0256,
      "mean_token_accuracy": 0.14454981982707976,
      "num_tokens": 20114431.0,
      "step": 9925
    },
    {
      "entropy": 6.150810289382934,
      "epoch": 0.8677794284715546,
      "grad_norm": 1.703125,
      "learning_rate": 0.000493153576729166,
      "loss": 5.9288,
      "mean_token_accuracy": 0.1526732012629509,
      "num_tokens": 20124161.0,
      "step": 9930
    },
    {
      "entropy": 6.182177495956421,
      "epoch": 0.8682163768242593,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004931459461355368,
      "loss": 5.9706,
      "mean_token_accuracy": 0.14706914722919465,
      "num_tokens": 20134227.0,
      "step": 9935
    },
    {
      "entropy": 6.215608930587768,
      "epoch": 0.868653325176964,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004931383113577541,
      "loss": 5.9952,
      "mean_token_accuracy": 0.14292289167642594,
      "num_tokens": 20143912.0,
      "step": 9940
    },
    {
      "entropy": 6.139667749404907,
      "epoch": 0.8690902735296688,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004931306723959641,
      "loss": 5.9164,
      "mean_token_accuracy": 0.14590430483222008,
      "num_tokens": 20153858.0,
      "step": 9945
    },
    {
      "entropy": 6.267096376419067,
      "epoch": 0.8695272218823735,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004931230292503137,
      "loss": 6.0553,
      "mean_token_accuracy": 0.13928072899580002,
      "num_tokens": 20164309.0,
      "step": 9950
    },
    {
      "entropy": 6.214211893081665,
      "epoch": 0.8699641702350782,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004931153819209492,
      "loss": 6.0088,
      "mean_token_accuracy": 0.1478944942355156,
      "num_tokens": 20173949.0,
      "step": 9955
    },
    {
      "entropy": 6.223191499710083,
      "epoch": 0.8704011185877829,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004931077304080175,
      "loss": 5.9925,
      "mean_token_accuracy": 0.14299240857362747,
      "num_tokens": 20184206.0,
      "step": 9960
    },
    {
      "entropy": 6.253048753738403,
      "epoch": 0.8708380669404876,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004931000747116651,
      "loss": 6.1353,
      "mean_token_accuracy": 0.13737881183624268,
      "num_tokens": 20194430.0,
      "step": 9965
    },
    {
      "entropy": 6.2516481399536135,
      "epoch": 0.8712750152931923,
      "grad_norm": 6.8125,
      "learning_rate": 0.000493092414832039,
      "loss": 6.0182,
      "mean_token_accuracy": 0.14453519731760026,
      "num_tokens": 20205004.0,
      "step": 9970
    },
    {
      "entropy": 6.221957063674926,
      "epoch": 0.8717119636458971,
      "grad_norm": 2.25,
      "learning_rate": 0.0004930847507692862,
      "loss": 5.9628,
      "mean_token_accuracy": 0.14764900654554367,
      "num_tokens": 20214698.0,
      "step": 9975
    },
    {
      "entropy": 6.22438383102417,
      "epoch": 0.8721489119986018,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004930770825235536,
      "loss": 6.0505,
      "mean_token_accuracy": 0.14198303520679473,
      "num_tokens": 20226534.0,
      "step": 9980
    },
    {
      "entropy": 6.255250358581543,
      "epoch": 0.8725858603513065,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004930694100949882,
      "loss": 6.02,
      "mean_token_accuracy": 0.14456572830677034,
      "num_tokens": 20237414.0,
      "step": 9985
    },
    {
      "entropy": 6.179359436035156,
      "epoch": 0.8730228087040112,
      "grad_norm": 4.625,
      "learning_rate": 0.0004930617334837373,
      "loss": 5.8932,
      "mean_token_accuracy": 0.14946675151586533,
      "num_tokens": 20246948.0,
      "step": 9990
    },
    {
      "entropy": 6.218437337875367,
      "epoch": 0.8734597570567159,
      "grad_norm": 3.40625,
      "learning_rate": 0.000493054052689948,
      "loss": 6.0853,
      "mean_token_accuracy": 0.14085782393813134,
      "num_tokens": 20257281.0,
      "step": 9995
    },
    {
      "entropy": 6.253528738021851,
      "epoch": 0.8738967054094207,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004930463677137679,
      "loss": 6.044,
      "mean_token_accuracy": 0.14160678461194037,
      "num_tokens": 20269041.0,
      "step": 10000
    },
    {
      "entropy": 6.248606109619141,
      "epoch": 0.8743336537621254,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004930386785553441,
      "loss": 6.0845,
      "mean_token_accuracy": 0.13608923330903053,
      "num_tokens": 20279688.0,
      "step": 10005
    },
    {
      "entropy": 6.159049129486084,
      "epoch": 0.87477060211483,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004930309852148242,
      "loss": 5.998,
      "mean_token_accuracy": 0.1461903363466263,
      "num_tokens": 20288913.0,
      "step": 10010
    },
    {
      "entropy": 6.21247410774231,
      "epoch": 0.8752075504675347,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004930232876923558,
      "loss": 6.0464,
      "mean_token_accuracy": 0.14632102996110916,
      "num_tokens": 20299009.0,
      "step": 10015
    },
    {
      "entropy": 6.23969349861145,
      "epoch": 0.8756444988202394,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004930155859880865,
      "loss": 6.052,
      "mean_token_accuracy": 0.13175813630223274,
      "num_tokens": 20309478.0,
      "step": 10020
    },
    {
      "entropy": 6.240252161026001,
      "epoch": 0.8760814471729441,
      "grad_norm": 2.140625,
      "learning_rate": 0.000493007880102164,
      "loss": 6.0511,
      "mean_token_accuracy": 0.1454978756606579,
      "num_tokens": 20319451.0,
      "step": 10025
    },
    {
      "entropy": 6.232294225692749,
      "epoch": 0.8765183955256489,
      "grad_norm": 1.9765625,
      "learning_rate": 0.000493000170034736,
      "loss": 5.9978,
      "mean_token_accuracy": 0.1465945139527321,
      "num_tokens": 20328497.0,
      "step": 10030
    },
    {
      "entropy": 6.127724742889404,
      "epoch": 0.8769553438783536,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004929924557859507,
      "loss": 5.8955,
      "mean_token_accuracy": 0.15832905918359758,
      "num_tokens": 20338584.0,
      "step": 10035
    },
    {
      "entropy": 6.156444787979126,
      "epoch": 0.8773922922310583,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004929847373559557,
      "loss": 5.9949,
      "mean_token_accuracy": 0.145755597949028,
      "num_tokens": 20348808.0,
      "step": 10040
    },
    {
      "entropy": 6.194709253311157,
      "epoch": 0.877829240583763,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004929770147448992,
      "loss": 5.9821,
      "mean_token_accuracy": 0.15028578341007232,
      "num_tokens": 20358301.0,
      "step": 10045
    },
    {
      "entropy": 6.242437314987183,
      "epoch": 0.8782661889364677,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004929692879529295,
      "loss": 5.981,
      "mean_token_accuracy": 0.14376169592142105,
      "num_tokens": 20369032.0,
      "step": 10050
    },
    {
      "entropy": 6.257175779342651,
      "epoch": 0.8787031372891724,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004929615569801946,
      "loss": 6.0148,
      "mean_token_accuracy": 0.14054974392056466,
      "num_tokens": 20379101.0,
      "step": 10055
    },
    {
      "entropy": 6.165028858184814,
      "epoch": 0.8791400856418772,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004929538218268428,
      "loss": 5.9747,
      "mean_token_accuracy": 0.14643891006708146,
      "num_tokens": 20388864.0,
      "step": 10060
    },
    {
      "entropy": 6.246957397460937,
      "epoch": 0.8795770339945819,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004929460824930226,
      "loss": 6.1249,
      "mean_token_accuracy": 0.13385197520256042,
      "num_tokens": 20398402.0,
      "step": 10065
    },
    {
      "entropy": 6.1573272228240965,
      "epoch": 0.8800139823472866,
      "grad_norm": 6.6875,
      "learning_rate": 0.0004929383389788821,
      "loss": 6.0106,
      "mean_token_accuracy": 0.14462155923247338,
      "num_tokens": 20408209.0,
      "step": 10070
    },
    {
      "entropy": 6.194298601150512,
      "epoch": 0.8804509306999913,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004929305912845703,
      "loss": 6.0292,
      "mean_token_accuracy": 0.14462321996688843,
      "num_tokens": 20418117.0,
      "step": 10075
    },
    {
      "entropy": 6.2855524063110355,
      "epoch": 0.8808878790526959,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004929228394102355,
      "loss": 6.0533,
      "mean_token_accuracy": 0.14158154129981995,
      "num_tokens": 20428347.0,
      "step": 10080
    },
    {
      "entropy": 6.237124443054199,
      "epoch": 0.8813248274054006,
      "grad_norm": 2.125,
      "learning_rate": 0.0004929150833560263,
      "loss": 5.983,
      "mean_token_accuracy": 0.1424064613878727,
      "num_tokens": 20437804.0,
      "step": 10085
    },
    {
      "entropy": 6.226949882507324,
      "epoch": 0.8817617757581054,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004929073231220917,
      "loss": 6.0289,
      "mean_token_accuracy": 0.13792968541383743,
      "num_tokens": 20447600.0,
      "step": 10090
    },
    {
      "entropy": 6.274075603485107,
      "epoch": 0.8821987241108101,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004928995587085806,
      "loss": 6.049,
      "mean_token_accuracy": 0.14034995287656785,
      "num_tokens": 20458023.0,
      "step": 10095
    },
    {
      "entropy": 6.213431310653687,
      "epoch": 0.8826356724635148,
      "grad_norm": 2.5,
      "learning_rate": 0.0004928917901156416,
      "loss": 5.9754,
      "mean_token_accuracy": 0.1537353515625,
      "num_tokens": 20468195.0,
      "step": 10100
    },
    {
      "entropy": 6.16763710975647,
      "epoch": 0.8830726208162195,
      "grad_norm": 2.296875,
      "learning_rate": 0.000492884017343424,
      "loss": 6.108,
      "mean_token_accuracy": 0.1395591288805008,
      "num_tokens": 20479250.0,
      "step": 10105
    },
    {
      "entropy": 6.253421497344971,
      "epoch": 0.8835095691689242,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004928762403920767,
      "loss": 5.9994,
      "mean_token_accuracy": 0.14397843629121782,
      "num_tokens": 20489587.0,
      "step": 10110
    },
    {
      "entropy": 6.253739929199218,
      "epoch": 0.8839465175216289,
      "grad_norm": 4.875,
      "learning_rate": 0.000492868459261749,
      "loss": 6.0401,
      "mean_token_accuracy": 0.1451330915093422,
      "num_tokens": 20499788.0,
      "step": 10115
    },
    {
      "entropy": 6.213185977935791,
      "epoch": 0.8843834658743337,
      "grad_norm": 4.90625,
      "learning_rate": 0.00049286067395259,
      "loss": 6.0309,
      "mean_token_accuracy": 0.14450110420584678,
      "num_tokens": 20509716.0,
      "step": 10120
    },
    {
      "entropy": 6.175362205505371,
      "epoch": 0.8848204142270384,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004928528844647493,
      "loss": 5.9808,
      "mean_token_accuracy": 0.14676104336977006,
      "num_tokens": 20519020.0,
      "step": 10125
    },
    {
      "entropy": 6.152131748199463,
      "epoch": 0.8852573625797431,
      "grad_norm": 2.03125,
      "learning_rate": 0.000492845090798376,
      "loss": 6.0376,
      "mean_token_accuracy": 0.14153722375631334,
      "num_tokens": 20528997.0,
      "step": 10130
    },
    {
      "entropy": 6.235318994522094,
      "epoch": 0.8856943109324478,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004928372929536197,
      "loss": 6.0194,
      "mean_token_accuracy": 0.14579871967434882,
      "num_tokens": 20538599.0,
      "step": 10135
    },
    {
      "entropy": 6.305400848388672,
      "epoch": 0.8861312592851525,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004928294909306301,
      "loss": 5.9692,
      "mean_token_accuracy": 0.14578816890716553,
      "num_tokens": 20547664.0,
      "step": 10140
    },
    {
      "entropy": 6.2061316013336185,
      "epoch": 0.8865682076378572,
      "grad_norm": 5.28125,
      "learning_rate": 0.0004928216847295565,
      "loss": 6.0832,
      "mean_token_accuracy": 0.14088777601718902,
      "num_tokens": 20557860.0,
      "step": 10145
    },
    {
      "entropy": 6.1760553359985355,
      "epoch": 0.887005155990562,
      "grad_norm": 1.953125,
      "learning_rate": 0.000492813874350549,
      "loss": 6.0209,
      "mean_token_accuracy": 0.14260354042053222,
      "num_tokens": 20569003.0,
      "step": 10150
    },
    {
      "entropy": 6.226287126541138,
      "epoch": 0.8874421043432666,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004928060597937573,
      "loss": 6.0092,
      "mean_token_accuracy": 0.14746515527367593,
      "num_tokens": 20578402.0,
      "step": 10155
    },
    {
      "entropy": 6.222630262374878,
      "epoch": 0.8878790526959713,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004927982410593312,
      "loss": 5.9212,
      "mean_token_accuracy": 0.14485912397503853,
      "num_tokens": 20587874.0,
      "step": 10160
    },
    {
      "entropy": 6.263999652862549,
      "epoch": 0.888316001048676,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004927904181474207,
      "loss": 5.9561,
      "mean_token_accuracy": 0.1467773661017418,
      "num_tokens": 20597270.0,
      "step": 10165
    },
    {
      "entropy": 6.237065267562866,
      "epoch": 0.8887529494013807,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004927825910581758,
      "loss": 6.0517,
      "mean_token_accuracy": 0.1435149371623993,
      "num_tokens": 20607898.0,
      "step": 10170
    },
    {
      "entropy": 6.189601135253906,
      "epoch": 0.8891898977540854,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004927747597917467,
      "loss": 5.9449,
      "mean_token_accuracy": 0.15103435814380645,
      "num_tokens": 20617300.0,
      "step": 10175
    },
    {
      "entropy": 6.236210060119629,
      "epoch": 0.8896268461067902,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004927669243482836,
      "loss": 6.0624,
      "mean_token_accuracy": 0.14391943737864493,
      "num_tokens": 20626582.0,
      "step": 10180
    },
    {
      "entropy": 6.154780483245849,
      "epoch": 0.8900637944594949,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004927590847279369,
      "loss": 5.9976,
      "mean_token_accuracy": 0.14631350710988045,
      "num_tokens": 20635802.0,
      "step": 10185
    },
    {
      "entropy": 6.191054105758667,
      "epoch": 0.8905007428121996,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004927512409308568,
      "loss": 6.0265,
      "mean_token_accuracy": 0.1469324618577957,
      "num_tokens": 20646057.0,
      "step": 10190
    },
    {
      "entropy": 6.282738447189331,
      "epoch": 0.8909376911649043,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004927433929571937,
      "loss": 6.0941,
      "mean_token_accuracy": 0.13414522036910057,
      "num_tokens": 20656511.0,
      "step": 10195
    },
    {
      "entropy": 6.295567369461059,
      "epoch": 0.891374639517609,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004927355408070982,
      "loss": 6.0067,
      "mean_token_accuracy": 0.1483195722103119,
      "num_tokens": 20666479.0,
      "step": 10200
    },
    {
      "entropy": 6.2385200500488285,
      "epoch": 0.8918115878703138,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004927276844807211,
      "loss": 5.9471,
      "mean_token_accuracy": 0.1456519827246666,
      "num_tokens": 20676830.0,
      "step": 10205
    },
    {
      "entropy": 6.122660112380982,
      "epoch": 0.8922485362230185,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004927198239782129,
      "loss": 5.9858,
      "mean_token_accuracy": 0.14475072771310807,
      "num_tokens": 20687316.0,
      "step": 10210
    },
    {
      "entropy": 6.191984367370606,
      "epoch": 0.8926854845757232,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004927119592997242,
      "loss": 6.0709,
      "mean_token_accuracy": 0.14245957657694816,
      "num_tokens": 20697901.0,
      "step": 10215
    },
    {
      "entropy": 6.2576416492462155,
      "epoch": 0.8931224329284279,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004927040904454062,
      "loss": 5.9502,
      "mean_token_accuracy": 0.15248084366321563,
      "num_tokens": 20707468.0,
      "step": 10220
    },
    {
      "entropy": 6.195013809204101,
      "epoch": 0.8935593812811325,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004926962174154097,
      "loss": 6.0015,
      "mean_token_accuracy": 0.14423225894570352,
      "num_tokens": 20717098.0,
      "step": 10225
    },
    {
      "entropy": 6.233042621612549,
      "epoch": 0.8939963296338372,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004926883402098857,
      "loss": 6.1132,
      "mean_token_accuracy": 0.13845815137028694,
      "num_tokens": 20727217.0,
      "step": 10230
    },
    {
      "entropy": 6.220581674575806,
      "epoch": 0.894433277986542,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004926804588289852,
      "loss": 6.0701,
      "mean_token_accuracy": 0.14081439152359962,
      "num_tokens": 20737397.0,
      "step": 10235
    },
    {
      "entropy": 6.156305837631225,
      "epoch": 0.8948702263392467,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004926725732728595,
      "loss": 5.8797,
      "mean_token_accuracy": 0.15089936554431915,
      "num_tokens": 20746712.0,
      "step": 10240
    },
    {
      "entropy": 6.206583786010742,
      "epoch": 0.8953071746919514,
      "grad_norm": 2.75,
      "learning_rate": 0.0004926646835416597,
      "loss": 6.0237,
      "mean_token_accuracy": 0.14401394873857498,
      "num_tokens": 20757161.0,
      "step": 10245
    },
    {
      "entropy": 6.2091819763183596,
      "epoch": 0.8957441230446561,
      "grad_norm": 2.25,
      "learning_rate": 0.0004926567896355372,
      "loss": 5.9504,
      "mean_token_accuracy": 0.15837636291980745,
      "num_tokens": 20766478.0,
      "step": 10250
    },
    {
      "entropy": 6.1995820045471195,
      "epoch": 0.8961810713973608,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004926488915546435,
      "loss": 6.0022,
      "mean_token_accuracy": 0.1455075055360794,
      "num_tokens": 20776317.0,
      "step": 10255
    },
    {
      "entropy": 6.219730091094971,
      "epoch": 0.8966180197500655,
      "grad_norm": 2.828125,
      "learning_rate": 0.00049264098929913,
      "loss": 5.9286,
      "mean_token_accuracy": 0.14915867894887924,
      "num_tokens": 20786534.0,
      "step": 10260
    },
    {
      "entropy": 6.155456066131592,
      "epoch": 0.8970549681027703,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004926330828691482,
      "loss": 5.9654,
      "mean_token_accuracy": 0.14729184210300444,
      "num_tokens": 20796791.0,
      "step": 10265
    },
    {
      "entropy": 6.213042879104615,
      "epoch": 0.897491916455475,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004926251722648499,
      "loss": 5.9681,
      "mean_token_accuracy": 0.14904457479715347,
      "num_tokens": 20806834.0,
      "step": 10270
    },
    {
      "entropy": 6.143362188339234,
      "epoch": 0.8979288648081797,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004926172574863868,
      "loss": 6.0001,
      "mean_token_accuracy": 0.1442363977432251,
      "num_tokens": 20816664.0,
      "step": 10275
    },
    {
      "entropy": 6.176703071594238,
      "epoch": 0.8983658131608844,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004926093385339105,
      "loss": 6.0923,
      "mean_token_accuracy": 0.14293062537908555,
      "num_tokens": 20826936.0,
      "step": 10280
    },
    {
      "entropy": 6.235605621337891,
      "epoch": 0.8988027615135891,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004926014154075732,
      "loss": 5.9701,
      "mean_token_accuracy": 0.1459277093410492,
      "num_tokens": 20836732.0,
      "step": 10285
    },
    {
      "entropy": 6.298902893066407,
      "epoch": 0.8992397098662938,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004925934881075268,
      "loss": 6.0548,
      "mean_token_accuracy": 0.13570440113544463,
      "num_tokens": 20848093.0,
      "step": 10290
    },
    {
      "entropy": 6.265059566497802,
      "epoch": 0.8996766582189986,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004925855566339232,
      "loss": 5.9708,
      "mean_token_accuracy": 0.14378736466169356,
      "num_tokens": 20858678.0,
      "step": 10295
    },
    {
      "entropy": 6.2076538562774655,
      "epoch": 0.9001136065717033,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004925776209869146,
      "loss": 5.9674,
      "mean_token_accuracy": 0.1465497247874737,
      "num_tokens": 20869319.0,
      "step": 10300
    },
    {
      "entropy": 6.204934883117676,
      "epoch": 0.9005505549244079,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004925696811666532,
      "loss": 5.9654,
      "mean_token_accuracy": 0.1483704574406147,
      "num_tokens": 20879475.0,
      "step": 10305
    },
    {
      "entropy": 6.1841202735900875,
      "epoch": 0.9009875032771126,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004925617371732914,
      "loss": 5.9606,
      "mean_token_accuracy": 0.15277539938688278,
      "num_tokens": 20889479.0,
      "step": 10310
    },
    {
      "entropy": 6.2324748039245605,
      "epoch": 0.9014244516298173,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004925537890069815,
      "loss": 6.0491,
      "mean_token_accuracy": 0.14187959283590318,
      "num_tokens": 20899864.0,
      "step": 10315
    },
    {
      "entropy": 6.162887859344482,
      "epoch": 0.901861399982522,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004925458366678758,
      "loss": 6.0477,
      "mean_token_accuracy": 0.14296629130840302,
      "num_tokens": 20909006.0,
      "step": 10320
    },
    {
      "entropy": 6.2473695278167725,
      "epoch": 0.9022983483352268,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004925378801561269,
      "loss": 5.9568,
      "mean_token_accuracy": 0.14621812105178833,
      "num_tokens": 20917998.0,
      "step": 10325
    },
    {
      "entropy": 6.145819425582886,
      "epoch": 0.9027352966879315,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004925299194718876,
      "loss": 5.9807,
      "mean_token_accuracy": 0.14791434556245803,
      "num_tokens": 20928407.0,
      "step": 10330
    },
    {
      "entropy": 6.103126573562622,
      "epoch": 0.9031722450406362,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004925219546153105,
      "loss": 6.0123,
      "mean_token_accuracy": 0.14059873446822166,
      "num_tokens": 20938016.0,
      "step": 10335
    },
    {
      "entropy": 6.274331665039062,
      "epoch": 0.9036091933933409,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004925139855865483,
      "loss": 6.0656,
      "mean_token_accuracy": 0.1379294253885746,
      "num_tokens": 20948578.0,
      "step": 10340
    },
    {
      "entropy": 6.305205392837524,
      "epoch": 0.9040461417460456,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004925060123857539,
      "loss": 5.9313,
      "mean_token_accuracy": 0.14447855800390244,
      "num_tokens": 20958193.0,
      "step": 10345
    },
    {
      "entropy": 6.234714555740356,
      "epoch": 0.9044830900987503,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004924980350130802,
      "loss": 5.9883,
      "mean_token_accuracy": 0.15243747979402542,
      "num_tokens": 20968092.0,
      "step": 10350
    },
    {
      "entropy": 6.135723304748535,
      "epoch": 0.9049200384514551,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004924900534686801,
      "loss": 6.0206,
      "mean_token_accuracy": 0.1399074487388134,
      "num_tokens": 20978303.0,
      "step": 10355
    },
    {
      "entropy": 6.191471338272095,
      "epoch": 0.9053569868041598,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004924820677527069,
      "loss": 6.0115,
      "mean_token_accuracy": 0.14377782195806504,
      "num_tokens": 20988331.0,
      "step": 10360
    },
    {
      "entropy": 6.150998830795288,
      "epoch": 0.9057939351568645,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004924740778653137,
      "loss": 5.936,
      "mean_token_accuracy": 0.14535974115133285,
      "num_tokens": 20997785.0,
      "step": 10365
    },
    {
      "entropy": 6.274278736114502,
      "epoch": 0.9062308835095692,
      "grad_norm": 2.875,
      "learning_rate": 0.0004924660838066538,
      "loss": 6.0422,
      "mean_token_accuracy": 0.1434810273349285,
      "num_tokens": 21008026.0,
      "step": 10370
    },
    {
      "entropy": 6.2038064956665036,
      "epoch": 0.9066678318622738,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004924580855768803,
      "loss": 6.0152,
      "mean_token_accuracy": 0.14854384064674378,
      "num_tokens": 21017993.0,
      "step": 10375
    },
    {
      "entropy": 6.159410095214843,
      "epoch": 0.9071047802149785,
      "grad_norm": 2.5,
      "learning_rate": 0.0004924500831761468,
      "loss": 5.967,
      "mean_token_accuracy": 0.14842465221881868,
      "num_tokens": 21028149.0,
      "step": 10380
    },
    {
      "entropy": 6.285621452331543,
      "epoch": 0.9075417285676833,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004924420766046069,
      "loss": 6.1227,
      "mean_token_accuracy": 0.13110206723213197,
      "num_tokens": 21039229.0,
      "step": 10385
    },
    {
      "entropy": 6.298278141021728,
      "epoch": 0.907978676920388,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004924340658624139,
      "loss": 6.0476,
      "mean_token_accuracy": 0.14461030662059784,
      "num_tokens": 21049358.0,
      "step": 10390
    },
    {
      "entropy": 6.239208078384399,
      "epoch": 0.9084156252730927,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004924260509497217,
      "loss": 6.0859,
      "mean_token_accuracy": 0.14283088743686675,
      "num_tokens": 21060385.0,
      "step": 10395
    },
    {
      "entropy": 6.213239908218384,
      "epoch": 0.9088525736257974,
      "grad_norm": 5.0625,
      "learning_rate": 0.0004924180318666838,
      "loss": 6.0752,
      "mean_token_accuracy": 0.14117654636502266,
      "num_tokens": 21070423.0,
      "step": 10400
    },
    {
      "entropy": 6.394452857971191,
      "epoch": 0.9092895219785021,
      "grad_norm": 3.125,
      "learning_rate": 0.0004924100086134541,
      "loss": 6.0393,
      "mean_token_accuracy": 0.13764989152550697,
      "num_tokens": 21080505.0,
      "step": 10405
    },
    {
      "entropy": 6.265415191650391,
      "epoch": 0.9097264703312069,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004924019811901866,
      "loss": 6.0346,
      "mean_token_accuracy": 0.13928126394748688,
      "num_tokens": 21090875.0,
      "step": 10410
    },
    {
      "entropy": 6.21007890701294,
      "epoch": 0.9101634186839116,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004923939495970352,
      "loss": 5.9771,
      "mean_token_accuracy": 0.14366873800754548,
      "num_tokens": 21101427.0,
      "step": 10415
    },
    {
      "entropy": 6.279690456390381,
      "epoch": 0.9106003670366163,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004923859138341539,
      "loss": 6.0991,
      "mean_token_accuracy": 0.14020335972309111,
      "num_tokens": 21112691.0,
      "step": 10420
    },
    {
      "entropy": 6.297071075439453,
      "epoch": 0.911037315389321,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004923778739016969,
      "loss": 6.0907,
      "mean_token_accuracy": 0.1366792932152748,
      "num_tokens": 21123932.0,
      "step": 10425
    },
    {
      "entropy": 6.215231418609619,
      "epoch": 0.9114742637420257,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004923698297998183,
      "loss": 6.0952,
      "mean_token_accuracy": 0.13873008415102958,
      "num_tokens": 21135380.0,
      "step": 10430
    },
    {
      "entropy": 6.207882308959961,
      "epoch": 0.9119112120947304,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004923617815286725,
      "loss": 5.99,
      "mean_token_accuracy": 0.14756772220134734,
      "num_tokens": 21144951.0,
      "step": 10435
    },
    {
      "entropy": 6.2428864479064945,
      "epoch": 0.9123481604474352,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004923537290884138,
      "loss": 5.8974,
      "mean_token_accuracy": 0.146316459774971,
      "num_tokens": 21155919.0,
      "step": 10440
    },
    {
      "entropy": 6.137046241760254,
      "epoch": 0.9127851088001399,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004923456724791967,
      "loss": 5.9503,
      "mean_token_accuracy": 0.14737611785531043,
      "num_tokens": 21166186.0,
      "step": 10445
    },
    {
      "entropy": 6.138951063156128,
      "epoch": 0.9132220571528445,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004923376117011758,
      "loss": 5.9316,
      "mean_token_accuracy": 0.1520561769604683,
      "num_tokens": 21176043.0,
      "step": 10450
    },
    {
      "entropy": 6.14461030960083,
      "epoch": 0.9136590055055492,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004923295467545055,
      "loss": 5.9324,
      "mean_token_accuracy": 0.15320966392755508,
      "num_tokens": 21185677.0,
      "step": 10455
    },
    {
      "entropy": 6.212748861312866,
      "epoch": 0.9140959538582539,
      "grad_norm": 3.375,
      "learning_rate": 0.0004923214776393407,
      "loss": 6.0528,
      "mean_token_accuracy": 0.1364740625023842,
      "num_tokens": 21195977.0,
      "step": 10460
    },
    {
      "entropy": 6.260949802398682,
      "epoch": 0.9145329022109586,
      "grad_norm": 5.03125,
      "learning_rate": 0.000492313404355836,
      "loss": 6.0192,
      "mean_token_accuracy": 0.1471311330795288,
      "num_tokens": 21205454.0,
      "step": 10465
    },
    {
      "entropy": 6.20062403678894,
      "epoch": 0.9149698505636634,
      "grad_norm": 3.75,
      "learning_rate": 0.0004923053269041464,
      "loss": 5.9291,
      "mean_token_accuracy": 0.15106230676174165,
      "num_tokens": 21215700.0,
      "step": 10470
    },
    {
      "entropy": 6.171433305740356,
      "epoch": 0.9154067989163681,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004922972452844267,
      "loss": 6.0425,
      "mean_token_accuracy": 0.14153960198163987,
      "num_tokens": 21226544.0,
      "step": 10475
    },
    {
      "entropy": 6.244486856460571,
      "epoch": 0.9158437472690728,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004922891594968321,
      "loss": 6.0621,
      "mean_token_accuracy": 0.14115132316946982,
      "num_tokens": 21236792.0,
      "step": 10480
    },
    {
      "entropy": 6.262734365463257,
      "epoch": 0.9162806956217775,
      "grad_norm": 4.9375,
      "learning_rate": 0.0004922810695415175,
      "loss": 5.9803,
      "mean_token_accuracy": 0.14429373666644096,
      "num_tokens": 21247039.0,
      "step": 10485
    },
    {
      "entropy": 6.187035703659058,
      "epoch": 0.9167176439744822,
      "grad_norm": 3.875,
      "learning_rate": 0.0004922729754186381,
      "loss": 5.9895,
      "mean_token_accuracy": 0.14789291173219682,
      "num_tokens": 21257008.0,
      "step": 10490
    },
    {
      "entropy": 6.202964210510254,
      "epoch": 0.9171545923271869,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004922648771283492,
      "loss": 6.1188,
      "mean_token_accuracy": 0.13898467347025872,
      "num_tokens": 21267638.0,
      "step": 10495
    },
    {
      "entropy": 6.265435743331909,
      "epoch": 0.9175915406798917,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004922567746708062,
      "loss": 5.9908,
      "mean_token_accuracy": 0.14649815410375594,
      "num_tokens": 21277327.0,
      "step": 10500
    },
    {
      "entropy": 6.216078615188598,
      "epoch": 0.9180284890325964,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004922486680461642,
      "loss": 6.0713,
      "mean_token_accuracy": 0.14119886606931686,
      "num_tokens": 21287579.0,
      "step": 10505
    },
    {
      "entropy": 6.215159559249878,
      "epoch": 0.9184654373853011,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004922405572545792,
      "loss": 6.003,
      "mean_token_accuracy": 0.13807881399989128,
      "num_tokens": 21297888.0,
      "step": 10510
    },
    {
      "entropy": 6.323606300354004,
      "epoch": 0.9189023857380058,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004922324422962063,
      "loss": 6.0497,
      "mean_token_accuracy": 0.14395225718617438,
      "num_tokens": 21308506.0,
      "step": 10515
    },
    {
      "entropy": 6.228168869018555,
      "epoch": 0.9193393340907104,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004922243231712013,
      "loss": 6.0943,
      "mean_token_accuracy": 0.13690708130598067,
      "num_tokens": 21319252.0,
      "step": 10520
    },
    {
      "entropy": 6.194595241546631,
      "epoch": 0.9197762824434151,
      "grad_norm": 3.015625,
      "learning_rate": 0.00049221619987972,
      "loss": 6.0507,
      "mean_token_accuracy": 0.15066403448581694,
      "num_tokens": 21328829.0,
      "step": 10525
    },
    {
      "entropy": 6.192099523544312,
      "epoch": 0.9202132307961199,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004922080724219182,
      "loss": 5.8998,
      "mean_token_accuracy": 0.1474005937576294,
      "num_tokens": 21338524.0,
      "step": 10530
    },
    {
      "entropy": 6.214118003845215,
      "epoch": 0.9206501791488246,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004921999407979518,
      "loss": 6.0217,
      "mean_token_accuracy": 0.14827122092247008,
      "num_tokens": 21348559.0,
      "step": 10535
    },
    {
      "entropy": 6.310662937164307,
      "epoch": 0.9210871275015293,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004921918050079766,
      "loss": 6.0529,
      "mean_token_accuracy": 0.14093535169959068,
      "num_tokens": 21360763.0,
      "step": 10540
    },
    {
      "entropy": 6.241768074035645,
      "epoch": 0.921524075854234,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004921836650521489,
      "loss": 6.0257,
      "mean_token_accuracy": 0.13818306252360343,
      "num_tokens": 21370048.0,
      "step": 10545
    },
    {
      "entropy": 6.127975034713745,
      "epoch": 0.9219610242069387,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004921755209306247,
      "loss": 6.0247,
      "mean_token_accuracy": 0.14913525879383088,
      "num_tokens": 21379760.0,
      "step": 10550
    },
    {
      "entropy": 6.254342365264892,
      "epoch": 0.9223979725596434,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004921673726435601,
      "loss": 6.0006,
      "mean_token_accuracy": 0.1414869636297226,
      "num_tokens": 21389712.0,
      "step": 10555
    },
    {
      "entropy": 6.301798152923584,
      "epoch": 0.9228349209123482,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004921592201911115,
      "loss": 5.985,
      "mean_token_accuracy": 0.139986489713192,
      "num_tokens": 21399515.0,
      "step": 10560
    },
    {
      "entropy": 6.14472541809082,
      "epoch": 0.9232718692650529,
      "grad_norm": 12.0,
      "learning_rate": 0.0004921510635734353,
      "loss": 5.9671,
      "mean_token_accuracy": 0.14641227424144745,
      "num_tokens": 21409704.0,
      "step": 10565
    },
    {
      "entropy": 6.149060153961182,
      "epoch": 0.9237088176177576,
      "grad_norm": 2.953125,
      "learning_rate": 0.000492142902790688,
      "loss": 5.969,
      "mean_token_accuracy": 0.14714572876691817,
      "num_tokens": 21419182.0,
      "step": 10570
    },
    {
      "entropy": 6.152178955078125,
      "epoch": 0.9241457659704623,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004921347378430259,
      "loss": 5.8465,
      "mean_token_accuracy": 0.14979463219642639,
      "num_tokens": 21429749.0,
      "step": 10575
    },
    {
      "entropy": 6.303376817703247,
      "epoch": 0.924582714323167,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004921265687306059,
      "loss": 6.1263,
      "mean_token_accuracy": 0.133523565530777,
      "num_tokens": 21440514.0,
      "step": 10580
    },
    {
      "entropy": 6.260319566726684,
      "epoch": 0.9250196626758717,
      "grad_norm": 3.375,
      "learning_rate": 0.0004921183954535844,
      "loss": 5.9849,
      "mean_token_accuracy": 0.14747928380966185,
      "num_tokens": 21450647.0,
      "step": 10585
    },
    {
      "entropy": 6.1637708187103275,
      "epoch": 0.9254566110285765,
      "grad_norm": 2.25,
      "learning_rate": 0.0004921102180121184,
      "loss": 6.0183,
      "mean_token_accuracy": 0.14358814507722856,
      "num_tokens": 21460233.0,
      "step": 10590
    },
    {
      "entropy": 6.205360841751099,
      "epoch": 0.9258935593812812,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004921020364063646,
      "loss": 6.0094,
      "mean_token_accuracy": 0.1464311197400093,
      "num_tokens": 21470207.0,
      "step": 10595
    },
    {
      "entropy": 6.226392030715942,
      "epoch": 0.9263305077339858,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004920938506364801,
      "loss": 6.041,
      "mean_token_accuracy": 0.14325212314724922,
      "num_tokens": 21480658.0,
      "step": 10600
    },
    {
      "entropy": 6.157357835769654,
      "epoch": 0.9267674560866905,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004920856607026216,
      "loss": 5.9372,
      "mean_token_accuracy": 0.15598447024822235,
      "num_tokens": 21490566.0,
      "step": 10605
    },
    {
      "entropy": 6.259216451644898,
      "epoch": 0.9272044044393952,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004920774666049465,
      "loss": 6.0181,
      "mean_token_accuracy": 0.14652312323451042,
      "num_tokens": 21500752.0,
      "step": 10610
    },
    {
      "entropy": 6.210867071151734,
      "epoch": 0.9276413527921,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004920692683436118,
      "loss": 6.0122,
      "mean_token_accuracy": 0.13795026540756225,
      "num_tokens": 21509802.0,
      "step": 10615
    },
    {
      "entropy": 6.196670293807983,
      "epoch": 0.9280783011448047,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004920610659187748,
      "loss": 6.0976,
      "mean_token_accuracy": 0.13529762476682664,
      "num_tokens": 21520962.0,
      "step": 10620
    },
    {
      "entropy": 6.2840722560882565,
      "epoch": 0.9285152494975094,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004920528593305929,
      "loss": 6.0552,
      "mean_token_accuracy": 0.1432752065360546,
      "num_tokens": 21530580.0,
      "step": 10625
    },
    {
      "entropy": 6.223748350143433,
      "epoch": 0.9289521978502141,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004920446485792233,
      "loss": 5.9924,
      "mean_token_accuracy": 0.14267162457108498,
      "num_tokens": 21541431.0,
      "step": 10630
    },
    {
      "entropy": 6.166488218307495,
      "epoch": 0.9293891462029188,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004920364336648236,
      "loss": 5.9956,
      "mean_token_accuracy": 0.14596092104911804,
      "num_tokens": 21551012.0,
      "step": 10635
    },
    {
      "entropy": 6.165671348571777,
      "epoch": 0.9298260945556235,
      "grad_norm": 6.78125,
      "learning_rate": 0.0004920282145875514,
      "loss": 5.968,
      "mean_token_accuracy": 0.14599599987268447,
      "num_tokens": 21560676.0,
      "step": 10640
    },
    {
      "entropy": 6.180420780181885,
      "epoch": 0.9302630429083283,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004920199913475644,
      "loss": 5.9498,
      "mean_token_accuracy": 0.15370635986328124,
      "num_tokens": 21570848.0,
      "step": 10645
    },
    {
      "entropy": 6.210285902023315,
      "epoch": 0.930699991261033,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004920117639450202,
      "loss": 6.0171,
      "mean_token_accuracy": 0.14113985002040863,
      "num_tokens": 21582358.0,
      "step": 10650
    },
    {
      "entropy": 6.191497421264648,
      "epoch": 0.9311369396137377,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004920035323800767,
      "loss": 5.9812,
      "mean_token_accuracy": 0.14366309270262717,
      "num_tokens": 21591709.0,
      "step": 10655
    },
    {
      "entropy": 6.220680570602417,
      "epoch": 0.9315738879664424,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004919952966528917,
      "loss": 5.9334,
      "mean_token_accuracy": 0.14706862941384316,
      "num_tokens": 21601092.0,
      "step": 10660
    },
    {
      "entropy": 6.2127635955810545,
      "epoch": 0.932010836319147,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004919870567636233,
      "loss": 5.9613,
      "mean_token_accuracy": 0.14712963551282882,
      "num_tokens": 21611003.0,
      "step": 10665
    },
    {
      "entropy": 6.218704462051392,
      "epoch": 0.9324477846718517,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004919788127124293,
      "loss": 6.0442,
      "mean_token_accuracy": 0.143191160261631,
      "num_tokens": 21621528.0,
      "step": 10670
    },
    {
      "entropy": 6.140312194824219,
      "epoch": 0.9328847330245565,
      "grad_norm": 2.09375,
      "learning_rate": 0.000491970564499468,
      "loss": 5.9407,
      "mean_token_accuracy": 0.15630738213658332,
      "num_tokens": 21631431.0,
      "step": 10675
    },
    {
      "entropy": 6.170972919464111,
      "epoch": 0.9333216813772612,
      "grad_norm": 2.375,
      "learning_rate": 0.0004919623121248976,
      "loss": 5.9649,
      "mean_token_accuracy": 0.14786597937345505,
      "num_tokens": 21640717.0,
      "step": 10680
    },
    {
      "entropy": 6.199743843078613,
      "epoch": 0.9337586297299659,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004919540555888765,
      "loss": 5.9988,
      "mean_token_accuracy": 0.14796182438731192,
      "num_tokens": 21650493.0,
      "step": 10685
    },
    {
      "entropy": 6.18906946182251,
      "epoch": 0.9341955780826706,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004919457948915628,
      "loss": 5.9918,
      "mean_token_accuracy": 0.14778658896684646,
      "num_tokens": 21660062.0,
      "step": 10690
    },
    {
      "entropy": 6.130363082885742,
      "epoch": 0.9346325264353753,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004919375300331151,
      "loss": 5.98,
      "mean_token_accuracy": 0.14802612364292145,
      "num_tokens": 21669364.0,
      "step": 10695
    },
    {
      "entropy": 6.211278104782105,
      "epoch": 0.93506947478808,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004919292610136919,
      "loss": 6.0904,
      "mean_token_accuracy": 0.13631946146488189,
      "num_tokens": 21680793.0,
      "step": 10700
    },
    {
      "entropy": 6.300205183029175,
      "epoch": 0.9355064231407848,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004919209878334518,
      "loss": 6.0556,
      "mean_token_accuracy": 0.14969588071107864,
      "num_tokens": 21690862.0,
      "step": 10705
    },
    {
      "entropy": 6.218013858795166,
      "epoch": 0.9359433714934895,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004919127104925535,
      "loss": 5.9557,
      "mean_token_accuracy": 0.1505468435585499,
      "num_tokens": 21700665.0,
      "step": 10710
    },
    {
      "entropy": 6.21826138496399,
      "epoch": 0.9363803198461942,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004919044289911557,
      "loss": 6.009,
      "mean_token_accuracy": 0.15433460250496864,
      "num_tokens": 21711128.0,
      "step": 10715
    },
    {
      "entropy": 6.298015975952149,
      "epoch": 0.9368172681988989,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004918961433294173,
      "loss": 6.0595,
      "mean_token_accuracy": 0.14322931915521622,
      "num_tokens": 21720233.0,
      "step": 10720
    },
    {
      "entropy": 6.219951391220093,
      "epoch": 0.9372542165516036,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004918878535074973,
      "loss": 6.0113,
      "mean_token_accuracy": 0.14601922184228897,
      "num_tokens": 21730980.0,
      "step": 10725
    },
    {
      "entropy": 6.1353987693786625,
      "epoch": 0.9376911649043083,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004918795595255545,
      "loss": 6.006,
      "mean_token_accuracy": 0.15070429146289827,
      "num_tokens": 21742802.0,
      "step": 10730
    },
    {
      "entropy": 6.202398681640625,
      "epoch": 0.9381281132570131,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004918712613837483,
      "loss": 6.0165,
      "mean_token_accuracy": 0.14104835465550422,
      "num_tokens": 21752743.0,
      "step": 10735
    },
    {
      "entropy": 6.2312483310699465,
      "epoch": 0.9385650616097178,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004918629590822375,
      "loss": 6.0082,
      "mean_token_accuracy": 0.15075039565563203,
      "num_tokens": 21763198.0,
      "step": 10740
    },
    {
      "entropy": 6.286467838287353,
      "epoch": 0.9390020099624224,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004918546526211816,
      "loss": 5.9961,
      "mean_token_accuracy": 0.14858868047595025,
      "num_tokens": 21773030.0,
      "step": 10745
    },
    {
      "entropy": 6.260855102539063,
      "epoch": 0.9394389583151271,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004918463420007398,
      "loss": 5.9461,
      "mean_token_accuracy": 0.15268792808055878,
      "num_tokens": 21783796.0,
      "step": 10750
    },
    {
      "entropy": 6.1994904518127445,
      "epoch": 0.9398759066678318,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004918380272210715,
      "loss": 5.9826,
      "mean_token_accuracy": 0.14605018645524978,
      "num_tokens": 21792504.0,
      "step": 10755
    },
    {
      "entropy": 6.211883544921875,
      "epoch": 0.9403128550205365,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004918297082823362,
      "loss": 6.0604,
      "mean_token_accuracy": 0.14198035895824432,
      "num_tokens": 21802764.0,
      "step": 10760
    },
    {
      "entropy": 6.255032253265381,
      "epoch": 0.9407498033732413,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004918213851846935,
      "loss": 6.0291,
      "mean_token_accuracy": 0.1414106547832489,
      "num_tokens": 21813051.0,
      "step": 10765
    },
    {
      "entropy": 6.171677970886231,
      "epoch": 0.941186751725946,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004918130579283031,
      "loss": 6.0547,
      "mean_token_accuracy": 0.14869959205389022,
      "num_tokens": 21822941.0,
      "step": 10770
    },
    {
      "entropy": 6.220159387588501,
      "epoch": 0.9416237000786507,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004918047265133246,
      "loss": 5.9944,
      "mean_token_accuracy": 0.14000218361616135,
      "num_tokens": 21833225.0,
      "step": 10775
    },
    {
      "entropy": 6.193222761154175,
      "epoch": 0.9420606484313554,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004917963909399178,
      "loss": 5.9935,
      "mean_token_accuracy": 0.1491098329424858,
      "num_tokens": 21842360.0,
      "step": 10780
    },
    {
      "entropy": 6.158725357055664,
      "epoch": 0.9424975967840601,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004917880512082427,
      "loss": 6.0061,
      "mean_token_accuracy": 0.1409321278333664,
      "num_tokens": 21852515.0,
      "step": 10785
    },
    {
      "entropy": 6.232230710983276,
      "epoch": 0.9429345451367648,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004917797073184591,
      "loss": 6.0969,
      "mean_token_accuracy": 0.13883135467767715,
      "num_tokens": 21863040.0,
      "step": 10790
    },
    {
      "entropy": 6.283722257614135,
      "epoch": 0.9433714934894696,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004917713592707271,
      "loss": 6.0239,
      "mean_token_accuracy": 0.14643628969788552,
      "num_tokens": 21872101.0,
      "step": 10795
    },
    {
      "entropy": 6.276474285125732,
      "epoch": 0.9438084418421743,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004917630070652069,
      "loss": 6.012,
      "mean_token_accuracy": 0.1447010673582554,
      "num_tokens": 21882148.0,
      "step": 10800
    },
    {
      "entropy": 6.178737545013428,
      "epoch": 0.944245390194879,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004917546507020587,
      "loss": 5.9793,
      "mean_token_accuracy": 0.1562783569097519,
      "num_tokens": 21891832.0,
      "step": 10805
    },
    {
      "entropy": 6.177520990371704,
      "epoch": 0.9446823385475837,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004917462901814427,
      "loss": 5.9844,
      "mean_token_accuracy": 0.1424844041466713,
      "num_tokens": 21901544.0,
      "step": 10810
    },
    {
      "entropy": 6.253967809677124,
      "epoch": 0.9451192869002883,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004917379255035192,
      "loss": 6.0482,
      "mean_token_accuracy": 0.13904709592461587,
      "num_tokens": 21911690.0,
      "step": 10815
    },
    {
      "entropy": 6.2717604637146,
      "epoch": 0.9455562352529931,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004917295566684487,
      "loss": 6.0767,
      "mean_token_accuracy": 0.14239536672830583,
      "num_tokens": 21922968.0,
      "step": 10820
    },
    {
      "entropy": 6.135378122329712,
      "epoch": 0.9459931836056978,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004917211836763919,
      "loss": 5.9054,
      "mean_token_accuracy": 0.15248963683843614,
      "num_tokens": 21932904.0,
      "step": 10825
    },
    {
      "entropy": 6.117606496810913,
      "epoch": 0.9464301319584025,
      "grad_norm": 2.0,
      "learning_rate": 0.0004917128065275091,
      "loss": 5.9183,
      "mean_token_accuracy": 0.15463434532284737,
      "num_tokens": 21943065.0,
      "step": 10830
    },
    {
      "entropy": 6.218717002868653,
      "epoch": 0.9468670803111072,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004917044252219612,
      "loss": 5.9868,
      "mean_token_accuracy": 0.13802486956119536,
      "num_tokens": 21953448.0,
      "step": 10835
    },
    {
      "entropy": 6.252996778488159,
      "epoch": 0.9473040286638119,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004916960397599088,
      "loss": 5.9783,
      "mean_token_accuracy": 0.14509997367858887,
      "num_tokens": 21963374.0,
      "step": 10840
    },
    {
      "entropy": 6.228418684005737,
      "epoch": 0.9477409770165166,
      "grad_norm": 3.25,
      "learning_rate": 0.0004916876501415129,
      "loss": 6.0081,
      "mean_token_accuracy": 0.1398419111967087,
      "num_tokens": 21973726.0,
      "step": 10845
    },
    {
      "entropy": 6.186339902877807,
      "epoch": 0.9481779253692214,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004916792563669342,
      "loss": 5.9187,
      "mean_token_accuracy": 0.15003404915332794,
      "num_tokens": 21983203.0,
      "step": 10850
    },
    {
      "entropy": 6.186272287368775,
      "epoch": 0.9486148737219261,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004916708584363339,
      "loss": 6.0156,
      "mean_token_accuracy": 0.14147771149873734,
      "num_tokens": 21993435.0,
      "step": 10855
    },
    {
      "entropy": 6.2610009670257565,
      "epoch": 0.9490518220746308,
      "grad_norm": 2.609375,
      "learning_rate": 0.000491662456349873,
      "loss": 6.051,
      "mean_token_accuracy": 0.14023432284593582,
      "num_tokens": 22003230.0,
      "step": 10860
    },
    {
      "entropy": 6.237763166427612,
      "epoch": 0.9494887704273355,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004916540501077127,
      "loss": 5.9938,
      "mean_token_accuracy": 0.14545562863349915,
      "num_tokens": 22012910.0,
      "step": 10865
    },
    {
      "entropy": 6.191306114196777,
      "epoch": 0.9499257187800402,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004916456397100141,
      "loss": 5.9582,
      "mean_token_accuracy": 0.14597126245498657,
      "num_tokens": 22022802.0,
      "step": 10870
    },
    {
      "entropy": 6.189124631881714,
      "epoch": 0.9503626671327449,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004916372251569386,
      "loss": 6.0449,
      "mean_token_accuracy": 0.14138342589139938,
      "num_tokens": 22032669.0,
      "step": 10875
    },
    {
      "entropy": 6.185146379470825,
      "epoch": 0.9507996154854497,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004916288064486477,
      "loss": 5.9525,
      "mean_token_accuracy": 0.1547721117734909,
      "num_tokens": 22042477.0,
      "step": 10880
    },
    {
      "entropy": 6.22675633430481,
      "epoch": 0.9512365638381544,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004916203835853027,
      "loss": 6.0294,
      "mean_token_accuracy": 0.14525218829512596,
      "num_tokens": 22052891.0,
      "step": 10885
    },
    {
      "entropy": 6.172472190856934,
      "epoch": 0.951673512190859,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004916119565670652,
      "loss": 5.9587,
      "mean_token_accuracy": 0.15339114665985107,
      "num_tokens": 22062413.0,
      "step": 10890
    },
    {
      "entropy": 6.226963949203491,
      "epoch": 0.9521104605435637,
      "grad_norm": 3.0625,
      "learning_rate": 0.000491603525394097,
      "loss": 6.0312,
      "mean_token_accuracy": 0.1441036880016327,
      "num_tokens": 22072565.0,
      "step": 10895
    },
    {
      "entropy": 6.263289117813111,
      "epoch": 0.9525474088962684,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004915950900665596,
      "loss": 5.9955,
      "mean_token_accuracy": 0.1423460550606251,
      "num_tokens": 22082910.0,
      "step": 10900
    },
    {
      "entropy": 6.138033008575439,
      "epoch": 0.9529843572489731,
      "grad_norm": 2.5625,
      "learning_rate": 0.000491586650584615,
      "loss": 5.9194,
      "mean_token_accuracy": 0.15537372156977652,
      "num_tokens": 22092636.0,
      "step": 10905
    },
    {
      "entropy": 6.159360361099243,
      "epoch": 0.9534213056016779,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004915782069484248,
      "loss": 5.9898,
      "mean_token_accuracy": 0.14539300203323363,
      "num_tokens": 22102232.0,
      "step": 10910
    },
    {
      "entropy": 6.246322154998779,
      "epoch": 0.9538582539543826,
      "grad_norm": 9.375,
      "learning_rate": 0.0004915697591581511,
      "loss": 6.0666,
      "mean_token_accuracy": 0.14191672429442406,
      "num_tokens": 22113592.0,
      "step": 10915
    },
    {
      "entropy": 6.223143243789673,
      "epoch": 0.9542952023070873,
      "grad_norm": 2.78125,
      "learning_rate": 0.000491561307213956,
      "loss": 6.0029,
      "mean_token_accuracy": 0.1490973487496376,
      "num_tokens": 22123731.0,
      "step": 10920
    },
    {
      "entropy": 6.195102214813232,
      "epoch": 0.954732150659792,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004915528511160016,
      "loss": 6.0466,
      "mean_token_accuracy": 0.13526889234781264,
      "num_tokens": 22133801.0,
      "step": 10925
    },
    {
      "entropy": 6.245194959640503,
      "epoch": 0.9551690990124967,
      "grad_norm": 3.375,
      "learning_rate": 0.0004915443908644501,
      "loss": 6.0112,
      "mean_token_accuracy": 0.14422514140605927,
      "num_tokens": 22143939.0,
      "step": 10930
    },
    {
      "entropy": 6.184619903564453,
      "epoch": 0.9556060473652014,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004915359264594636,
      "loss": 6.0344,
      "mean_token_accuracy": 0.14297499433159827,
      "num_tokens": 22154954.0,
      "step": 10935
    },
    {
      "entropy": 6.171615982055664,
      "epoch": 0.9560429957179062,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004915274579012045,
      "loss": 6.0184,
      "mean_token_accuracy": 0.14542377069592477,
      "num_tokens": 22165225.0,
      "step": 10940
    },
    {
      "entropy": 6.2445436954498295,
      "epoch": 0.9564799440706109,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004915189851898355,
      "loss": 6.0379,
      "mean_token_accuracy": 0.13991244956851007,
      "num_tokens": 22174891.0,
      "step": 10945
    },
    {
      "entropy": 6.175931406021118,
      "epoch": 0.9569168924233156,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004915105083255187,
      "loss": 6.0164,
      "mean_token_accuracy": 0.14690523222088814,
      "num_tokens": 22185255.0,
      "step": 10950
    },
    {
      "entropy": 6.201735305786133,
      "epoch": 0.9573538407760203,
      "grad_norm": 2.8125,
      "learning_rate": 0.000491502027308417,
      "loss": 6.0653,
      "mean_token_accuracy": 0.14451221227645875,
      "num_tokens": 22196124.0,
      "step": 10955
    },
    {
      "entropy": 6.230536365509034,
      "epoch": 0.957790789128725,
      "grad_norm": 2.171875,
      "learning_rate": 0.000491493542138693,
      "loss": 6.0112,
      "mean_token_accuracy": 0.14705559909343718,
      "num_tokens": 22207563.0,
      "step": 10960
    },
    {
      "entropy": 6.2241034507751465,
      "epoch": 0.9582277374814296,
      "grad_norm": 3.25,
      "learning_rate": 0.0004914850528165094,
      "loss": 6.0454,
      "mean_token_accuracy": 0.14533495381474495,
      "num_tokens": 22218527.0,
      "step": 10965
    },
    {
      "entropy": 6.2043381690979,
      "epoch": 0.9586646858341344,
      "grad_norm": 2.609375,
      "learning_rate": 0.000491476559342029,
      "loss": 6.0183,
      "mean_token_accuracy": 0.14308564662933348,
      "num_tokens": 22229538.0,
      "step": 10970
    },
    {
      "entropy": 6.18302173614502,
      "epoch": 0.9591016341868391,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004914680617154148,
      "loss": 5.932,
      "mean_token_accuracy": 0.15026368349790573,
      "num_tokens": 22240047.0,
      "step": 10975
    },
    {
      "entropy": 6.127550363540649,
      "epoch": 0.9595385825395438,
      "grad_norm": 6.25,
      "learning_rate": 0.0004914595599368297,
      "loss": 5.9069,
      "mean_token_accuracy": 0.15337428599596023,
      "num_tokens": 22250440.0,
      "step": 10980
    },
    {
      "entropy": 6.183742475509644,
      "epoch": 0.9599755308922485,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004914510540064369,
      "loss": 5.9875,
      "mean_token_accuracy": 0.1468888506293297,
      "num_tokens": 22259909.0,
      "step": 10985
    },
    {
      "entropy": 6.203589534759521,
      "epoch": 0.9604124792449532,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004914425439243994,
      "loss": 6.0268,
      "mean_token_accuracy": 0.14455460160970687,
      "num_tokens": 22270358.0,
      "step": 10990
    },
    {
      "entropy": 6.184572982788086,
      "epoch": 0.9608494275976579,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004914340296908805,
      "loss": 6.0497,
      "mean_token_accuracy": 0.14026312083005904,
      "num_tokens": 22280591.0,
      "step": 10995
    },
    {
      "entropy": 6.257546615600586,
      "epoch": 0.9612863759503627,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004914255113060435,
      "loss": 6.0256,
      "mean_token_accuracy": 0.14363593012094497,
      "num_tokens": 22291067.0,
      "step": 11000
    },
    {
      "entropy": 6.183000373840332,
      "epoch": 0.9617233243030674,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004914169887700519,
      "loss": 5.9595,
      "mean_token_accuracy": 0.15076697915792464,
      "num_tokens": 22300919.0,
      "step": 11005
    },
    {
      "entropy": 6.209692049026489,
      "epoch": 0.9621602726557721,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004914084620830689,
      "loss": 6.0558,
      "mean_token_accuracy": 0.13879420086741448,
      "num_tokens": 22311562.0,
      "step": 11010
    },
    {
      "entropy": 6.280151510238648,
      "epoch": 0.9625972210084768,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004913999312452583,
      "loss": 6.0154,
      "mean_token_accuracy": 0.14432545006275177,
      "num_tokens": 22322980.0,
      "step": 11015
    },
    {
      "entropy": 6.294468259811401,
      "epoch": 0.9630341693611815,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004913913962567836,
      "loss": 5.9451,
      "mean_token_accuracy": 0.15182111859321595,
      "num_tokens": 22333242.0,
      "step": 11020
    },
    {
      "entropy": 6.259706830978393,
      "epoch": 0.9634711177138863,
      "grad_norm": 2.5,
      "learning_rate": 0.0004913828571178085,
      "loss": 6.0167,
      "mean_token_accuracy": 0.1432610757648945,
      "num_tokens": 22343618.0,
      "step": 11025
    },
    {
      "entropy": 6.175912189483642,
      "epoch": 0.963908066066591,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004913743138284969,
      "loss": 6.0274,
      "mean_token_accuracy": 0.1369355410337448,
      "num_tokens": 22354249.0,
      "step": 11030
    },
    {
      "entropy": 6.2488789558410645,
      "epoch": 0.9643450144192957,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004913657663890124,
      "loss": 5.9867,
      "mean_token_accuracy": 0.14141578301787378,
      "num_tokens": 22363824.0,
      "step": 11035
    },
    {
      "entropy": 6.244002914428711,
      "epoch": 0.9647819627720003,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004913572147995193,
      "loss": 5.9553,
      "mean_token_accuracy": 0.14645989313721658,
      "num_tokens": 22373470.0,
      "step": 11040
    },
    {
      "entropy": 6.158661794662476,
      "epoch": 0.965218911124705,
      "grad_norm": 2.125,
      "learning_rate": 0.0004913486590601814,
      "loss": 5.9427,
      "mean_token_accuracy": 0.14150407910346985,
      "num_tokens": 22382999.0,
      "step": 11045
    },
    {
      "entropy": 6.129118585586548,
      "epoch": 0.9656558594774097,
      "grad_norm": 3.875,
      "learning_rate": 0.0004913400991711629,
      "loss": 5.9972,
      "mean_token_accuracy": 0.15310502126812936,
      "num_tokens": 22393415.0,
      "step": 11050
    },
    {
      "entropy": 6.135149049758911,
      "epoch": 0.9660928078301145,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004913315351326279,
      "loss": 5.9181,
      "mean_token_accuracy": 0.1508888080716133,
      "num_tokens": 22403040.0,
      "step": 11055
    },
    {
      "entropy": 6.231525945663452,
      "epoch": 0.9665297561828192,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004913229669447406,
      "loss": 5.9638,
      "mean_token_accuracy": 0.14777958244085312,
      "num_tokens": 22413434.0,
      "step": 11060
    },
    {
      "entropy": 6.213392639160157,
      "epoch": 0.9669667045355239,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004913143946076656,
      "loss": 6.0661,
      "mean_token_accuracy": 0.14121007025241852,
      "num_tokens": 22424446.0,
      "step": 11065
    },
    {
      "entropy": 6.190732192993164,
      "epoch": 0.9674036528882286,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004913058181215671,
      "loss": 6.0079,
      "mean_token_accuracy": 0.15233338326215745,
      "num_tokens": 22433928.0,
      "step": 11070
    },
    {
      "entropy": 6.224525260925293,
      "epoch": 0.9678406012409333,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004912972374866096,
      "loss": 5.9824,
      "mean_token_accuracy": 0.1454777702689171,
      "num_tokens": 22443778.0,
      "step": 11075
    },
    {
      "entropy": 6.227717351913452,
      "epoch": 0.968277549593638,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004912886527029579,
      "loss": 6.048,
      "mean_token_accuracy": 0.14311131238937377,
      "num_tokens": 22453423.0,
      "step": 11080
    },
    {
      "entropy": 6.248215675354004,
      "epoch": 0.9687144979463428,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004912800637707764,
      "loss": 5.9791,
      "mean_token_accuracy": 0.1495063677430153,
      "num_tokens": 22464067.0,
      "step": 11085
    },
    {
      "entropy": 6.164710330963135,
      "epoch": 0.9691514462990475,
      "grad_norm": 3.15625,
      "learning_rate": 0.00049127147069023,
      "loss": 5.9641,
      "mean_token_accuracy": 0.14813558384776115,
      "num_tokens": 22474332.0,
      "step": 11090
    },
    {
      "entropy": 6.222461557388305,
      "epoch": 0.9695883946517522,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004912628734614834,
      "loss": 6.0174,
      "mean_token_accuracy": 0.1444977566599846,
      "num_tokens": 22485304.0,
      "step": 11095
    },
    {
      "entropy": 6.194020080566406,
      "epoch": 0.9700253430044569,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004912542720847016,
      "loss": 5.9109,
      "mean_token_accuracy": 0.1493035674095154,
      "num_tokens": 22494447.0,
      "step": 11100
    },
    {
      "entropy": 6.250943279266357,
      "epoch": 0.9704622913571616,
      "grad_norm": 2.5,
      "learning_rate": 0.0004912456665600495,
      "loss": 5.9866,
      "mean_token_accuracy": 0.14843482226133348,
      "num_tokens": 22503936.0,
      "step": 11105
    },
    {
      "entropy": 6.224056243896484,
      "epoch": 0.9708992397098662,
      "grad_norm": 2.25,
      "learning_rate": 0.0004912370568876923,
      "loss": 5.9035,
      "mean_token_accuracy": 0.14860760420560837,
      "num_tokens": 22513960.0,
      "step": 11110
    },
    {
      "entropy": 6.187701988220215,
      "epoch": 0.971336188062571,
      "grad_norm": 2.375,
      "learning_rate": 0.0004912284430677949,
      "loss": 6.0262,
      "mean_token_accuracy": 0.14538998007774354,
      "num_tokens": 22522309.0,
      "step": 11115
    },
    {
      "entropy": 6.186139631271362,
      "epoch": 0.9717731364152757,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004912198251005228,
      "loss": 5.971,
      "mean_token_accuracy": 0.1455381765961647,
      "num_tokens": 22533510.0,
      "step": 11120
    },
    {
      "entropy": 6.1655395984649655,
      "epoch": 0.9722100847679804,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004912112029860411,
      "loss": 5.9542,
      "mean_token_accuracy": 0.14665468484163285,
      "num_tokens": 22543644.0,
      "step": 11125
    },
    {
      "entropy": 6.285950803756714,
      "epoch": 0.9726470331206851,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004912025767245153,
      "loss": 6.0938,
      "mean_token_accuracy": 0.13609763905405997,
      "num_tokens": 22553475.0,
      "step": 11130
    },
    {
      "entropy": 6.172232151031494,
      "epoch": 0.9730839814733898,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004911939463161108,
      "loss": 5.9258,
      "mean_token_accuracy": 0.14573410749435425,
      "num_tokens": 22563190.0,
      "step": 11135
    },
    {
      "entropy": 6.144096708297729,
      "epoch": 0.9735209298260945,
      "grad_norm": 2.21875,
      "learning_rate": 0.000491185311760993,
      "loss": 5.9591,
      "mean_token_accuracy": 0.14986318498849868,
      "num_tokens": 22572069.0,
      "step": 11140
    },
    {
      "entropy": 6.161580896377563,
      "epoch": 0.9739578781787993,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004911766730593276,
      "loss": 6.0439,
      "mean_token_accuracy": 0.13760749623179436,
      "num_tokens": 22581597.0,
      "step": 11145
    },
    {
      "entropy": 6.188813734054565,
      "epoch": 0.974394826531504,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004911680302112806,
      "loss": 5.9402,
      "mean_token_accuracy": 0.15137511417269706,
      "num_tokens": 22591486.0,
      "step": 11150
    },
    {
      "entropy": 6.244290208816528,
      "epoch": 0.9748317748842087,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004911593832170173,
      "loss": 6.014,
      "mean_token_accuracy": 0.1380392387509346,
      "num_tokens": 22601615.0,
      "step": 11155
    },
    {
      "entropy": 6.232906675338745,
      "epoch": 0.9752687232369134,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004911507320767038,
      "loss": 5.9956,
      "mean_token_accuracy": 0.1435872994363308,
      "num_tokens": 22611397.0,
      "step": 11160
    },
    {
      "entropy": 6.128848695755005,
      "epoch": 0.9757056715896181,
      "grad_norm": 2.109375,
      "learning_rate": 0.000491142076790506,
      "loss": 5.9786,
      "mean_token_accuracy": 0.14970599263906478,
      "num_tokens": 22621269.0,
      "step": 11165
    },
    {
      "entropy": 6.222478818893433,
      "epoch": 0.9761426199423228,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004911334173585899,
      "loss": 5.9652,
      "mean_token_accuracy": 0.14476251602172852,
      "num_tokens": 22631782.0,
      "step": 11170
    },
    {
      "entropy": 6.20287971496582,
      "epoch": 0.9765795682950276,
      "grad_norm": 2.625,
      "learning_rate": 0.0004911247537811216,
      "loss": 6.0432,
      "mean_token_accuracy": 0.14233123064041137,
      "num_tokens": 22642240.0,
      "step": 11175
    },
    {
      "entropy": 6.226010990142822,
      "epoch": 0.9770165166477323,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004911160860582674,
      "loss": 5.9912,
      "mean_token_accuracy": 0.1437778577208519,
      "num_tokens": 22651977.0,
      "step": 11180
    },
    {
      "entropy": 6.279437589645386,
      "epoch": 0.977453465000437,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004911074141901933,
      "loss": 6.0121,
      "mean_token_accuracy": 0.14684271663427353,
      "num_tokens": 22662871.0,
      "step": 11185
    },
    {
      "entropy": 6.217707872390747,
      "epoch": 0.9778904133531416,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004910987381770657,
      "loss": 5.9665,
      "mean_token_accuracy": 0.14834269061684607,
      "num_tokens": 22672083.0,
      "step": 11190
    },
    {
      "entropy": 6.124328088760376,
      "epoch": 0.9783273617058463,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004910900580190512,
      "loss": 5.9799,
      "mean_token_accuracy": 0.14960613399744033,
      "num_tokens": 22681247.0,
      "step": 11195
    },
    {
      "entropy": 6.159181261062622,
      "epoch": 0.9787643100585511,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004910813737163161,
      "loss": 6.0077,
      "mean_token_accuracy": 0.14306883215904237,
      "num_tokens": 22691231.0,
      "step": 11200
    },
    {
      "entropy": 6.213525772094727,
      "epoch": 0.9792012584112558,
      "grad_norm": 2.09375,
      "learning_rate": 0.000491072685269027,
      "loss": 5.9996,
      "mean_token_accuracy": 0.14373035728931427,
      "num_tokens": 22700712.0,
      "step": 11205
    },
    {
      "entropy": 6.301746463775634,
      "epoch": 0.9796382067639605,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004910639926773506,
      "loss": 6.0959,
      "mean_token_accuracy": 0.1427776262164116,
      "num_tokens": 22710961.0,
      "step": 11210
    },
    {
      "entropy": 6.22751030921936,
      "epoch": 0.9800751551166652,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004910552959414538,
      "loss": 5.9628,
      "mean_token_accuracy": 0.14933306127786636,
      "num_tokens": 22720450.0,
      "step": 11215
    },
    {
      "entropy": 6.106460380554199,
      "epoch": 0.9805121034693699,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004910465950615029,
      "loss": 5.9218,
      "mean_token_accuracy": 0.15042502135038377,
      "num_tokens": 22730337.0,
      "step": 11220
    },
    {
      "entropy": 6.137769746780395,
      "epoch": 0.9809490518220746,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004910378900376651,
      "loss": 6.05,
      "mean_token_accuracy": 0.140307005494833,
      "num_tokens": 22742379.0,
      "step": 11225
    },
    {
      "entropy": 6.262981986999511,
      "epoch": 0.9813860001747794,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004910291808701077,
      "loss": 6.0635,
      "mean_token_accuracy": 0.14013471826910973,
      "num_tokens": 22753176.0,
      "step": 11230
    },
    {
      "entropy": 6.246870422363282,
      "epoch": 0.9818229485274841,
      "grad_norm": 4.875,
      "learning_rate": 0.0004910204675589972,
      "loss": 6.0184,
      "mean_token_accuracy": 0.14495594948530197,
      "num_tokens": 22763263.0,
      "step": 11235
    },
    {
      "entropy": 6.135342979431153,
      "epoch": 0.9822598968801888,
      "grad_norm": 3.25,
      "learning_rate": 0.0004910117501045009,
      "loss": 6.0356,
      "mean_token_accuracy": 0.14509285241365433,
      "num_tokens": 22774630.0,
      "step": 11240
    },
    {
      "entropy": 6.194659757614136,
      "epoch": 0.9826968452328935,
      "grad_norm": 14.375,
      "learning_rate": 0.0004910030285067861,
      "loss": 6.0125,
      "mean_token_accuracy": 0.14676431119441985,
      "num_tokens": 22785027.0,
      "step": 11245
    },
    {
      "entropy": 6.24728946685791,
      "epoch": 0.9831337935855982,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004909943027660201,
      "loss": 6.0574,
      "mean_token_accuracy": 0.13427724316716194,
      "num_tokens": 22796399.0,
      "step": 11250
    },
    {
      "entropy": 6.267506265640259,
      "epoch": 0.9835707419383029,
      "grad_norm": 2.34375,
      "learning_rate": 0.00049098557288237,
      "loss": 6.0008,
      "mean_token_accuracy": 0.14368091523647308,
      "num_tokens": 22806352.0,
      "step": 11255
    },
    {
      "entropy": 6.247375726699829,
      "epoch": 0.9840076902910077,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004909768388560036,
      "loss": 6.0621,
      "mean_token_accuracy": 0.14165429025888443,
      "num_tokens": 22818369.0,
      "step": 11260
    },
    {
      "entropy": 6.21934142112732,
      "epoch": 0.9844446386437123,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004909681006870882,
      "loss": 5.9801,
      "mean_token_accuracy": 0.14450762867927552,
      "num_tokens": 22828063.0,
      "step": 11265
    },
    {
      "entropy": 6.245118045806885,
      "epoch": 0.984881586996417,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004909593583757915,
      "loss": 6.0414,
      "mean_token_accuracy": 0.13885763436555862,
      "num_tokens": 22839265.0,
      "step": 11270
    },
    {
      "entropy": 6.202704000473022,
      "epoch": 0.9853185353491217,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004909506119222812,
      "loss": 5.9891,
      "mean_token_accuracy": 0.15027930736541747,
      "num_tokens": 22848501.0,
      "step": 11275
    },
    {
      "entropy": 6.171825742721557,
      "epoch": 0.9857554837018264,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004909418613267248,
      "loss": 5.9268,
      "mean_token_accuracy": 0.1502961188554764,
      "num_tokens": 22859176.0,
      "step": 11280
    },
    {
      "entropy": 6.187764453887939,
      "epoch": 0.9861924320545311,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004909331065892905,
      "loss": 5.9456,
      "mean_token_accuracy": 0.14518322348594664,
      "num_tokens": 22869312.0,
      "step": 11285
    },
    {
      "entropy": 6.227721261978149,
      "epoch": 0.9866293804072359,
      "grad_norm": 3.28125,
      "learning_rate": 0.000490924347710146,
      "loss": 6.0531,
      "mean_token_accuracy": 0.14625362008810044,
      "num_tokens": 22878699.0,
      "step": 11290
    },
    {
      "entropy": 6.1834508895874025,
      "epoch": 0.9870663287599406,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004909155846894595,
      "loss": 6.0875,
      "mean_token_accuracy": 0.14202336668968202,
      "num_tokens": 22889147.0,
      "step": 11295
    },
    {
      "entropy": 6.241399526596069,
      "epoch": 0.9875032771126453,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004909068175273989,
      "loss": 5.9764,
      "mean_token_accuracy": 0.14701655730605126,
      "num_tokens": 22899085.0,
      "step": 11300
    },
    {
      "entropy": 6.2564019680023195,
      "epoch": 0.98794022546535,
      "grad_norm": 4.875,
      "learning_rate": 0.0004908980462241323,
      "loss": 6.1063,
      "mean_token_accuracy": 0.14100436270236968,
      "num_tokens": 22909924.0,
      "step": 11305
    },
    {
      "entropy": 6.254339504241943,
      "epoch": 0.9883771738180547,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004908892707798281,
      "loss": 6.0698,
      "mean_token_accuracy": 0.14327240511775016,
      "num_tokens": 22920426.0,
      "step": 11310
    },
    {
      "entropy": 6.254977798461914,
      "epoch": 0.9888141221707594,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004908804911946546,
      "loss": 5.9912,
      "mean_token_accuracy": 0.1458214096724987,
      "num_tokens": 22929929.0,
      "step": 11315
    },
    {
      "entropy": 6.199986457824707,
      "epoch": 0.9892510705234642,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004908717074687801,
      "loss": 5.9296,
      "mean_token_accuracy": 0.15381366908550262,
      "num_tokens": 22940102.0,
      "step": 11320
    },
    {
      "entropy": 6.232103538513184,
      "epoch": 0.9896880188761689,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004908629196023732,
      "loss": 5.9649,
      "mean_token_accuracy": 0.15618996918201447,
      "num_tokens": 22949490.0,
      "step": 11325
    },
    {
      "entropy": 6.224821615219116,
      "epoch": 0.9901249672288736,
      "grad_norm": 5.21875,
      "learning_rate": 0.0004908541275956024,
      "loss": 5.9302,
      "mean_token_accuracy": 0.14935585260391235,
      "num_tokens": 22958119.0,
      "step": 11330
    },
    {
      "entropy": 6.201994943618774,
      "epoch": 0.9905619155815782,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004908453314486363,
      "loss": 5.9957,
      "mean_token_accuracy": 0.143864306807518,
      "num_tokens": 22969447.0,
      "step": 11335
    },
    {
      "entropy": 6.118885803222656,
      "epoch": 0.9909988639342829,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004908365311616437,
      "loss": 5.875,
      "mean_token_accuracy": 0.152784363925457,
      "num_tokens": 22978925.0,
      "step": 11340
    },
    {
      "entropy": 6.194061374664306,
      "epoch": 0.9914358122869876,
      "grad_norm": 4.125,
      "learning_rate": 0.0004908277267347931,
      "loss": 5.9936,
      "mean_token_accuracy": 0.14802219718694687,
      "num_tokens": 22988861.0,
      "step": 11345
    },
    {
      "entropy": 6.21967887878418,
      "epoch": 0.9918727606396924,
      "grad_norm": 2.75,
      "learning_rate": 0.0004908189181682538,
      "loss": 6.0361,
      "mean_token_accuracy": 0.1465169072151184,
      "num_tokens": 22998700.0,
      "step": 11350
    },
    {
      "entropy": 6.207000207901001,
      "epoch": 0.9923097089923971,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004908101054621946,
      "loss": 5.9662,
      "mean_token_accuracy": 0.1498105153441429,
      "num_tokens": 23009095.0,
      "step": 11355
    },
    {
      "entropy": 6.11841926574707,
      "epoch": 0.9927466573451018,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004908012886167845,
      "loss": 5.9851,
      "mean_token_accuracy": 0.1481009840965271,
      "num_tokens": 23019484.0,
      "step": 11360
    },
    {
      "entropy": 6.14017276763916,
      "epoch": 0.9931836056978065,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004907924676321924,
      "loss": 5.9555,
      "mean_token_accuracy": 0.14604686200618744,
      "num_tokens": 23029867.0,
      "step": 11365
    },
    {
      "entropy": 6.212933015823364,
      "epoch": 0.9936205540505112,
      "grad_norm": 4.90625,
      "learning_rate": 0.000490783642508588,
      "loss": 5.9252,
      "mean_token_accuracy": 0.15722475349903106,
      "num_tokens": 23040179.0,
      "step": 11370
    },
    {
      "entropy": 6.217982435226441,
      "epoch": 0.9940575024032159,
      "grad_norm": 3.046875,
      "learning_rate": 0.00049077481324614,
      "loss": 5.9209,
      "mean_token_accuracy": 0.15354355573654174,
      "num_tokens": 23050389.0,
      "step": 11375
    },
    {
      "entropy": 6.113583707809449,
      "epoch": 0.9944944507559207,
      "grad_norm": 2.96875,
      "learning_rate": 0.000490765979845018,
      "loss": 5.9516,
      "mean_token_accuracy": 0.15064753219485283,
      "num_tokens": 23059984.0,
      "step": 11380
    },
    {
      "entropy": 6.197798204421997,
      "epoch": 0.9949313991086254,
      "grad_norm": 16.875,
      "learning_rate": 0.0004907571423053917,
      "loss": 5.9715,
      "mean_token_accuracy": 0.14383725970983505,
      "num_tokens": 23069828.0,
      "step": 11385
    },
    {
      "entropy": 6.298434257507324,
      "epoch": 0.9953683474613301,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004907483006274302,
      "loss": 6.0085,
      "mean_token_accuracy": 0.14273910969495773,
      "num_tokens": 23079223.0,
      "step": 11390
    },
    {
      "entropy": 6.198758459091186,
      "epoch": 0.9958052958140348,
      "grad_norm": 2.125,
      "learning_rate": 0.0004907394548113033,
      "loss": 6.0323,
      "mean_token_accuracy": 0.14386327266693116,
      "num_tokens": 23088540.0,
      "step": 11395
    },
    {
      "entropy": 6.150919389724732,
      "epoch": 0.9962422441667395,
      "grad_norm": 3.0,
      "learning_rate": 0.0004907306048571806,
      "loss": 6.0861,
      "mean_token_accuracy": 0.14062410742044448,
      "num_tokens": 23098617.0,
      "step": 11400
    },
    {
      "entropy": 6.188783264160156,
      "epoch": 0.9966791925194443,
      "grad_norm": 2.625,
      "learning_rate": 0.0004907217507652317,
      "loss": 5.9777,
      "mean_token_accuracy": 0.14869602024555206,
      "num_tokens": 23109702.0,
      "step": 11405
    },
    {
      "entropy": 6.269557237625122,
      "epoch": 0.997116140872149,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004907128925356268,
      "loss": 6.0179,
      "mean_token_accuracy": 0.14628712385892867,
      "num_tokens": 23120295.0,
      "step": 11410
    },
    {
      "entropy": 6.251343870162964,
      "epoch": 0.9975530892248536,
      "grad_norm": 5.25,
      "learning_rate": 0.0004907040301685355,
      "loss": 6.005,
      "mean_token_accuracy": 0.141085384786129,
      "num_tokens": 23131999.0,
      "step": 11415
    },
    {
      "entropy": 6.186858892440796,
      "epoch": 0.9979900375775583,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004906951636641279,
      "loss": 6.004,
      "mean_token_accuracy": 0.14834101349115372,
      "num_tokens": 23141060.0,
      "step": 11420
    },
    {
      "entropy": 6.163265514373779,
      "epoch": 0.998426985930263,
      "grad_norm": 4.0625,
      "learning_rate": 0.000490686293022574,
      "loss": 5.9792,
      "mean_token_accuracy": 0.14975388795137407,
      "num_tokens": 23151607.0,
      "step": 11425
    },
    {
      "entropy": 6.212855100631714,
      "epoch": 0.9988639342829677,
      "grad_norm": 2.609375,
      "learning_rate": 0.000490677418244044,
      "loss": 6.0458,
      "mean_token_accuracy": 0.13977433294057845,
      "num_tokens": 23161892.0,
      "step": 11430
    },
    {
      "entropy": 6.243049621582031,
      "epoch": 0.9993008826356725,
      "grad_norm": 6.9375,
      "learning_rate": 0.0004906685393287081,
      "loss": 6.0301,
      "mean_token_accuracy": 0.14472009390592575,
      "num_tokens": 23172112.0,
      "step": 11435
    },
    {
      "entropy": 6.29594612121582,
      "epoch": 0.9997378309883772,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004906596562767367,
      "loss": 5.9398,
      "mean_token_accuracy": 0.14860506355762482,
      "num_tokens": 23181435.0,
      "step": 11440
    },
    {
      "entropy": 6.125890064239502,
      "epoch": 1.0001747793410818,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004906507690883,
      "loss": 5.8701,
      "mean_token_accuracy": 0.153719761967659,
      "num_tokens": 23191084.0,
      "step": 11445
    },
    {
      "entropy": 6.128403663635254,
      "epoch": 1.0006117276937867,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004906418777635685,
      "loss": 5.9783,
      "mean_token_accuracy": 0.14208925738930703,
      "num_tokens": 23202372.0,
      "step": 11450
    },
    {
      "entropy": 6.237107753753662,
      "epoch": 1.0010486760464914,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004906329823027129,
      "loss": 5.9734,
      "mean_token_accuracy": 0.1449448399245739,
      "num_tokens": 23214113.0,
      "step": 11455
    },
    {
      "entropy": 6.249896621704101,
      "epoch": 1.001485624399196,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004906240827059036,
      "loss": 5.8868,
      "mean_token_accuracy": 0.1506553828716278,
      "num_tokens": 23223562.0,
      "step": 11460
    },
    {
      "entropy": 6.213391733169556,
      "epoch": 1.0019225727519008,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004906151789733114,
      "loss": 5.9835,
      "mean_token_accuracy": 0.14346898794174195,
      "num_tokens": 23232901.0,
      "step": 11465
    },
    {
      "entropy": 6.19885401725769,
      "epoch": 1.0023595211046055,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004906062711051073,
      "loss": 5.998,
      "mean_token_accuracy": 0.1453511595726013,
      "num_tokens": 23242869.0,
      "step": 11470
    },
    {
      "entropy": 6.220397806167602,
      "epoch": 1.0027964694573102,
      "grad_norm": 2.25,
      "learning_rate": 0.0004905973591014617,
      "loss": 5.8953,
      "mean_token_accuracy": 0.15457765609025956,
      "num_tokens": 23253653.0,
      "step": 11475
    },
    {
      "entropy": 6.164454364776612,
      "epoch": 1.0032334178100148,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004905884429625459,
      "loss": 5.9233,
      "mean_token_accuracy": 0.1484118938446045,
      "num_tokens": 23263530.0,
      "step": 11480
    },
    {
      "entropy": 6.224922561645508,
      "epoch": 1.0036703661627195,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004905795226885308,
      "loss": 5.9625,
      "mean_token_accuracy": 0.1458275243639946,
      "num_tokens": 23274946.0,
      "step": 11485
    },
    {
      "entropy": 6.260493659973145,
      "epoch": 1.0041073145154242,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004905705982795874,
      "loss": 5.9598,
      "mean_token_accuracy": 0.15124345943331718,
      "num_tokens": 23283811.0,
      "step": 11490
    },
    {
      "entropy": 6.132644557952881,
      "epoch": 1.004544262868129,
      "grad_norm": 2.5625,
      "learning_rate": 0.000490561669735887,
      "loss": 5.9931,
      "mean_token_accuracy": 0.14670841172337531,
      "num_tokens": 23294554.0,
      "step": 11495
    },
    {
      "entropy": 6.100069761276245,
      "epoch": 1.0049812112208336,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004905527370576009,
      "loss": 5.8439,
      "mean_token_accuracy": 0.15170976519584656,
      "num_tokens": 23304245.0,
      "step": 11500
    },
    {
      "entropy": 6.321207714080811,
      "epoch": 1.0054181595735383,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004905438002449004,
      "loss": 5.9934,
      "mean_token_accuracy": 0.1374276116490364,
      "num_tokens": 23315169.0,
      "step": 11505
    },
    {
      "entropy": 6.165538024902344,
      "epoch": 1.0058551079262432,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004905348592979567,
      "loss": 5.91,
      "mean_token_accuracy": 0.14459408447146416,
      "num_tokens": 23325528.0,
      "step": 11510
    },
    {
      "entropy": 6.1551250457763675,
      "epoch": 1.006292056278948,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004905259142169415,
      "loss": 6.026,
      "mean_token_accuracy": 0.14281592890620232,
      "num_tokens": 23336744.0,
      "step": 11515
    },
    {
      "entropy": 6.20447883605957,
      "epoch": 1.0067290046316526,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004905169650020264,
      "loss": 5.8811,
      "mean_token_accuracy": 0.15230331122875213,
      "num_tokens": 23346092.0,
      "step": 11520
    },
    {
      "entropy": 6.143009042739868,
      "epoch": 1.0071659529843573,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004905080116533831,
      "loss": 5.9392,
      "mean_token_accuracy": 0.13944583237171174,
      "num_tokens": 23357326.0,
      "step": 11525
    },
    {
      "entropy": 6.1353497982025145,
      "epoch": 1.007602901337062,
      "grad_norm": 5.78125,
      "learning_rate": 0.000490499054171183,
      "loss": 5.9435,
      "mean_token_accuracy": 0.14872722774744035,
      "num_tokens": 23367789.0,
      "step": 11530
    },
    {
      "entropy": 6.225711488723755,
      "epoch": 1.0080398496897667,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004904900925555982,
      "loss": 5.9225,
      "mean_token_accuracy": 0.14835504293441773,
      "num_tokens": 23377499.0,
      "step": 11535
    },
    {
      "entropy": 6.253147792816162,
      "epoch": 1.0084767980424714,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004904811268068005,
      "loss": 5.9179,
      "mean_token_accuracy": 0.14387512803077698,
      "num_tokens": 23387867.0,
      "step": 11540
    },
    {
      "entropy": 6.188341188430786,
      "epoch": 1.008913746395176,
      "grad_norm": 3.0,
      "learning_rate": 0.0004904721569249618,
      "loss": 6.0106,
      "mean_token_accuracy": 0.142968862503767,
      "num_tokens": 23398179.0,
      "step": 11545
    },
    {
      "entropy": 6.174725770950317,
      "epoch": 1.0093506947478808,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004904631829102543,
      "loss": 6.0121,
      "mean_token_accuracy": 0.14683498293161393,
      "num_tokens": 23407407.0,
      "step": 11550
    },
    {
      "entropy": 6.186086559295655,
      "epoch": 1.0097876431005854,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004904542047628501,
      "loss": 5.8626,
      "mean_token_accuracy": 0.15031913071870803,
      "num_tokens": 23417324.0,
      "step": 11555
    },
    {
      "entropy": 6.215742206573486,
      "epoch": 1.0102245914532901,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004904452224829212,
      "loss": 5.932,
      "mean_token_accuracy": 0.14663962423801422,
      "num_tokens": 23427157.0,
      "step": 11560
    },
    {
      "entropy": 6.16818528175354,
      "epoch": 1.0106615398059948,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004904362360706403,
      "loss": 5.9659,
      "mean_token_accuracy": 0.14321382492780685,
      "num_tokens": 23437712.0,
      "step": 11565
    },
    {
      "entropy": 6.234272527694702,
      "epoch": 1.0110984881586997,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004904272455261793,
      "loss": 5.9244,
      "mean_token_accuracy": 0.15019643679261208,
      "num_tokens": 23447949.0,
      "step": 11570
    },
    {
      "entropy": 6.123412847518921,
      "epoch": 1.0115354365114044,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004904182508497109,
      "loss": 5.8796,
      "mean_token_accuracy": 0.14437250867486,
      "num_tokens": 23457319.0,
      "step": 11575
    },
    {
      "entropy": 6.214909410476684,
      "epoch": 1.0119723848641091,
      "grad_norm": 2.625,
      "learning_rate": 0.0004904092520414075,
      "loss": 5.9888,
      "mean_token_accuracy": 0.14348120987415314,
      "num_tokens": 23468245.0,
      "step": 11580
    },
    {
      "entropy": 6.255124092102051,
      "epoch": 1.0124093332168138,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004904002491014418,
      "loss": 5.9534,
      "mean_token_accuracy": 0.1529581293463707,
      "num_tokens": 23477049.0,
      "step": 11585
    },
    {
      "entropy": 6.189140129089355,
      "epoch": 1.0128462815695185,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004903912420299866,
      "loss": 6.0188,
      "mean_token_accuracy": 0.14091739803552628,
      "num_tokens": 23488574.0,
      "step": 11590
    },
    {
      "entropy": 6.213904619216919,
      "epoch": 1.0132832299222232,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004903822308272144,
      "loss": 5.9408,
      "mean_token_accuracy": 0.15238529890775682,
      "num_tokens": 23499282.0,
      "step": 11595
    },
    {
      "entropy": 6.188621473312378,
      "epoch": 1.0137201782749279,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004903732154932982,
      "loss": 5.9359,
      "mean_token_accuracy": 0.15845069140195847,
      "num_tokens": 23508777.0,
      "step": 11600
    },
    {
      "entropy": 6.18798246383667,
      "epoch": 1.0141571266276326,
      "grad_norm": 3.375,
      "learning_rate": 0.0004903641960284108,
      "loss": 5.9186,
      "mean_token_accuracy": 0.15181176364421844,
      "num_tokens": 23518818.0,
      "step": 11605
    },
    {
      "entropy": 6.164873456954956,
      "epoch": 1.0145940749803373,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004903551724327254,
      "loss": 5.9269,
      "mean_token_accuracy": 0.1478419706225395,
      "num_tokens": 23529668.0,
      "step": 11610
    },
    {
      "entropy": 6.253747272491455,
      "epoch": 1.015031023333042,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004903461447064147,
      "loss": 5.9657,
      "mean_token_accuracy": 0.15092179626226426,
      "num_tokens": 23540418.0,
      "step": 11615
    },
    {
      "entropy": 6.229628038406372,
      "epoch": 1.0154679716857467,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004903371128496523,
      "loss": 5.9238,
      "mean_token_accuracy": 0.1532558262348175,
      "num_tokens": 23550035.0,
      "step": 11620
    },
    {
      "entropy": 6.139720869064331,
      "epoch": 1.0159049200384516,
      "grad_norm": 4.5,
      "learning_rate": 0.0004903280768626113,
      "loss": 5.9612,
      "mean_token_accuracy": 0.1426764450967312,
      "num_tokens": 23559487.0,
      "step": 11625
    },
    {
      "entropy": 6.167341470718384,
      "epoch": 1.0163418683911563,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004903190367454648,
      "loss": 5.9033,
      "mean_token_accuracy": 0.15389101952314377,
      "num_tokens": 23569146.0,
      "step": 11630
    },
    {
      "entropy": 6.159114265441895,
      "epoch": 1.016778816743861,
      "grad_norm": 5.9375,
      "learning_rate": 0.0004903099924983864,
      "loss": 5.8807,
      "mean_token_accuracy": 0.14899331256747245,
      "num_tokens": 23579458.0,
      "step": 11635
    },
    {
      "entropy": 6.204789733886718,
      "epoch": 1.0172157650965656,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004903009441215495,
      "loss": 5.8922,
      "mean_token_accuracy": 0.14243703931570054,
      "num_tokens": 23589416.0,
      "step": 11640
    },
    {
      "entropy": 6.151660490036011,
      "epoch": 1.0176527134492703,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004902918916151278,
      "loss": 5.9843,
      "mean_token_accuracy": 0.1495899051427841,
      "num_tokens": 23599588.0,
      "step": 11645
    },
    {
      "entropy": 6.232296943664551,
      "epoch": 1.018089661801975,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004902828349792947,
      "loss": 5.9681,
      "mean_token_accuracy": 0.14577097371220588,
      "num_tokens": 23610116.0,
      "step": 11650
    },
    {
      "entropy": 6.173538064956665,
      "epoch": 1.0185266101546797,
      "grad_norm": 2.140625,
      "learning_rate": 0.000490273774214224,
      "loss": 5.91,
      "mean_token_accuracy": 0.1557459205389023,
      "num_tokens": 23619906.0,
      "step": 11655
    },
    {
      "entropy": 6.142177963256836,
      "epoch": 1.0189635585073844,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004902647093200896,
      "loss": 5.9258,
      "mean_token_accuracy": 0.15174940824508668,
      "num_tokens": 23630351.0,
      "step": 11660
    },
    {
      "entropy": 6.165373945236206,
      "epoch": 1.019400506860089,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004902556402970652,
      "loss": 5.9869,
      "mean_token_accuracy": 0.14259752109646798,
      "num_tokens": 23639696.0,
      "step": 11665
    },
    {
      "entropy": 6.214051818847656,
      "epoch": 1.0198374552127938,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004902465671453248,
      "loss": 5.9313,
      "mean_token_accuracy": 0.14681773036718368,
      "num_tokens": 23649388.0,
      "step": 11670
    },
    {
      "entropy": 6.206135511398315,
      "epoch": 1.0202744035654985,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004902374898650425,
      "loss": 5.9775,
      "mean_token_accuracy": 0.14501645565032958,
      "num_tokens": 23659125.0,
      "step": 11675
    },
    {
      "entropy": 6.299280166625977,
      "epoch": 1.0207113519182032,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004902284084563924,
      "loss": 5.9878,
      "mean_token_accuracy": 0.1447453536093235,
      "num_tokens": 23668817.0,
      "step": 11680
    },
    {
      "entropy": 6.153827667236328,
      "epoch": 1.021148300270908,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004902193229195487,
      "loss": 5.9407,
      "mean_token_accuracy": 0.1484477698802948,
      "num_tokens": 23679378.0,
      "step": 11685
    },
    {
      "entropy": 6.11264123916626,
      "epoch": 1.0215852486236128,
      "grad_norm": 3.125,
      "learning_rate": 0.0004902102332546855,
      "loss": 5.9248,
      "mean_token_accuracy": 0.14696565568447112,
      "num_tokens": 23689400.0,
      "step": 11690
    },
    {
      "entropy": 6.288630819320678,
      "epoch": 1.0220221969763175,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004902011394619772,
      "loss": 6.0662,
      "mean_token_accuracy": 0.13742880672216415,
      "num_tokens": 23699503.0,
      "step": 11695
    },
    {
      "entropy": 6.254185009002685,
      "epoch": 1.0224591453290222,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004901920415415985,
      "loss": 6.0063,
      "mean_token_accuracy": 0.14093072190880776,
      "num_tokens": 23710913.0,
      "step": 11700
    },
    {
      "entropy": 6.160375022888184,
      "epoch": 1.0228960936817268,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004901829394937234,
      "loss": 5.9092,
      "mean_token_accuracy": 0.14887271896004678,
      "num_tokens": 23720730.0,
      "step": 11705
    },
    {
      "entropy": 6.199987697601318,
      "epoch": 1.0233330420344315,
      "grad_norm": 1.921875,
      "learning_rate": 0.000490173833318527,
      "loss": 5.9701,
      "mean_token_accuracy": 0.1405739016830921,
      "num_tokens": 23730902.0,
      "step": 11710
    },
    {
      "entropy": 6.196088171005249,
      "epoch": 1.0237699903871362,
      "grad_norm": 2.875,
      "learning_rate": 0.0004901647230161837,
      "loss": 6.005,
      "mean_token_accuracy": 0.14492653608322142,
      "num_tokens": 23741901.0,
      "step": 11715
    },
    {
      "entropy": 6.220611524581909,
      "epoch": 1.024206938739841,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004901556085868681,
      "loss": 5.9108,
      "mean_token_accuracy": 0.14587837159633638,
      "num_tokens": 23752029.0,
      "step": 11720
    },
    {
      "entropy": 6.147278547286987,
      "epoch": 1.0246438870925456,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004901464900307553,
      "loss": 5.9036,
      "mean_token_accuracy": 0.15617354661226274,
      "num_tokens": 23762125.0,
      "step": 11725
    },
    {
      "entropy": 6.174932861328125,
      "epoch": 1.0250808354452503,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004901373673480201,
      "loss": 5.9619,
      "mean_token_accuracy": 0.15121276825666427,
      "num_tokens": 23772056.0,
      "step": 11730
    },
    {
      "entropy": 6.160613059997559,
      "epoch": 1.025517783797955,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004901282405388375,
      "loss": 5.8968,
      "mean_token_accuracy": 0.15151197090744972,
      "num_tokens": 23781441.0,
      "step": 11735
    },
    {
      "entropy": 6.1676229476928714,
      "epoch": 1.0259547321506597,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004901191096033825,
      "loss": 5.8869,
      "mean_token_accuracy": 0.15124904215335847,
      "num_tokens": 23791472.0,
      "step": 11740
    },
    {
      "entropy": 6.137651062011718,
      "epoch": 1.0263916805033646,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004901099745418302,
      "loss": 5.9157,
      "mean_token_accuracy": 0.14580238312482835,
      "num_tokens": 23801940.0,
      "step": 11745
    },
    {
      "entropy": 6.1130469799041744,
      "epoch": 1.0268286288560693,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004901008353543559,
      "loss": 5.8517,
      "mean_token_accuracy": 0.15534514486789702,
      "num_tokens": 23811515.0,
      "step": 11750
    },
    {
      "entropy": 6.162692975997925,
      "epoch": 1.027265577208774,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004900916920411349,
      "loss": 5.9132,
      "mean_token_accuracy": 0.1488305225968361,
      "num_tokens": 23821871.0,
      "step": 11755
    },
    {
      "entropy": 6.107495069503784,
      "epoch": 1.0277025255614787,
      "grad_norm": 3.5,
      "learning_rate": 0.0004900825446023426,
      "loss": 5.8774,
      "mean_token_accuracy": 0.15171604752540588,
      "num_tokens": 23831677.0,
      "step": 11760
    },
    {
      "entropy": 6.172711324691773,
      "epoch": 1.0281394739141834,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004900733930381543,
      "loss": 5.9426,
      "mean_token_accuracy": 0.14815112799406052,
      "num_tokens": 23841415.0,
      "step": 11765
    },
    {
      "entropy": 6.236703205108642,
      "epoch": 1.028576422266888,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004900642373487455,
      "loss": 5.9806,
      "mean_token_accuracy": 0.14841483831405639,
      "num_tokens": 23851809.0,
      "step": 11770
    },
    {
      "entropy": 6.196333360671997,
      "epoch": 1.0290133706195927,
      "grad_norm": 6.875,
      "learning_rate": 0.0004900550775342921,
      "loss": 5.9257,
      "mean_token_accuracy": 0.14207281544804573,
      "num_tokens": 23861981.0,
      "step": 11775
    },
    {
      "entropy": 6.177800512313842,
      "epoch": 1.0294503189722974,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004900459135949697,
      "loss": 5.9113,
      "mean_token_accuracy": 0.14654712975025178,
      "num_tokens": 23872437.0,
      "step": 11780
    },
    {
      "entropy": 6.179175281524659,
      "epoch": 1.0298872673250021,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004900367455309537,
      "loss": 5.9405,
      "mean_token_accuracy": 0.14729941487312317,
      "num_tokens": 23881650.0,
      "step": 11785
    },
    {
      "entropy": 6.145957088470459,
      "epoch": 1.0303242156777068,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004900275733424205,
      "loss": 5.9554,
      "mean_token_accuracy": 0.1503347083926201,
      "num_tokens": 23891631.0,
      "step": 11790
    },
    {
      "entropy": 6.130884408950806,
      "epoch": 1.0307611640304115,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004900183970295457,
      "loss": 5.8837,
      "mean_token_accuracy": 0.1534467577934265,
      "num_tokens": 23901725.0,
      "step": 11795
    },
    {
      "entropy": 6.1669008255004885,
      "epoch": 1.0311981123831164,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004900092165925053,
      "loss": 5.9087,
      "mean_token_accuracy": 0.14660380333662032,
      "num_tokens": 23911850.0,
      "step": 11800
    },
    {
      "entropy": 6.215511226654053,
      "epoch": 1.0316350607358211,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004900000320314757,
      "loss": 6.0501,
      "mean_token_accuracy": 0.14064744338393212,
      "num_tokens": 23922571.0,
      "step": 11805
    },
    {
      "entropy": 6.2230743885040285,
      "epoch": 1.0320720090885258,
      "grad_norm": 1.875,
      "learning_rate": 0.0004899908433466326,
      "loss": 5.9259,
      "mean_token_accuracy": 0.1500337116420269,
      "num_tokens": 23932637.0,
      "step": 11810
    },
    {
      "entropy": 6.243884754180908,
      "epoch": 1.0325089574412305,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004899816505381525,
      "loss": 5.9987,
      "mean_token_accuracy": 0.14186873733997346,
      "num_tokens": 23943288.0,
      "step": 11815
    },
    {
      "entropy": 6.133871936798096,
      "epoch": 1.0329459057939352,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004899724536062117,
      "loss": 5.9084,
      "mean_token_accuracy": 0.1506035581231117,
      "num_tokens": 23952853.0,
      "step": 11820
    },
    {
      "entropy": 6.1379118919372555,
      "epoch": 1.0333828541466399,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004899632525509867,
      "loss": 5.935,
      "mean_token_accuracy": 0.1465439938008785,
      "num_tokens": 23963111.0,
      "step": 11825
    },
    {
      "entropy": 6.206754207611084,
      "epoch": 1.0338198024993446,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004899540473726538,
      "loss": 5.8826,
      "mean_token_accuracy": 0.15071972832083702,
      "num_tokens": 23972468.0,
      "step": 11830
    },
    {
      "entropy": 6.21347222328186,
      "epoch": 1.0342567508520493,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004899448380713897,
      "loss": 5.9035,
      "mean_token_accuracy": 0.15384145379066466,
      "num_tokens": 23982567.0,
      "step": 11835
    },
    {
      "entropy": 6.12376971244812,
      "epoch": 1.034693699204754,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004899356246473709,
      "loss": 5.9214,
      "mean_token_accuracy": 0.1518683910369873,
      "num_tokens": 23993283.0,
      "step": 11840
    },
    {
      "entropy": 6.186660671234131,
      "epoch": 1.0351306475574586,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004899264071007741,
      "loss": 5.8942,
      "mean_token_accuracy": 0.15201990753412248,
      "num_tokens": 24004504.0,
      "step": 11845
    },
    {
      "entropy": 6.169562196731567,
      "epoch": 1.0355675959101633,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004899171854317764,
      "loss": 5.9334,
      "mean_token_accuracy": 0.15003803074359895,
      "num_tokens": 24014557.0,
      "step": 11850
    },
    {
      "entropy": 6.148725986480713,
      "epoch": 1.036004544262868,
      "grad_norm": 4.375,
      "learning_rate": 0.0004899079596405543,
      "loss": 5.9134,
      "mean_token_accuracy": 0.14588637053966522,
      "num_tokens": 24024495.0,
      "step": 11855
    },
    {
      "entropy": 6.113318681716919,
      "epoch": 1.036441492615573,
      "grad_norm": 2.921875,
      "learning_rate": 0.000489898729727285,
      "loss": 5.9295,
      "mean_token_accuracy": 0.14993472695350646,
      "num_tokens": 24034984.0,
      "step": 11860
    },
    {
      "entropy": 6.190798377990722,
      "epoch": 1.0368784409682776,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004898894956921455,
      "loss": 5.9036,
      "mean_token_accuracy": 0.15183772817254065,
      "num_tokens": 24044379.0,
      "step": 11865
    },
    {
      "entropy": 6.232519006729126,
      "epoch": 1.0373153893209823,
      "grad_norm": 2.375,
      "learning_rate": 0.0004898802575353129,
      "loss": 5.925,
      "mean_token_accuracy": 0.14870004057884217,
      "num_tokens": 24054133.0,
      "step": 11870
    },
    {
      "entropy": 6.080832529067993,
      "epoch": 1.037752337673687,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004898710152569642,
      "loss": 5.8852,
      "mean_token_accuracy": 0.14960063993930817,
      "num_tokens": 24064868.0,
      "step": 11875
    },
    {
      "entropy": 6.201964950561523,
      "epoch": 1.0381892860263917,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004898617688572769,
      "loss": 5.9326,
      "mean_token_accuracy": 0.14799922406673433,
      "num_tokens": 24074977.0,
      "step": 11880
    },
    {
      "entropy": 6.23028597831726,
      "epoch": 1.0386262343790964,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004898525183364284,
      "loss": 5.9849,
      "mean_token_accuracy": 0.15064526498317718,
      "num_tokens": 24085673.0,
      "step": 11885
    },
    {
      "entropy": 6.209465217590332,
      "epoch": 1.039063182731801,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004898432636945959,
      "loss": 5.977,
      "mean_token_accuracy": 0.15092883259058,
      "num_tokens": 24095465.0,
      "step": 11890
    },
    {
      "entropy": 6.144705772399902,
      "epoch": 1.0395001310845058,
      "grad_norm": 3.59375,
      "learning_rate": 0.000489834004931957,
      "loss": 5.9241,
      "mean_token_accuracy": 0.1513504572212696,
      "num_tokens": 24105576.0,
      "step": 11895
    },
    {
      "entropy": 6.171712017059326,
      "epoch": 1.0399370794372105,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004898247420486895,
      "loss": 5.9129,
      "mean_token_accuracy": 0.1441113233566284,
      "num_tokens": 24114998.0,
      "step": 11900
    },
    {
      "entropy": 6.17154221534729,
      "epoch": 1.0403740277899152,
      "grad_norm": 6.0625,
      "learning_rate": 0.0004898154750449707,
      "loss": 5.9085,
      "mean_token_accuracy": 0.15080181807279586,
      "num_tokens": 24125059.0,
      "step": 11905
    },
    {
      "entropy": 6.254666042327881,
      "epoch": 1.0408109761426199,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004898062039209786,
      "loss": 5.9263,
      "mean_token_accuracy": 0.14852346032857894,
      "num_tokens": 24134062.0,
      "step": 11910
    },
    {
      "entropy": 6.197595024108887,
      "epoch": 1.0412479244953246,
      "grad_norm": 3.515625,
      "learning_rate": 0.000489796928676891,
      "loss": 5.9407,
      "mean_token_accuracy": 0.14671702384948732,
      "num_tokens": 24144016.0,
      "step": 11915
    },
    {
      "entropy": 6.18574538230896,
      "epoch": 1.0416848728480295,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004897876493128856,
      "loss": 5.9095,
      "mean_token_accuracy": 0.14869166910648346,
      "num_tokens": 24154337.0,
      "step": 11920
    },
    {
      "entropy": 6.145417070388794,
      "epoch": 1.0421218212007342,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004897783658291406,
      "loss": 5.8958,
      "mean_token_accuracy": 0.15060929954051971,
      "num_tokens": 24164440.0,
      "step": 11925
    },
    {
      "entropy": 6.201339292526245,
      "epoch": 1.0425587695534388,
      "grad_norm": 2.640625,
      "learning_rate": 0.000489769078225834,
      "loss": 6.0275,
      "mean_token_accuracy": 0.14404802173376083,
      "num_tokens": 24174388.0,
      "step": 11930
    },
    {
      "entropy": 6.180160903930664,
      "epoch": 1.0429957179061435,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004897597865031441,
      "loss": 5.8999,
      "mean_token_accuracy": 0.15236803889274597,
      "num_tokens": 24183543.0,
      "step": 11935
    },
    {
      "entropy": 6.1767994403839115,
      "epoch": 1.0434326662588482,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004897504906612487,
      "loss": 5.9378,
      "mean_token_accuracy": 0.14996734261512756,
      "num_tokens": 24193847.0,
      "step": 11940
    },
    {
      "entropy": 6.154396438598633,
      "epoch": 1.043869614611553,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004897411907003265,
      "loss": 5.8986,
      "mean_token_accuracy": 0.1502672702074051,
      "num_tokens": 24203072.0,
      "step": 11945
    },
    {
      "entropy": 6.178869199752808,
      "epoch": 1.0443065629642576,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004897318866205558,
      "loss": 5.8518,
      "mean_token_accuracy": 0.15772329717874528,
      "num_tokens": 24213610.0,
      "step": 11950
    },
    {
      "entropy": 6.079909324645996,
      "epoch": 1.0447435113169623,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004897225784221149,
      "loss": 5.888,
      "mean_token_accuracy": 0.15807106643915175,
      "num_tokens": 24223571.0,
      "step": 11955
    },
    {
      "entropy": 6.1383027076721195,
      "epoch": 1.045180459669667,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004897132661051826,
      "loss": 5.9346,
      "mean_token_accuracy": 0.1453519806265831,
      "num_tokens": 24233867.0,
      "step": 11960
    },
    {
      "entropy": 6.146329164505005,
      "epoch": 1.0456174080223717,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004897039496699372,
      "loss": 5.9411,
      "mean_token_accuracy": 0.14822976738214494,
      "num_tokens": 24243853.0,
      "step": 11965
    },
    {
      "entropy": 6.166684150695801,
      "epoch": 1.0460543563750764,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004896946291165575,
      "loss": 5.9266,
      "mean_token_accuracy": 0.14673857092857362,
      "num_tokens": 24254049.0,
      "step": 11970
    },
    {
      "entropy": 6.150206089019775,
      "epoch": 1.0464913047277813,
      "grad_norm": 4.125,
      "learning_rate": 0.0004896853044452224,
      "loss": 5.9171,
      "mean_token_accuracy": 0.14790754914283752,
      "num_tokens": 24265166.0,
      "step": 11975
    },
    {
      "entropy": 6.174460506439209,
      "epoch": 1.046928253080486,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004896759756561106,
      "loss": 5.9162,
      "mean_token_accuracy": 0.15014341622591018,
      "num_tokens": 24274981.0,
      "step": 11980
    },
    {
      "entropy": 6.233915996551514,
      "epoch": 1.0473652014331907,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004896666427494012,
      "loss": 5.9401,
      "mean_token_accuracy": 0.14949152767658233,
      "num_tokens": 24285009.0,
      "step": 11985
    },
    {
      "entropy": 6.312009716033936,
      "epoch": 1.0478021497858954,
      "grad_norm": 2.609375,
      "learning_rate": 0.000489657305725273,
      "loss": 6.0029,
      "mean_token_accuracy": 0.14560900554060935,
      "num_tokens": 24294669.0,
      "step": 11990
    },
    {
      "entropy": 6.282224369049072,
      "epoch": 1.0482390981386,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004896479645839052,
      "loss": 5.9548,
      "mean_token_accuracy": 0.14923173934221268,
      "num_tokens": 24304647.0,
      "step": 11995
    },
    {
      "entropy": 6.093004846572876,
      "epoch": 1.0486760464913047,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004896386193254771,
      "loss": 5.8477,
      "mean_token_accuracy": 0.1570341557264328,
      "num_tokens": 24314266.0,
      "step": 12000
    },
    {
      "epoch": 1.0486760464913047,
      "eval_entropy": 5.884056688720525,
      "eval_loss": 6.006004333496094,
      "eval_mean_token_accuracy": 0.15266030561480842,
      "eval_num_tokens": 24314266.0,
      "eval_runtime": 28.2209,
      "eval_samples_per_second": 1311.58,
      "eval_steps_per_second": 163.956,
      "step": 12000
    },
    {
      "entropy": 6.100390911102295,
      "epoch": 1.0491129948440094,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004896292699501677,
      "loss": 5.9324,
      "mean_token_accuracy": 0.14589439630508422,
      "num_tokens": 24323770.0,
      "step": 12005
    },
    {
      "entropy": 6.182142543792724,
      "epoch": 1.0495499431967141,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004896199164581564,
      "loss": 5.8569,
      "mean_token_accuracy": 0.15646701753139497,
      "num_tokens": 24333438.0,
      "step": 12010
    },
    {
      "entropy": 6.1544561862945555,
      "epoch": 1.0499868915494188,
      "grad_norm": 12.625,
      "learning_rate": 0.0004896105588496227,
      "loss": 5.9399,
      "mean_token_accuracy": 0.14933661073446275,
      "num_tokens": 24342929.0,
      "step": 12015
    },
    {
      "entropy": 6.16118106842041,
      "epoch": 1.0504238399021235,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004896011971247461,
      "loss": 5.9848,
      "mean_token_accuracy": 0.14823028296232224,
      "num_tokens": 24352617.0,
      "step": 12020
    },
    {
      "entropy": 6.17752161026001,
      "epoch": 1.0508607882548282,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004895918312837061,
      "loss": 5.9224,
      "mean_token_accuracy": 0.15364284068346024,
      "num_tokens": 24363940.0,
      "step": 12025
    },
    {
      "entropy": 6.245773077011108,
      "epoch": 1.051297736607533,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004895824613266822,
      "loss": 5.9765,
      "mean_token_accuracy": 0.14229566529393195,
      "num_tokens": 24373168.0,
      "step": 12030
    },
    {
      "entropy": 6.2633058547973635,
      "epoch": 1.0517346849602376,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004895730872538543,
      "loss": 5.9968,
      "mean_token_accuracy": 0.14618628546595575,
      "num_tokens": 24383067.0,
      "step": 12035
    },
    {
      "entropy": 6.203412818908691,
      "epoch": 1.0521716333129425,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004895637090654021,
      "loss": 5.9115,
      "mean_token_accuracy": 0.15422668531537057,
      "num_tokens": 24392734.0,
      "step": 12040
    },
    {
      "entropy": 6.227333211898804,
      "epoch": 1.0526085816656472,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004895543267615057,
      "loss": 5.9264,
      "mean_token_accuracy": 0.1483489379286766,
      "num_tokens": 24403970.0,
      "step": 12045
    },
    {
      "entropy": 6.178986930847168,
      "epoch": 1.0530455300183519,
      "grad_norm": 2.625,
      "learning_rate": 0.0004895449403423447,
      "loss": 5.9256,
      "mean_token_accuracy": 0.15066254287958145,
      "num_tokens": 24413614.0,
      "step": 12050
    },
    {
      "entropy": 6.123789644241333,
      "epoch": 1.0534824783710566,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004895355498080994,
      "loss": 5.8726,
      "mean_token_accuracy": 0.16127931773662568,
      "num_tokens": 24424237.0,
      "step": 12055
    },
    {
      "entropy": 6.142438364028931,
      "epoch": 1.0539194267237613,
      "grad_norm": 4.0,
      "learning_rate": 0.0004895261551589498,
      "loss": 5.8771,
      "mean_token_accuracy": 0.1540844053030014,
      "num_tokens": 24433220.0,
      "step": 12060
    },
    {
      "entropy": 6.242302989959716,
      "epoch": 1.054356375076466,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004895167563950763,
      "loss": 6.0254,
      "mean_token_accuracy": 0.14323251694440842,
      "num_tokens": 24444224.0,
      "step": 12065
    },
    {
      "entropy": 6.2590171813964846,
      "epoch": 1.0547933234291706,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004895073535166588,
      "loss": 6.0311,
      "mean_token_accuracy": 0.1430962413549423,
      "num_tokens": 24454431.0,
      "step": 12070
    },
    {
      "entropy": 6.243322563171387,
      "epoch": 1.0552302717818753,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004894979465238779,
      "loss": 5.9115,
      "mean_token_accuracy": 0.14977423772215842,
      "num_tokens": 24464006.0,
      "step": 12075
    },
    {
      "entropy": 6.174377155303955,
      "epoch": 1.05566722013458,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004894885354169141,
      "loss": 5.8648,
      "mean_token_accuracy": 0.14857657104730607,
      "num_tokens": 24472706.0,
      "step": 12080
    },
    {
      "entropy": 6.15708589553833,
      "epoch": 1.0561041684872847,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004894791201959477,
      "loss": 5.8885,
      "mean_token_accuracy": 0.1566796638071537,
      "num_tokens": 24482785.0,
      "step": 12085
    },
    {
      "entropy": 6.160139656066894,
      "epoch": 1.0565411168399894,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004894697008611594,
      "loss": 5.9754,
      "mean_token_accuracy": 0.14761830270290374,
      "num_tokens": 24493318.0,
      "step": 12090
    },
    {
      "entropy": 6.148003339767456,
      "epoch": 1.0569780651926943,
      "grad_norm": 3.75,
      "learning_rate": 0.0004894602774127299,
      "loss": 5.9403,
      "mean_token_accuracy": 0.1446126401424408,
      "num_tokens": 24503780.0,
      "step": 12095
    },
    {
      "entropy": 6.285747718811035,
      "epoch": 1.057415013545399,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004894508498508399,
      "loss": 5.9948,
      "mean_token_accuracy": 0.14292212575674057,
      "num_tokens": 24513905.0,
      "step": 12100
    },
    {
      "entropy": 6.241286563873291,
      "epoch": 1.0578519618981037,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004894414181756702,
      "loss": 5.9368,
      "mean_token_accuracy": 0.15671780258417128,
      "num_tokens": 24524194.0,
      "step": 12105
    },
    {
      "entropy": 6.186323499679565,
      "epoch": 1.0582889102508084,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004894319823874017,
      "loss": 5.9556,
      "mean_token_accuracy": 0.1467944286763668,
      "num_tokens": 24533456.0,
      "step": 12110
    },
    {
      "entropy": 6.2142415046691895,
      "epoch": 1.058725858603513,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004894225424862154,
      "loss": 5.9558,
      "mean_token_accuracy": 0.14688051193952562,
      "num_tokens": 24544191.0,
      "step": 12115
    },
    {
      "entropy": 6.14606819152832,
      "epoch": 1.0591628069562178,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004894130984722926,
      "loss": 5.8814,
      "mean_token_accuracy": 0.15194664970040322,
      "num_tokens": 24554353.0,
      "step": 12120
    },
    {
      "entropy": 6.147780084609986,
      "epoch": 1.0595997553089225,
      "grad_norm": 2.65625,
      "learning_rate": 0.000489403650345814,
      "loss": 5.98,
      "mean_token_accuracy": 0.14456124007701873,
      "num_tokens": 24564313.0,
      "step": 12125
    },
    {
      "entropy": 6.200004577636719,
      "epoch": 1.0600367036616272,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004893941981069612,
      "loss": 5.9647,
      "mean_token_accuracy": 0.1462851420044899,
      "num_tokens": 24573129.0,
      "step": 12130
    },
    {
      "entropy": 6.153363609313965,
      "epoch": 1.0604736520143319,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004893847417559152,
      "loss": 5.8485,
      "mean_token_accuracy": 0.1546036720275879,
      "num_tokens": 24582324.0,
      "step": 12135
    },
    {
      "entropy": 6.2165992736816404,
      "epoch": 1.0609106003670365,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004893752812928576,
      "loss": 5.9972,
      "mean_token_accuracy": 0.14059825390577316,
      "num_tokens": 24593360.0,
      "step": 12140
    },
    {
      "entropy": 6.2215382099151615,
      "epoch": 1.0613475487197412,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004893658167179697,
      "loss": 5.9317,
      "mean_token_accuracy": 0.15346293598413469,
      "num_tokens": 24603597.0,
      "step": 12145
    },
    {
      "entropy": 6.166983270645142,
      "epoch": 1.061784497072446,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004893563480314331,
      "loss": 5.8568,
      "mean_token_accuracy": 0.15244588255882263,
      "num_tokens": 24612426.0,
      "step": 12150
    },
    {
      "entropy": 6.158516788482666,
      "epoch": 1.0622214454251508,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004893468752334294,
      "loss": 5.8797,
      "mean_token_accuracy": 0.14777790904045104,
      "num_tokens": 24623031.0,
      "step": 12155
    },
    {
      "entropy": 6.157711505889893,
      "epoch": 1.0626583937778555,
      "grad_norm": 2.0,
      "learning_rate": 0.0004893373983241404,
      "loss": 5.9401,
      "mean_token_accuracy": 0.1478666692972183,
      "num_tokens": 24633763.0,
      "step": 12160
    },
    {
      "entropy": 6.1805846214294435,
      "epoch": 1.0630953421305602,
      "grad_norm": 2.625,
      "learning_rate": 0.0004893279173037478,
      "loss": 5.9689,
      "mean_token_accuracy": 0.14446838200092316,
      "num_tokens": 24644340.0,
      "step": 12165
    },
    {
      "entropy": 6.208295249938965,
      "epoch": 1.063532290483265,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004893184321724335,
      "loss": 5.9607,
      "mean_token_accuracy": 0.15800270587205886,
      "num_tokens": 24654751.0,
      "step": 12170
    },
    {
      "entropy": 6.2193244934082035,
      "epoch": 1.0639692388359696,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004893089429303792,
      "loss": 5.9103,
      "mean_token_accuracy": 0.15098214745521546,
      "num_tokens": 24665134.0,
      "step": 12175
    },
    {
      "entropy": 6.155243730545044,
      "epoch": 1.0644061871886743,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004892994495777672,
      "loss": 5.9023,
      "mean_token_accuracy": 0.14569930657744407,
      "num_tokens": 24673796.0,
      "step": 12180
    },
    {
      "entropy": 6.106982469558716,
      "epoch": 1.064843135541379,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004892899521147794,
      "loss": 5.8513,
      "mean_token_accuracy": 0.16166339367628096,
      "num_tokens": 24684104.0,
      "step": 12185
    },
    {
      "entropy": 6.0830785751342775,
      "epoch": 1.0652800838940837,
      "grad_norm": 2.40625,
      "learning_rate": 0.000489280450541598,
      "loss": 5.8652,
      "mean_token_accuracy": 0.1524901419878006,
      "num_tokens": 24693994.0,
      "step": 12190
    },
    {
      "entropy": 6.155972719192505,
      "epoch": 1.0657170322467884,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004892709448584054,
      "loss": 5.9318,
      "mean_token_accuracy": 0.14918644726276398,
      "num_tokens": 24703815.0,
      "step": 12195
    },
    {
      "entropy": 6.23654956817627,
      "epoch": 1.066153980599493,
      "grad_norm": 2.5,
      "learning_rate": 0.0004892614350653837,
      "loss": 5.8798,
      "mean_token_accuracy": 0.15056822299957276,
      "num_tokens": 24713631.0,
      "step": 12200
    },
    {
      "entropy": 6.240795707702636,
      "epoch": 1.0665909289521978,
      "grad_norm": 2.125,
      "learning_rate": 0.0004892519211627156,
      "loss": 5.9281,
      "mean_token_accuracy": 0.15143959522247313,
      "num_tokens": 24723353.0,
      "step": 12205
    },
    {
      "entropy": 6.17115216255188,
      "epoch": 1.0670278773049025,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004892424031505832,
      "loss": 5.9122,
      "mean_token_accuracy": 0.14854995906352997,
      "num_tokens": 24733246.0,
      "step": 12210
    },
    {
      "entropy": 6.153678703308105,
      "epoch": 1.0674648256576074,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004892328810291693,
      "loss": 5.8932,
      "mean_token_accuracy": 0.1565846860408783,
      "num_tokens": 24743097.0,
      "step": 12215
    },
    {
      "entropy": 6.109102964401245,
      "epoch": 1.067901774010312,
      "grad_norm": 3.25,
      "learning_rate": 0.0004892233547986564,
      "loss": 5.8664,
      "mean_token_accuracy": 0.15516779869794844,
      "num_tokens": 24752618.0,
      "step": 12220
    },
    {
      "entropy": 6.216071701049804,
      "epoch": 1.0683387223630167,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004892138244592274,
      "loss": 5.9443,
      "mean_token_accuracy": 0.15100240260362624,
      "num_tokens": 24763960.0,
      "step": 12225
    },
    {
      "entropy": 6.182392311096192,
      "epoch": 1.0687756707157214,
      "grad_norm": 5.46875,
      "learning_rate": 0.0004892042900110649,
      "loss": 5.8288,
      "mean_token_accuracy": 0.15177802741527557,
      "num_tokens": 24773040.0,
      "step": 12230
    },
    {
      "entropy": 6.1401246070861815,
      "epoch": 1.0692126190684261,
      "grad_norm": 2.9375,
      "learning_rate": 0.000489194751454352,
      "loss": 5.9465,
      "mean_token_accuracy": 0.14840585142374038,
      "num_tokens": 24783049.0,
      "step": 12235
    },
    {
      "entropy": 6.225646018981934,
      "epoch": 1.0696495674211308,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004891852087892714,
      "loss": 5.9229,
      "mean_token_accuracy": 0.14651037156581878,
      "num_tokens": 24794590.0,
      "step": 12240
    },
    {
      "entropy": 6.158383131027222,
      "epoch": 1.0700865157738355,
      "grad_norm": 2.75,
      "learning_rate": 0.0004891756620160064,
      "loss": 5.8439,
      "mean_token_accuracy": 0.15536906868219375,
      "num_tokens": 24805146.0,
      "step": 12245
    },
    {
      "entropy": 6.129940366744995,
      "epoch": 1.0705234641265402,
      "grad_norm": 4.375,
      "learning_rate": 0.0004891661111347399,
      "loss": 5.9397,
      "mean_token_accuracy": 0.14204553216695787,
      "num_tokens": 24815573.0,
      "step": 12250
    },
    {
      "entropy": 6.173180103302002,
      "epoch": 1.070960412479245,
      "grad_norm": 6.375,
      "learning_rate": 0.0004891565561456552,
      "loss": 5.8587,
      "mean_token_accuracy": 0.15503737926483155,
      "num_tokens": 24825417.0,
      "step": 12255
    },
    {
      "entropy": 6.095524835586548,
      "epoch": 1.0713973608319496,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004891469970489355,
      "loss": 5.8864,
      "mean_token_accuracy": 0.15385202467441558,
      "num_tokens": 24835589.0,
      "step": 12260
    },
    {
      "entropy": 6.116691255569458,
      "epoch": 1.0718343091846543,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004891374338447642,
      "loss": 5.9988,
      "mean_token_accuracy": 0.1440843388438225,
      "num_tokens": 24846603.0,
      "step": 12265
    },
    {
      "entropy": 6.219188451766968,
      "epoch": 1.0722712575373592,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004891278665333247,
      "loss": 5.9095,
      "mean_token_accuracy": 0.14616507589817046,
      "num_tokens": 24856368.0,
      "step": 12270
    },
    {
      "entropy": 6.1318151473999025,
      "epoch": 1.0727082058900639,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004891182951148005,
      "loss": 5.8091,
      "mean_token_accuracy": 0.15435133129358292,
      "num_tokens": 24866322.0,
      "step": 12275
    },
    {
      "entropy": 6.134053373336792,
      "epoch": 1.0731451542427686,
      "grad_norm": 5.0625,
      "learning_rate": 0.0004891087195893753,
      "loss": 5.9351,
      "mean_token_accuracy": 0.14491613283753396,
      "num_tokens": 24876686.0,
      "step": 12280
    },
    {
      "entropy": 6.1039286136627195,
      "epoch": 1.0735821025954733,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004890991399572325,
      "loss": 5.8914,
      "mean_token_accuracy": 0.15078727751970292,
      "num_tokens": 24886561.0,
      "step": 12285
    },
    {
      "entropy": 6.209589958190918,
      "epoch": 1.074019050948178,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004890895562185562,
      "loss": 5.9077,
      "mean_token_accuracy": 0.14919467717409135,
      "num_tokens": 24896010.0,
      "step": 12290
    },
    {
      "entropy": 6.2686793327331545,
      "epoch": 1.0744559993008826,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004890799683735299,
      "loss": 5.9375,
      "mean_token_accuracy": 0.14493292868137359,
      "num_tokens": 24904852.0,
      "step": 12295
    },
    {
      "entropy": 6.19042387008667,
      "epoch": 1.0748929476535873,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004890703764223377,
      "loss": 5.9294,
      "mean_token_accuracy": 0.1475605010986328,
      "num_tokens": 24914828.0,
      "step": 12300
    },
    {
      "entropy": 6.180906343460083,
      "epoch": 1.075329896006292,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004890607803651635,
      "loss": 5.9805,
      "mean_token_accuracy": 0.1379154935479164,
      "num_tokens": 24925602.0,
      "step": 12305
    },
    {
      "entropy": 6.17429141998291,
      "epoch": 1.0757668443589967,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004890511802021914,
      "loss": 5.9529,
      "mean_token_accuracy": 0.15566674172878264,
      "num_tokens": 24935175.0,
      "step": 12310
    },
    {
      "entropy": 6.208604955673218,
      "epoch": 1.0762037927117014,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004890415759336053,
      "loss": 5.9589,
      "mean_token_accuracy": 0.14076291993260384,
      "num_tokens": 24945828.0,
      "step": 12315
    },
    {
      "entropy": 6.209710884094238,
      "epoch": 1.076640741064406,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004890319675595899,
      "loss": 5.907,
      "mean_token_accuracy": 0.15007260888814927,
      "num_tokens": 24955400.0,
      "step": 12320
    },
    {
      "entropy": 6.172647190093994,
      "epoch": 1.0770776894171108,
      "grad_norm": 3.0625,
      "learning_rate": 0.000489022355080329,
      "loss": 5.972,
      "mean_token_accuracy": 0.14173820465803147,
      "num_tokens": 24966079.0,
      "step": 12325
    },
    {
      "entropy": 6.148740291595459,
      "epoch": 1.0775146377698157,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004890127384960073,
      "loss": 5.8579,
      "mean_token_accuracy": 0.15463101342320443,
      "num_tokens": 24975090.0,
      "step": 12330
    },
    {
      "entropy": 6.265727901458741,
      "epoch": 1.0779515861225204,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004890031178068091,
      "loss": 6.0046,
      "mean_token_accuracy": 0.1382139004766941,
      "num_tokens": 24986068.0,
      "step": 12335
    },
    {
      "entropy": 6.197646760940552,
      "epoch": 1.078388534475225,
      "grad_norm": 3.96875,
      "learning_rate": 0.000488993493012919,
      "loss": 5.9696,
      "mean_token_accuracy": 0.15166736245155335,
      "num_tokens": 24995697.0,
      "step": 12340
    },
    {
      "entropy": 6.169878768920898,
      "epoch": 1.0788254828279298,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004889838641145215,
      "loss": 5.9155,
      "mean_token_accuracy": 0.14484099447727203,
      "num_tokens": 25005602.0,
      "step": 12345
    },
    {
      "entropy": 6.252286529541015,
      "epoch": 1.0792624311806345,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004889742311118014,
      "loss": 5.9668,
      "mean_token_accuracy": 0.15457540899515151,
      "num_tokens": 25016407.0,
      "step": 12350
    },
    {
      "entropy": 6.308634328842163,
      "epoch": 1.0796993795333392,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004889645940049434,
      "loss": 5.982,
      "mean_token_accuracy": 0.14494881108403207,
      "num_tokens": 25025889.0,
      "step": 12355
    },
    {
      "entropy": 6.186164379119873,
      "epoch": 1.0801363278860439,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004889549527941324,
      "loss": 5.9301,
      "mean_token_accuracy": 0.15099339187145233,
      "num_tokens": 25036383.0,
      "step": 12360
    },
    {
      "entropy": 6.186786317825318,
      "epoch": 1.0805732762387485,
      "grad_norm": 4.125,
      "learning_rate": 0.0004889453074795534,
      "loss": 6.0263,
      "mean_token_accuracy": 0.13621387705206872,
      "num_tokens": 25045718.0,
      "step": 12365
    },
    {
      "entropy": 6.19301700592041,
      "epoch": 1.0810102245914532,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004889356580613911,
      "loss": 6.007,
      "mean_token_accuracy": 0.14771054461598396,
      "num_tokens": 25056244.0,
      "step": 12370
    },
    {
      "entropy": 6.242445611953736,
      "epoch": 1.081447172944158,
      "grad_norm": 2.484375,
      "learning_rate": 0.000488926004539831,
      "loss": 5.9363,
      "mean_token_accuracy": 0.15063196271657944,
      "num_tokens": 25065954.0,
      "step": 12375
    },
    {
      "entropy": 6.198996686935425,
      "epoch": 1.0818841212968626,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004889163469150578,
      "loss": 5.9832,
      "mean_token_accuracy": 0.14517340958118438,
      "num_tokens": 25075851.0,
      "step": 12380
    },
    {
      "entropy": 6.1527872562408445,
      "epoch": 1.0823210696495673,
      "grad_norm": 2.25,
      "learning_rate": 0.0004889066851872571,
      "loss": 5.938,
      "mean_token_accuracy": 0.1523074522614479,
      "num_tokens": 25085324.0,
      "step": 12385
    },
    {
      "entropy": 6.178440189361572,
      "epoch": 1.0827580180022722,
      "grad_norm": 2.046875,
      "learning_rate": 0.000488897019356614,
      "loss": 5.9744,
      "mean_token_accuracy": 0.14243398308753968,
      "num_tokens": 25096566.0,
      "step": 12390
    },
    {
      "entropy": 6.274006700515747,
      "epoch": 1.083194966354977,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004888873494233141,
      "loss": 6.0449,
      "mean_token_accuracy": 0.13726802542805672,
      "num_tokens": 25107642.0,
      "step": 12395
    },
    {
      "entropy": 6.204235172271728,
      "epoch": 1.0836319147076816,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004888776753875427,
      "loss": 5.9265,
      "mean_token_accuracy": 0.15124789029359817,
      "num_tokens": 25117488.0,
      "step": 12400
    },
    {
      "entropy": 6.14057731628418,
      "epoch": 1.0840688630603863,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004888679972494854,
      "loss": 5.8783,
      "mean_token_accuracy": 0.15421270728111267,
      "num_tokens": 25127107.0,
      "step": 12405
    },
    {
      "entropy": 6.108871221542358,
      "epoch": 1.084505811413091,
      "grad_norm": 2.28125,
      "learning_rate": 0.000488858315009328,
      "loss": 5.9041,
      "mean_token_accuracy": 0.14985944032669068,
      "num_tokens": 25138272.0,
      "step": 12410
    },
    {
      "entropy": 6.175687646865844,
      "epoch": 1.0849427597657957,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004888486286672559,
      "loss": 5.9652,
      "mean_token_accuracy": 0.14317135363817216,
      "num_tokens": 25147128.0,
      "step": 12415
    },
    {
      "entropy": 6.2086694717407225,
      "epoch": 1.0853797081185004,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004888389382234551,
      "loss": 5.867,
      "mean_token_accuracy": 0.1534759894013405,
      "num_tokens": 25157418.0,
      "step": 12420
    },
    {
      "entropy": 6.211604595184326,
      "epoch": 1.085816656471205,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004888292436781114,
      "loss": 5.9332,
      "mean_token_accuracy": 0.15103035420179367,
      "num_tokens": 25168123.0,
      "step": 12425
    },
    {
      "entropy": 6.107608413696289,
      "epoch": 1.0862536048239098,
      "grad_norm": 2.125,
      "learning_rate": 0.0004888195450314108,
      "loss": 5.9168,
      "mean_token_accuracy": 0.1523303858935833,
      "num_tokens": 25178445.0,
      "step": 12430
    },
    {
      "entropy": 6.144627666473388,
      "epoch": 1.0866905531766144,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004888098422835392,
      "loss": 5.9651,
      "mean_token_accuracy": 0.1374790109694004,
      "num_tokens": 25189246.0,
      "step": 12435
    },
    {
      "entropy": 6.1797661781311035,
      "epoch": 1.0871275015293191,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004888001354346828,
      "loss": 5.9687,
      "mean_token_accuracy": 0.14563697203993797,
      "num_tokens": 25201277.0,
      "step": 12440
    },
    {
      "entropy": 6.2508142471313475,
      "epoch": 1.087564449882024,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004887904244850279,
      "loss": 5.9527,
      "mean_token_accuracy": 0.14970300942659379,
      "num_tokens": 25211353.0,
      "step": 12445
    },
    {
      "entropy": 6.230410814285278,
      "epoch": 1.0880013982347287,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004887807094347606,
      "loss": 5.9231,
      "mean_token_accuracy": 0.1498468391597271,
      "num_tokens": 25222236.0,
      "step": 12450
    },
    {
      "entropy": 6.127790260314941,
      "epoch": 1.0884383465874334,
      "grad_norm": 2.875,
      "learning_rate": 0.0004887709902840672,
      "loss": 5.8448,
      "mean_token_accuracy": 0.15379466414451598,
      "num_tokens": 25232464.0,
      "step": 12455
    },
    {
      "entropy": 6.154881811141967,
      "epoch": 1.0888752949401381,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004887612670331342,
      "loss": 5.9224,
      "mean_token_accuracy": 0.14510940238833428,
      "num_tokens": 25243518.0,
      "step": 12460
    },
    {
      "entropy": 6.16659517288208,
      "epoch": 1.0893122432928428,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004887515396821481,
      "loss": 5.9331,
      "mean_token_accuracy": 0.14443064704537392,
      "num_tokens": 25253168.0,
      "step": 12465
    },
    {
      "entropy": 6.178592395782471,
      "epoch": 1.0897491916455475,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004887418082312955,
      "loss": 5.8977,
      "mean_token_accuracy": 0.15176382958889006,
      "num_tokens": 25262750.0,
      "step": 12470
    },
    {
      "entropy": 6.173739385604859,
      "epoch": 1.0901861399982522,
      "grad_norm": 2.703125,
      "learning_rate": 0.000488732072680763,
      "loss": 5.9172,
      "mean_token_accuracy": 0.14840522408485413,
      "num_tokens": 25273520.0,
      "step": 12475
    },
    {
      "entropy": 6.192647075653076,
      "epoch": 1.090623088350957,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004887223330307373,
      "loss": 5.96,
      "mean_token_accuracy": 0.14730576574802398,
      "num_tokens": 25284136.0,
      "step": 12480
    },
    {
      "entropy": 6.162393379211426,
      "epoch": 1.0910600367036616,
      "grad_norm": 3.5,
      "learning_rate": 0.0004887125892814053,
      "loss": 5.8696,
      "mean_token_accuracy": 0.1461636930704117,
      "num_tokens": 25294523.0,
      "step": 12485
    },
    {
      "entropy": 6.145088958740234,
      "epoch": 1.0914969850563663,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004887028414329538,
      "loss": 5.9389,
      "mean_token_accuracy": 0.1463019907474518,
      "num_tokens": 25305050.0,
      "step": 12490
    },
    {
      "entropy": 6.197679042816162,
      "epoch": 1.091933933409071,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048869308948557,
      "loss": 5.9093,
      "mean_token_accuracy": 0.1485550343990326,
      "num_tokens": 25314445.0,
      "step": 12495
    },
    {
      "entropy": 6.189301538467407,
      "epoch": 1.0923708817617757,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004886833334394406,
      "loss": 5.9182,
      "mean_token_accuracy": 0.1524156928062439,
      "num_tokens": 25324426.0,
      "step": 12500
    },
    {
      "entropy": 6.187959003448486,
      "epoch": 1.0928078301144806,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004886735732947529,
      "loss": 5.9239,
      "mean_token_accuracy": 0.15090341120958328,
      "num_tokens": 25335051.0,
      "step": 12505
    },
    {
      "entropy": 6.169259023666382,
      "epoch": 1.0932447784671853,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004886638090516942,
      "loss": 5.8399,
      "mean_token_accuracy": 0.15351396203041076,
      "num_tokens": 25343955.0,
      "step": 12510
    },
    {
      "entropy": 6.115030527114868,
      "epoch": 1.09368172681989,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004886540407104517,
      "loss": 5.958,
      "mean_token_accuracy": 0.14472192227840425,
      "num_tokens": 25353817.0,
      "step": 12515
    },
    {
      "entropy": 6.239952564239502,
      "epoch": 1.0941186751725946,
      "grad_norm": 2.625,
      "learning_rate": 0.0004886442682712126,
      "loss": 5.87,
      "mean_token_accuracy": 0.14718974381685257,
      "num_tokens": 25363882.0,
      "step": 12520
    },
    {
      "entropy": 6.227929401397705,
      "epoch": 1.0945556235252993,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004886344917341646,
      "loss": 5.8893,
      "mean_token_accuracy": 0.15139144361019136,
      "num_tokens": 25373721.0,
      "step": 12525
    },
    {
      "entropy": 6.140818977355957,
      "epoch": 1.094992571878004,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004886247110994952,
      "loss": 5.9173,
      "mean_token_accuracy": 0.149771024286747,
      "num_tokens": 25383520.0,
      "step": 12530
    },
    {
      "entropy": 6.140714836120606,
      "epoch": 1.0954295202307087,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004886149263673916,
      "loss": 5.8963,
      "mean_token_accuracy": 0.15616434812545776,
      "num_tokens": 25394039.0,
      "step": 12535
    },
    {
      "entropy": 6.17752161026001,
      "epoch": 1.0958664685834134,
      "grad_norm": 2.390625,
      "learning_rate": 0.000488605137538042,
      "loss": 5.9279,
      "mean_token_accuracy": 0.14835054725408553,
      "num_tokens": 25403404.0,
      "step": 12540
    },
    {
      "entropy": 6.11985149383545,
      "epoch": 1.096303416936118,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004885953446116339,
      "loss": 5.9226,
      "mean_token_accuracy": 0.14923788011074066,
      "num_tokens": 25413800.0,
      "step": 12545
    },
    {
      "entropy": 6.14561448097229,
      "epoch": 1.0967403652888228,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004885855475883551,
      "loss": 5.8656,
      "mean_token_accuracy": 0.15641434043645858,
      "num_tokens": 25424252.0,
      "step": 12550
    },
    {
      "entropy": 6.235609436035157,
      "epoch": 1.0971773136415275,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004885757464683937,
      "loss": 5.8802,
      "mean_token_accuracy": 0.1519485369324684,
      "num_tokens": 25433381.0,
      "step": 12555
    },
    {
      "entropy": 6.1403765201568605,
      "epoch": 1.0976142619942322,
      "grad_norm": 2.5,
      "learning_rate": 0.0004885659412519376,
      "loss": 5.8595,
      "mean_token_accuracy": 0.15306203961372375,
      "num_tokens": 25443253.0,
      "step": 12560
    },
    {
      "entropy": 6.111497020721435,
      "epoch": 1.098051210346937,
      "grad_norm": 7.78125,
      "learning_rate": 0.0004885561319391747,
      "loss": 5.9622,
      "mean_token_accuracy": 0.14385128617286683,
      "num_tokens": 25453590.0,
      "step": 12565
    },
    {
      "entropy": 6.168994331359864,
      "epoch": 1.0984881586996418,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004885463185302934,
      "loss": 5.9074,
      "mean_token_accuracy": 0.15635205805301666,
      "num_tokens": 25464065.0,
      "step": 12570
    },
    {
      "entropy": 6.264686918258667,
      "epoch": 1.0989251070523465,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004885365010254817,
      "loss": 6.0071,
      "mean_token_accuracy": 0.1437750354409218,
      "num_tokens": 25475236.0,
      "step": 12575
    },
    {
      "entropy": 6.2338732242584225,
      "epoch": 1.0993620554050512,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004885266794249281,
      "loss": 5.9851,
      "mean_token_accuracy": 0.1526201382279396,
      "num_tokens": 25485200.0,
      "step": 12580
    },
    {
      "entropy": 6.190126037597656,
      "epoch": 1.0997990037577559,
      "grad_norm": 2.5,
      "learning_rate": 0.0004885168537288209,
      "loss": 5.9362,
      "mean_token_accuracy": 0.15273040235042573,
      "num_tokens": 25494775.0,
      "step": 12585
    },
    {
      "entropy": 6.228781986236572,
      "epoch": 1.1002359521104605,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004885070239373484,
      "loss": 5.9358,
      "mean_token_accuracy": 0.14743367880582808,
      "num_tokens": 25504435.0,
      "step": 12590
    },
    {
      "entropy": 6.1857123374938965,
      "epoch": 1.1006729004631652,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004884971900506995,
      "loss": 5.941,
      "mean_token_accuracy": 0.1494644284248352,
      "num_tokens": 25514085.0,
      "step": 12595
    },
    {
      "entropy": 6.25676736831665,
      "epoch": 1.10110984881587,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004884873520690626,
      "loss": 5.9814,
      "mean_token_accuracy": 0.14542241394519806,
      "num_tokens": 25523988.0,
      "step": 12600
    },
    {
      "entropy": 6.197001504898071,
      "epoch": 1.1015467971685746,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004884775099926264,
      "loss": 5.9671,
      "mean_token_accuracy": 0.1485779508948326,
      "num_tokens": 25534469.0,
      "step": 12605
    },
    {
      "entropy": 6.212598848342895,
      "epoch": 1.1019837455212793,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004884676638215797,
      "loss": 6.0116,
      "mean_token_accuracy": 0.14274339526891708,
      "num_tokens": 25544854.0,
      "step": 12610
    },
    {
      "entropy": 6.2349677085876465,
      "epoch": 1.102420693873984,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004884578135561113,
      "loss": 5.9017,
      "mean_token_accuracy": 0.1462556853890419,
      "num_tokens": 25555465.0,
      "step": 12615
    },
    {
      "entropy": 6.174994707107544,
      "epoch": 1.102857642226689,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004884479591964103,
      "loss": 5.9508,
      "mean_token_accuracy": 0.15059880614280702,
      "num_tokens": 25565949.0,
      "step": 12620
    },
    {
      "entropy": 6.192708539962768,
      "epoch": 1.1032945905793936,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004884381007426654,
      "loss": 5.9354,
      "mean_token_accuracy": 0.14404048919677734,
      "num_tokens": 25576511.0,
      "step": 12625
    },
    {
      "entropy": 6.182388114929199,
      "epoch": 1.1037315389320983,
      "grad_norm": 2.375,
      "learning_rate": 0.0004884282381950661,
      "loss": 5.8686,
      "mean_token_accuracy": 0.1456463098526001,
      "num_tokens": 25587224.0,
      "step": 12630
    },
    {
      "entropy": 6.172303533554077,
      "epoch": 1.104168487284803,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004884183715538012,
      "loss": 5.9054,
      "mean_token_accuracy": 0.15106040462851525,
      "num_tokens": 25596696.0,
      "step": 12635
    },
    {
      "entropy": 6.1065895557403564,
      "epoch": 1.1046054356375077,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004884085008190602,
      "loss": 5.8387,
      "mean_token_accuracy": 0.1551930382847786,
      "num_tokens": 25606829.0,
      "step": 12640
    },
    {
      "entropy": 6.141040468215943,
      "epoch": 1.1050423839902124,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004883986259910323,
      "loss": 5.9142,
      "mean_token_accuracy": 0.15261392295360565,
      "num_tokens": 25617122.0,
      "step": 12645
    },
    {
      "entropy": 6.185095930099488,
      "epoch": 1.105479332342917,
      "grad_norm": 2.875,
      "learning_rate": 0.0004883887470699069,
      "loss": 5.8853,
      "mean_token_accuracy": 0.1480973869562149,
      "num_tokens": 25626485.0,
      "step": 12650
    },
    {
      "entropy": 6.163683032989502,
      "epoch": 1.1059162806956218,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004883788640558735,
      "loss": 5.8924,
      "mean_token_accuracy": 0.15037520676851274,
      "num_tokens": 25636695.0,
      "step": 12655
    },
    {
      "entropy": 6.015004873275757,
      "epoch": 1.1063532290483264,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004883689769491218,
      "loss": 5.768,
      "mean_token_accuracy": 0.16217269003391266,
      "num_tokens": 25646169.0,
      "step": 12660
    },
    {
      "entropy": 6.142358684539795,
      "epoch": 1.1067901774010311,
      "grad_norm": 3.125,
      "learning_rate": 0.0004883590857498413,
      "loss": 5.9635,
      "mean_token_accuracy": 0.1473284512758255,
      "num_tokens": 25656728.0,
      "step": 12665
    },
    {
      "entropy": 6.22428331375122,
      "epoch": 1.1072271257537358,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004883491904582217,
      "loss": 5.8926,
      "mean_token_accuracy": 0.15063771679997445,
      "num_tokens": 25666184.0,
      "step": 12670
    },
    {
      "entropy": 6.219676780700683,
      "epoch": 1.1076640741064405,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004883392910744528,
      "loss": 5.9002,
      "mean_token_accuracy": 0.15376341938972474,
      "num_tokens": 25675571.0,
      "step": 12675
    },
    {
      "entropy": 6.211104822158814,
      "epoch": 1.1081010224591452,
      "grad_norm": 5.78125,
      "learning_rate": 0.0004883293875987245,
      "loss": 5.9353,
      "mean_token_accuracy": 0.1464277945458889,
      "num_tokens": 25685485.0,
      "step": 12680
    },
    {
      "entropy": 6.087258625030517,
      "epoch": 1.1085379708118501,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004883194800312268,
      "loss": 5.8646,
      "mean_token_accuracy": 0.15398255437612535,
      "num_tokens": 25695278.0,
      "step": 12685
    },
    {
      "entropy": 6.1708001613616945,
      "epoch": 1.1089749191645548,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004883095683721497,
      "loss": 5.9207,
      "mean_token_accuracy": 0.15454749017953873,
      "num_tokens": 25704410.0,
      "step": 12690
    },
    {
      "entropy": 6.186515378952026,
      "epoch": 1.1094118675172595,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004882996526216833,
      "loss": 5.9206,
      "mean_token_accuracy": 0.14819520488381385,
      "num_tokens": 25715027.0,
      "step": 12695
    },
    {
      "entropy": 6.228687143325805,
      "epoch": 1.1098488158699642,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004882897327800178,
      "loss": 5.9518,
      "mean_token_accuracy": 0.1518068417906761,
      "num_tokens": 25723992.0,
      "step": 12700
    },
    {
      "entropy": 6.212923622131347,
      "epoch": 1.110285764222669,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004882798088473435,
      "loss": 5.912,
      "mean_token_accuracy": 0.1467521905899048,
      "num_tokens": 25733791.0,
      "step": 12705
    },
    {
      "entropy": 6.159442901611328,
      "epoch": 1.1107227125753736,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048826988082385065,
      "loss": 5.9207,
      "mean_token_accuracy": 0.14514946714043617,
      "num_tokens": 25743258.0,
      "step": 12710
    },
    {
      "entropy": 6.160631990432739,
      "epoch": 1.1111596609280783,
      "grad_norm": 2.125,
      "learning_rate": 0.00048825994870972976,
      "loss": 5.9438,
      "mean_token_accuracy": 0.14972086474299431,
      "num_tokens": 25753183.0,
      "step": 12715
    },
    {
      "entropy": 6.220805501937866,
      "epoch": 1.111596609280783,
      "grad_norm": 3.453125,
      "learning_rate": 0.00048825001250517124,
      "loss": 5.9722,
      "mean_token_accuracy": 0.14766730591654778,
      "num_tokens": 25763308.0,
      "step": 12720
    },
    {
      "entropy": 6.199866914749146,
      "epoch": 1.1120335576334877,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004882400722103658,
      "loss": 5.8527,
      "mean_token_accuracy": 0.1589813455939293,
      "num_tokens": 25774269.0,
      "step": 12725
    },
    {
      "entropy": 6.055586147308349,
      "epoch": 1.1124705059861923,
      "grad_norm": 3.03125,
      "learning_rate": 0.000488230127825504,
      "loss": 5.8731,
      "mean_token_accuracy": 0.15525724738836288,
      "num_tokens": 25785194.0,
      "step": 12730
    },
    {
      "entropy": 6.151578140258789,
      "epoch": 1.112907454338897,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048822017935077673,
      "loss": 5.8963,
      "mean_token_accuracy": 0.15453674048185348,
      "num_tokens": 25793927.0,
      "step": 12735
    },
    {
      "entropy": 6.212102222442627,
      "epoch": 1.113344402691602,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004882102267863746,
      "loss": 5.9591,
      "mean_token_accuracy": 0.14770420044660568,
      "num_tokens": 25805395.0,
      "step": 12740
    },
    {
      "entropy": 6.214001703262329,
      "epoch": 1.1137813510443066,
      "grad_norm": 1.984375,
      "learning_rate": 0.00048820027013248864,
      "loss": 6.0237,
      "mean_token_accuracy": 0.1472778335213661,
      "num_tokens": 25816594.0,
      "step": 12745
    },
    {
      "entropy": 6.243660402297974,
      "epoch": 1.1142182993970113,
      "grad_norm": 1.84375,
      "learning_rate": 0.00048819030938930974,
      "loss": 5.8968,
      "mean_token_accuracy": 0.15108416527509688,
      "num_tokens": 25825943.0,
      "step": 12750
    },
    {
      "entropy": 6.138241147994995,
      "epoch": 1.114655247749716,
      "grad_norm": 2.203125,
      "learning_rate": 0.000488180344557029,
      "loss": 5.8937,
      "mean_token_accuracy": 0.15438546389341354,
      "num_tokens": 25835928.0,
      "step": 12755
    },
    {
      "entropy": 6.1140360832214355,
      "epoch": 1.1150921961024207,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004881703756358376,
      "loss": 5.9289,
      "mean_token_accuracy": 0.14904947578907013,
      "num_tokens": 25847363.0,
      "step": 12760
    },
    {
      "entropy": 6.161057662963867,
      "epoch": 1.1155291444551254,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004881604026259266,
      "loss": 5.9381,
      "mean_token_accuracy": 0.1508537843823433,
      "num_tokens": 25857246.0,
      "step": 12765
    },
    {
      "entropy": 6.285070276260376,
      "epoch": 1.11596609280783,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048815042552748745,
      "loss": 5.952,
      "mean_token_accuracy": 0.1469726085662842,
      "num_tokens": 25866615.0,
      "step": 12770
    },
    {
      "entropy": 6.225005435943603,
      "epoch": 1.1164030411605348,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004881404443407115,
      "loss": 6.0025,
      "mean_token_accuracy": 0.14359909892082215,
      "num_tokens": 25876698.0,
      "step": 12775
    },
    {
      "entropy": 6.239930868148804,
      "epoch": 1.1168399895132395,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048813045906579,
      "loss": 5.9767,
      "mean_token_accuracy": 0.1381794475018978,
      "num_tokens": 25888029.0,
      "step": 12780
    },
    {
      "entropy": 6.168526124954224,
      "epoch": 1.1172769378659442,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048812046970291473,
      "loss": 6.0075,
      "mean_token_accuracy": 0.14545436352491378,
      "num_tokens": 25897472.0,
      "step": 12785
    },
    {
      "entropy": 6.277098894119263,
      "epoch": 1.1177138862186489,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004881104762522772,
      "loss": 5.967,
      "mean_token_accuracy": 0.14485169649124147,
      "num_tokens": 25908025.0,
      "step": 12790
    },
    {
      "entropy": 6.309070301055908,
      "epoch": 1.1181508345713538,
      "grad_norm": 2.515625,
      "learning_rate": 0.000488100478714069,
      "loss": 5.949,
      "mean_token_accuracy": 0.1473851904273033,
      "num_tokens": 25919110.0,
      "step": 12795
    },
    {
      "entropy": 6.157921648025512,
      "epoch": 1.1185877829240585,
      "grad_norm": 3.5625,
      "learning_rate": 0.00048809047708848194,
      "loss": 5.8712,
      "mean_token_accuracy": 0.15649706423282622,
      "num_tokens": 25928322.0,
      "step": 12800
    },
    {
      "entropy": 6.076091861724853,
      "epoch": 1.1190247312767632,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048808047137570795,
      "loss": 5.8379,
      "mean_token_accuracy": 0.14862755611538886,
      "num_tokens": 25937347.0,
      "step": 12805
    },
    {
      "entropy": 6.091773843765258,
      "epoch": 1.1194616796294679,
      "grad_norm": 5.5,
      "learning_rate": 0.0004880704615759388,
      "loss": 5.8602,
      "mean_token_accuracy": 0.15211082249879837,
      "num_tokens": 25948408.0,
      "step": 12810
    },
    {
      "entropy": 6.198835706710815,
      "epoch": 1.1198986279821725,
      "grad_norm": 5.9375,
      "learning_rate": 0.00048806044768936655,
      "loss": 5.9569,
      "mean_token_accuracy": 0.14384416937828065,
      "num_tokens": 25958249.0,
      "step": 12815
    },
    {
      "entropy": 6.22984299659729,
      "epoch": 1.1203355763348772,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004880504297161833,
      "loss": 5.9729,
      "mean_token_accuracy": 0.1435294672846794,
      "num_tokens": 25968849.0,
      "step": 12820
    },
    {
      "entropy": 6.186871814727783,
      "epoch": 1.120772524687582,
      "grad_norm": 8.1875,
      "learning_rate": 0.00048804040765658115,
      "loss": 5.9722,
      "mean_token_accuracy": 0.149842718988657,
      "num_tokens": 25978787.0,
      "step": 12825
    },
    {
      "entropy": 6.126379680633545,
      "epoch": 1.1212094730402866,
      "grad_norm": 1.8125,
      "learning_rate": 0.00048803038151075227,
      "loss": 5.9149,
      "mean_token_accuracy": 0.15154345333576202,
      "num_tokens": 25988619.0,
      "step": 12830
    },
    {
      "entropy": 6.208881092071533,
      "epoch": 1.1216464213929913,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004880203512788892,
      "loss": 5.9417,
      "mean_token_accuracy": 0.1463656470179558,
      "num_tokens": 25999080.0,
      "step": 12835
    },
    {
      "entropy": 6.21388988494873,
      "epoch": 1.122083369745696,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048801031696118406,
      "loss": 5.8685,
      "mean_token_accuracy": 0.1544377714395523,
      "num_tokens": 26010329.0,
      "step": 12840
    },
    {
      "entropy": 6.149548673629761,
      "epoch": 1.1225203180984007,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004880002785578295,
      "loss": 5.8705,
      "mean_token_accuracy": 0.14981589317321778,
      "num_tokens": 26019147.0,
      "step": 12845
    },
    {
      "entropy": 6.128245210647583,
      "epoch": 1.1229572664511054,
      "grad_norm": 2.140625,
      "learning_rate": 0.000487990236069018,
      "loss": 5.9103,
      "mean_token_accuracy": 0.1495534211397171,
      "num_tokens": 26029212.0,
      "step": 12850
    },
    {
      "entropy": 6.188356161117554,
      "epoch": 1.12339421480381,
      "grad_norm": 2.625,
      "learning_rate": 0.00048798018949494214,
      "loss": 5.7797,
      "mean_token_accuracy": 0.1502614289522171,
      "num_tokens": 26038059.0,
      "step": 12855
    },
    {
      "entropy": 6.195520257949829,
      "epoch": 1.123831163156515,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004879701388357946,
      "loss": 6.0213,
      "mean_token_accuracy": 0.139035477489233,
      "num_tokens": 26048076.0,
      "step": 12860
    },
    {
      "entropy": 6.141154003143311,
      "epoch": 1.1242681115092197,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004879600840917683,
      "loss": 5.9288,
      "mean_token_accuracy": 0.14810874313116074,
      "num_tokens": 26059006.0,
      "step": 12865
    },
    {
      "entropy": 6.15558557510376,
      "epoch": 1.1247050598619244,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004879500252630559,
      "loss": 5.8715,
      "mean_token_accuracy": 0.1544817343354225,
      "num_tokens": 26069275.0,
      "step": 12870
    },
    {
      "entropy": 6.182112789154052,
      "epoch": 1.125142008214629,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004879399623498504,
      "loss": 5.8542,
      "mean_token_accuracy": 0.15223054438829423,
      "num_tokens": 26078869.0,
      "step": 12875
    },
    {
      "entropy": 6.172968101501465,
      "epoch": 1.1255789565673338,
      "grad_norm": 4.09375,
      "learning_rate": 0.000487929895352345,
      "loss": 5.8624,
      "mean_token_accuracy": 0.14759116023778915,
      "num_tokens": 26088544.0,
      "step": 12880
    },
    {
      "entropy": 6.0676685810089115,
      "epoch": 1.1260159049200384,
      "grad_norm": 2.25,
      "learning_rate": 0.0004879198242707325,
      "loss": 5.836,
      "mean_token_accuracy": 0.1578221946954727,
      "num_tokens": 26097503.0,
      "step": 12885
    },
    {
      "entropy": 6.253616619110107,
      "epoch": 1.1264528532727431,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048790974910520627,
      "loss": 5.9791,
      "mean_token_accuracy": 0.1464863285422325,
      "num_tokens": 26107228.0,
      "step": 12890
    },
    {
      "entropy": 6.182883834838867,
      "epoch": 1.1268898016254478,
      "grad_norm": 2.8125,
      "learning_rate": 0.00048789966985595946,
      "loss": 5.9061,
      "mean_token_accuracy": 0.15523041635751725,
      "num_tokens": 26117633.0,
      "step": 12895
    },
    {
      "entropy": 6.113433408737182,
      "epoch": 1.1273267499781525,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004878895865231854,
      "loss": 5.9354,
      "mean_token_accuracy": 0.14570484161376954,
      "num_tokens": 26127781.0,
      "step": 12900
    },
    {
      "entropy": 6.192937850952148,
      "epoch": 1.1277636983308572,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004878794991070776,
      "loss": 5.8766,
      "mean_token_accuracy": 0.15598804056644439,
      "num_tokens": 26137675.0,
      "step": 12905
    },
    {
      "entropy": 6.211550092697143,
      "epoch": 1.128200646683562,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048786940760782937,
      "loss": 6.0002,
      "mean_token_accuracy": 0.14600116685032843,
      "num_tokens": 26148572.0,
      "step": 12910
    },
    {
      "entropy": 6.159525442123413,
      "epoch": 1.1286375950362668,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048785931202563443,
      "loss": 5.9757,
      "mean_token_accuracy": 0.15151511877775192,
      "num_tokens": 26159261.0,
      "step": 12915
    },
    {
      "entropy": 6.161359024047852,
      "epoch": 1.1290745433889715,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004878492123606863,
      "loss": 5.9543,
      "mean_token_accuracy": 0.14990262165665627,
      "num_tokens": 26170072.0,
      "step": 12920
    },
    {
      "entropy": 6.203384733200073,
      "epoch": 1.1295114917416762,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048783910861317873,
      "loss": 5.8786,
      "mean_token_accuracy": 0.14958043470978738,
      "num_tokens": 26179947.0,
      "step": 12925
    },
    {
      "entropy": 6.249737024307251,
      "epoch": 1.129948440094381,
      "grad_norm": 2.75,
      "learning_rate": 0.00048782900078330556,
      "loss": 5.9246,
      "mean_token_accuracy": 0.14690151065587997,
      "num_tokens": 26189982.0,
      "step": 12930
    },
    {
      "entropy": 6.246534252166748,
      "epoch": 1.1303853884470856,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004878188888712606,
      "loss": 6.0259,
      "mean_token_accuracy": 0.137100001424551,
      "num_tokens": 26200832.0,
      "step": 12935
    },
    {
      "entropy": 6.14182710647583,
      "epoch": 1.1308223367997903,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004878087728772379,
      "loss": 5.9764,
      "mean_token_accuracy": 0.14586392045021057,
      "num_tokens": 26211506.0,
      "step": 12940
    },
    {
      "entropy": 6.121871566772461,
      "epoch": 1.131259285152495,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004877986528014314,
      "loss": 5.8837,
      "mean_token_accuracy": 0.14971708208322526,
      "num_tokens": 26222045.0,
      "step": 12945
    },
    {
      "entropy": 6.124219799041748,
      "epoch": 1.1316962335051997,
      "grad_norm": 2.125,
      "learning_rate": 0.0004877885286440353,
      "loss": 5.8293,
      "mean_token_accuracy": 0.15299318581819535,
      "num_tokens": 26231411.0,
      "step": 12950
    },
    {
      "entropy": 6.176163387298584,
      "epoch": 1.1321331818579043,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048777840040524365,
      "loss": 5.9419,
      "mean_token_accuracy": 0.14727454632520676,
      "num_tokens": 26241491.0,
      "step": 12955
    },
    {
      "entropy": 6.1419812679290775,
      "epoch": 1.132570130210609,
      "grad_norm": 2.546875,
      "learning_rate": 0.00048776826808525074,
      "loss": 6.0072,
      "mean_token_accuracy": 0.14592743515968323,
      "num_tokens": 26252037.0,
      "step": 12960
    },
    {
      "entropy": 6.187282514572144,
      "epoch": 1.1330070785633137,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00048775813168425104,
      "loss": 6.0075,
      "mean_token_accuracy": 0.14510024040937425,
      "num_tokens": 26262310.0,
      "step": 12965
    },
    {
      "entropy": 6.220762872695923,
      "epoch": 1.1334440269160186,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048774799120243883,
      "loss": 5.8953,
      "mean_token_accuracy": 0.1512672409415245,
      "num_tokens": 26273001.0,
      "step": 12970
    },
    {
      "entropy": 6.204404687881469,
      "epoch": 1.1338809752687231,
      "grad_norm": 5.625,
      "learning_rate": 0.0004877378466400087,
      "loss": 5.9418,
      "mean_token_accuracy": 0.15215789824724196,
      "num_tokens": 26283500.0,
      "step": 12975
    },
    {
      "entropy": 6.102461242675782,
      "epoch": 1.134317923621428,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004877276979971552,
      "loss": 5.8917,
      "mean_token_accuracy": 0.14968496486544608,
      "num_tokens": 26294118.0,
      "step": 12980
    },
    {
      "entropy": 6.23026123046875,
      "epoch": 1.1347548719741327,
      "grad_norm": 1.8984375,
      "learning_rate": 0.000487717545274073,
      "loss": 6.0188,
      "mean_token_accuracy": 0.14182747900485992,
      "num_tokens": 26303833.0,
      "step": 12985
    },
    {
      "entropy": 6.237927532196045,
      "epoch": 1.1351918203268374,
      "grad_norm": 4.75,
      "learning_rate": 0.00048770738847095683,
      "loss": 5.9587,
      "mean_token_accuracy": 0.1449764922261238,
      "num_tokens": 26313542.0,
      "step": 12990
    },
    {
      "entropy": 6.137920904159546,
      "epoch": 1.135628768679542,
      "grad_norm": 2.125,
      "learning_rate": 0.00048769722758800144,
      "loss": 5.9115,
      "mean_token_accuracy": 0.1549493372440338,
      "num_tokens": 26322783.0,
      "step": 12995
    },
    {
      "entropy": 6.217292070388794,
      "epoch": 1.1360657170322468,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004876870626254018,
      "loss": 5.9672,
      "mean_token_accuracy": 0.14531565308570862,
      "num_tokens": 26333770.0,
      "step": 13000
    },
    {
      "entropy": 6.222366571426392,
      "epoch": 1.1365026653849515,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048767689358335286,
      "loss": 5.9406,
      "mean_token_accuracy": 0.14643460512161255,
      "num_tokens": 26344435.0,
      "step": 13005
    },
    {
      "entropy": 6.131643676757813,
      "epoch": 1.1369396137376562,
      "grad_norm": 5.28125,
      "learning_rate": 0.0004876667204620497,
      "loss": 5.8985,
      "mean_token_accuracy": 0.15525186583399772,
      "num_tokens": 26355410.0,
      "step": 13010
    },
    {
      "entropy": 6.17512469291687,
      "epoch": 1.1373765620903609,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004876565432616873,
      "loss": 5.9563,
      "mean_token_accuracy": 0.15009936392307283,
      "num_tokens": 26365785.0,
      "step": 13015
    },
    {
      "entropy": 6.234361839294434,
      "epoch": 1.1378135104430656,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048764636198246104,
      "loss": 5.8995,
      "mean_token_accuracy": 0.1503366306424141,
      "num_tokens": 26375489.0,
      "step": 13020
    },
    {
      "entropy": 6.13693790435791,
      "epoch": 1.1382504587957702,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004876361766245661,
      "loss": 5.8935,
      "mean_token_accuracy": 0.1502930074930191,
      "num_tokens": 26385126.0,
      "step": 13025
    },
    {
      "entropy": 6.141677713394165,
      "epoch": 1.138687407148475,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004876259871881979,
      "loss": 5.8918,
      "mean_token_accuracy": 0.14507785439491272,
      "num_tokens": 26395314.0,
      "step": 13030
    },
    {
      "entropy": 6.177396249771118,
      "epoch": 1.1391243555011799,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004876157936735519,
      "loss": 5.9339,
      "mean_token_accuracy": 0.14458275586366653,
      "num_tokens": 26404660.0,
      "step": 13035
    },
    {
      "entropy": 6.194282484054566,
      "epoch": 1.1395613038538845,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048760559608082343,
      "loss": 5.9359,
      "mean_token_accuracy": 0.15247158706188202,
      "num_tokens": 26415371.0,
      "step": 13040
    },
    {
      "entropy": 6.171847629547119,
      "epoch": 1.1399982522065892,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004875953944102084,
      "loss": 5.9378,
      "mean_token_accuracy": 0.14727843254804612,
      "num_tokens": 26424896.0,
      "step": 13045
    },
    {
      "entropy": 6.177714014053345,
      "epoch": 1.140435200559294,
      "grad_norm": 2.78125,
      "learning_rate": 0.00048758518866190214,
      "loss": 5.9013,
      "mean_token_accuracy": 0.15153519809246063,
      "num_tokens": 26434847.0,
      "step": 13050
    },
    {
      "entropy": 6.119139719009399,
      "epoch": 1.1408721489119986,
      "grad_norm": 4.625,
      "learning_rate": 0.0004875749788361006,
      "loss": 5.8668,
      "mean_token_accuracy": 0.14892391189932824,
      "num_tokens": 26444138.0,
      "step": 13055
    },
    {
      "entropy": 6.175608062744141,
      "epoch": 1.1413090972647033,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004875647649329997,
      "loss": 5.9071,
      "mean_token_accuracy": 0.1447135291993618,
      "num_tokens": 26453549.0,
      "step": 13060
    },
    {
      "entropy": 6.201389455795288,
      "epoch": 1.141746045617408,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004875545469527952,
      "loss": 5.9302,
      "mean_token_accuracy": 0.14497822076082229,
      "num_tokens": 26464386.0,
      "step": 13065
    },
    {
      "entropy": 6.16513843536377,
      "epoch": 1.1421829939701127,
      "grad_norm": 2.296875,
      "learning_rate": 0.000487544324895683,
      "loss": 5.8879,
      "mean_token_accuracy": 0.15028253346681594,
      "num_tokens": 26474146.0,
      "step": 13070
    },
    {
      "entropy": 6.158507823944092,
      "epoch": 1.1426199423228174,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048753409876185934,
      "loss": 5.8559,
      "mean_token_accuracy": 0.15528683811426164,
      "num_tokens": 26484408.0,
      "step": 13075
    },
    {
      "entropy": 6.1672851085662845,
      "epoch": 1.143056890675522,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004875238685515203,
      "loss": 5.8633,
      "mean_token_accuracy": 0.14090832620859145,
      "num_tokens": 26494769.0,
      "step": 13080
    },
    {
      "entropy": 6.129169368743897,
      "epoch": 1.1434938390282268,
      "grad_norm": 2.859375,
      "learning_rate": 0.00048751363426486206,
      "loss": 5.9309,
      "mean_token_accuracy": 0.14978424459695816,
      "num_tokens": 26505917.0,
      "step": 13085
    },
    {
      "entropy": 6.199118709564209,
      "epoch": 1.1439307873809317,
      "grad_norm": 3.3125,
      "learning_rate": 0.000487503395902081,
      "loss": 5.9595,
      "mean_token_accuracy": 0.13977217376232148,
      "num_tokens": 26516388.0,
      "step": 13090
    },
    {
      "entropy": 6.294061899185181,
      "epoch": 1.1443677357336364,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004874931534633735,
      "loss": 5.9543,
      "mean_token_accuracy": 0.14296025186777114,
      "num_tokens": 26527180.0,
      "step": 13095
    },
    {
      "entropy": 6.150160551071167,
      "epoch": 1.144804684086341,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004874829069489359,
      "loss": 5.9007,
      "mean_token_accuracy": 0.15269961953163147,
      "num_tokens": 26537705.0,
      "step": 13100
    },
    {
      "entropy": 6.115722799301148,
      "epoch": 1.1452416324390458,
      "grad_norm": 2.734375,
      "learning_rate": 0.00048747265635896487,
      "loss": 6.0743,
      "mean_token_accuracy": 0.13656316250562667,
      "num_tokens": 26548862.0,
      "step": 13105
    },
    {
      "entropy": 6.216728019714355,
      "epoch": 1.1456785807917504,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048746240169365683,
      "loss": 5.9483,
      "mean_token_accuracy": 0.14423804432153703,
      "num_tokens": 26559344.0,
      "step": 13110
    },
    {
      "entropy": 6.187315177917481,
      "epoch": 1.1461155291444551,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004874521429532087,
      "loss": 5.8891,
      "mean_token_accuracy": 0.14793665260076522,
      "num_tokens": 26568439.0,
      "step": 13115
    },
    {
      "entropy": 6.192612266540527,
      "epoch": 1.1465524774971598,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00048744188013781714,
      "loss": 5.9041,
      "mean_token_accuracy": 0.1501721516251564,
      "num_tokens": 26578588.0,
      "step": 13120
    },
    {
      "entropy": 6.147067213058472,
      "epoch": 1.1469894258498645,
      "grad_norm": 1.78125,
      "learning_rate": 0.000487431613247679,
      "loss": 5.9471,
      "mean_token_accuracy": 0.1519698679447174,
      "num_tokens": 26588491.0,
      "step": 13125
    },
    {
      "entropy": 6.160977935791015,
      "epoch": 1.1474263742025692,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048742134228299116,
      "loss": 5.9643,
      "mean_token_accuracy": 0.14815946519374848,
      "num_tokens": 26598945.0,
      "step": 13130
    },
    {
      "entropy": 6.172370958328247,
      "epoch": 1.147863322555274,
      "grad_norm": 5.21875,
      "learning_rate": 0.0004874110672439508,
      "loss": 5.8636,
      "mean_token_accuracy": 0.1555538922548294,
      "num_tokens": 26609842.0,
      "step": 13135
    },
    {
      "entropy": 6.2168840885162355,
      "epoch": 1.1483002709079786,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004874007881307547,
      "loss": 5.9902,
      "mean_token_accuracy": 0.1484801799058914,
      "num_tokens": 26620836.0,
      "step": 13140
    },
    {
      "entropy": 6.224114036560058,
      "epoch": 1.1487372192606835,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048739050494360026,
      "loss": 5.9127,
      "mean_token_accuracy": 0.14655357077717782,
      "num_tokens": 26630148.0,
      "step": 13145
    },
    {
      "entropy": 6.215399074554443,
      "epoch": 1.149174167613388,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048738021768268463,
      "loss": 5.941,
      "mean_token_accuracy": 0.14698300510644913,
      "num_tokens": 26640518.0,
      "step": 13150
    },
    {
      "entropy": 6.132278299331665,
      "epoch": 1.149611115966093,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004873699263482052,
      "loss": 5.8647,
      "mean_token_accuracy": 0.16045882403850556,
      "num_tokens": 26650156.0,
      "step": 13155
    },
    {
      "entropy": 6.229346513748169,
      "epoch": 1.1500480643187976,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048735963094035924,
      "loss": 5.9309,
      "mean_token_accuracy": 0.1450416103005409,
      "num_tokens": 26661251.0,
      "step": 13160
    },
    {
      "entropy": 6.20313835144043,
      "epoch": 1.1504850126715023,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004873493314593443,
      "loss": 5.9533,
      "mean_token_accuracy": 0.14792988076806068,
      "num_tokens": 26671851.0,
      "step": 13165
    },
    {
      "entropy": 6.1452638626098635,
      "epoch": 1.150921961024207,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004873390279053579,
      "loss": 5.9549,
      "mean_token_accuracy": 0.14189176633954048,
      "num_tokens": 26682984.0,
      "step": 13170
    },
    {
      "entropy": 6.1874671459198,
      "epoch": 1.1513589093769117,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004873287202785977,
      "loss": 5.8878,
      "mean_token_accuracy": 0.1542479231953621,
      "num_tokens": 26692706.0,
      "step": 13175
    },
    {
      "entropy": 6.203208780288696,
      "epoch": 1.1517958577296163,
      "grad_norm": 7.09375,
      "learning_rate": 0.0004873184085792614,
      "loss": 5.9249,
      "mean_token_accuracy": 0.14727155715227128,
      "num_tokens": 26702152.0,
      "step": 13180
    },
    {
      "entropy": 6.062010002136231,
      "epoch": 1.152232806082321,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048730809280754674,
      "loss": 5.8999,
      "mean_token_accuracy": 0.15685417205095292,
      "num_tokens": 26711404.0,
      "step": 13185
    },
    {
      "entropy": 6.177418375015259,
      "epoch": 1.1526697544350257,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004872977729636516,
      "loss": 5.8838,
      "mean_token_accuracy": 0.152011838555336,
      "num_tokens": 26720776.0,
      "step": 13190
    },
    {
      "entropy": 6.210717296600341,
      "epoch": 1.1531067027877304,
      "grad_norm": 3.234375,
      "learning_rate": 0.00048728744904777394,
      "loss": 5.9283,
      "mean_token_accuracy": 0.14755924940109252,
      "num_tokens": 26730700.0,
      "step": 13195
    },
    {
      "entropy": 6.22635588645935,
      "epoch": 1.1535436511404351,
      "grad_norm": 7.8125,
      "learning_rate": 0.00048727712106011185,
      "loss": 5.9903,
      "mean_token_accuracy": 0.14609125405550002,
      "num_tokens": 26741453.0,
      "step": 13200
    },
    {
      "entropy": 6.208837938308716,
      "epoch": 1.1539805994931398,
      "grad_norm": 2.9375,
      "learning_rate": 0.00048726678900086327,
      "loss": 5.9069,
      "mean_token_accuracy": 0.1469104468822479,
      "num_tokens": 26750627.0,
      "step": 13205
    },
    {
      "entropy": 6.204519844055175,
      "epoch": 1.1544175478458447,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048725645287022644,
      "loss": 5.9276,
      "mean_token_accuracy": 0.1467217855155468,
      "num_tokens": 26760820.0,
      "step": 13210
    },
    {
      "entropy": 6.1821324825286865,
      "epoch": 1.1548544961985494,
      "grad_norm": 2.25,
      "learning_rate": 0.0004872461126683996,
      "loss": 5.8699,
      "mean_token_accuracy": 0.1448241375386715,
      "num_tokens": 26771535.0,
      "step": 13215
    },
    {
      "entropy": 6.222924375534058,
      "epoch": 1.155291444551254,
      "grad_norm": 2.125,
      "learning_rate": 0.0004872357683955812,
      "loss": 5.9582,
      "mean_token_accuracy": 0.15162405669689177,
      "num_tokens": 26782755.0,
      "step": 13220
    },
    {
      "entropy": 6.199812126159668,
      "epoch": 1.1557283929039588,
      "grad_norm": 2.578125,
      "learning_rate": 0.00048722542005196954,
      "loss": 5.9128,
      "mean_token_accuracy": 0.1463736265897751,
      "num_tokens": 26792834.0,
      "step": 13225
    },
    {
      "entropy": 6.111869430541992,
      "epoch": 1.1561653412566635,
      "grad_norm": 2.90625,
      "learning_rate": 0.00048721506763776306,
      "loss": 5.9053,
      "mean_token_accuracy": 0.15205851346254348,
      "num_tokens": 26802865.0,
      "step": 13230
    },
    {
      "entropy": 6.121051549911499,
      "epoch": 1.1566022896093682,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004872047111531604,
      "loss": 5.8974,
      "mean_token_accuracy": 0.14660682380199433,
      "num_tokens": 26813566.0,
      "step": 13235
    },
    {
      "entropy": 6.183195304870606,
      "epoch": 1.1570392379620729,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004871943505983602,
      "loss": 5.8854,
      "mean_token_accuracy": 0.15813306868076324,
      "num_tokens": 26823885.0,
      "step": 13240
    },
    {
      "entropy": 6.063178014755249,
      "epoch": 1.1574761863147776,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004871839859735612,
      "loss": 5.8253,
      "mean_token_accuracy": 0.1624150589108467,
      "num_tokens": 26833236.0,
      "step": 13245
    },
    {
      "entropy": 6.172446298599243,
      "epoch": 1.1579131346674822,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048717361727896203,
      "loss": 5.9686,
      "mean_token_accuracy": 0.1461391344666481,
      "num_tokens": 26842651.0,
      "step": 13250
    },
    {
      "entropy": 6.175981140136718,
      "epoch": 1.158350083020187,
      "grad_norm": 3.078125,
      "learning_rate": 0.00048716324451476183,
      "loss": 5.883,
      "mean_token_accuracy": 0.14535028114914894,
      "num_tokens": 26852198.0,
      "step": 13255
    },
    {
      "entropy": 6.262139511108399,
      "epoch": 1.1587870313728916,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004871528676811594,
      "loss": 6.0089,
      "mean_token_accuracy": 0.14602211117744446,
      "num_tokens": 26862807.0,
      "step": 13260
    },
    {
      "entropy": 6.273232746124267,
      "epoch": 1.1592239797255965,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004871424867783538,
      "loss": 5.9583,
      "mean_token_accuracy": 0.1396594375371933,
      "num_tokens": 26871992.0,
      "step": 13265
    },
    {
      "entropy": 6.196260643005371,
      "epoch": 1.1596609280783012,
      "grad_norm": 3.84375,
      "learning_rate": 0.00048713210180654414,
      "loss": 5.8997,
      "mean_token_accuracy": 0.14591027721762656,
      "num_tokens": 26881417.0,
      "step": 13270
    },
    {
      "entropy": 6.188068580627442,
      "epoch": 1.160097876431006,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048712171276592964,
      "loss": 5.874,
      "mean_token_accuracy": 0.15216402113437652,
      "num_tokens": 26891879.0,
      "step": 13275
    },
    {
      "entropy": 6.183082151412964,
      "epoch": 1.1605348247837106,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004871113196567095,
      "loss": 6.0166,
      "mean_token_accuracy": 0.14431816339492798,
      "num_tokens": 26902729.0,
      "step": 13280
    },
    {
      "entropy": 6.152678155899048,
      "epoch": 1.1609717731364153,
      "grad_norm": 3.140625,
      "learning_rate": 0.00048710092247908316,
      "loss": 5.9107,
      "mean_token_accuracy": 0.15283699035644532,
      "num_tokens": 26914415.0,
      "step": 13285
    },
    {
      "entropy": 6.256611728668213,
      "epoch": 1.16140872148912,
      "grad_norm": 4.25,
      "learning_rate": 0.00048709052123324995,
      "loss": 5.9447,
      "mean_token_accuracy": 0.1510358914732933,
      "num_tokens": 26924513.0,
      "step": 13290
    },
    {
      "entropy": 6.193924140930176,
      "epoch": 1.1618456698418247,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048708011591940936,
      "loss": 5.8499,
      "mean_token_accuracy": 0.1548793777823448,
      "num_tokens": 26935039.0,
      "step": 13295
    },
    {
      "entropy": 6.129718065261841,
      "epoch": 1.1622826181945294,
      "grad_norm": 2.46875,
      "learning_rate": 0.000487069706537761,
      "loss": 5.838,
      "mean_token_accuracy": 0.15177211165428162,
      "num_tokens": 26945148.0,
      "step": 13300
    },
    {
      "entropy": 6.139807987213135,
      "epoch": 1.162719566547234,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004870592930885046,
      "loss": 5.8829,
      "mean_token_accuracy": 0.15055215656757354,
      "num_tokens": 26955227.0,
      "step": 13305
    },
    {
      "entropy": 6.194760179519653,
      "epoch": 1.1631565148999388,
      "grad_norm": 2.25,
      "learning_rate": 0.00048704887557183993,
      "loss": 5.8557,
      "mean_token_accuracy": 0.14913835525512695,
      "num_tokens": 26964389.0,
      "step": 13310
    },
    {
      "entropy": 6.2291806697845455,
      "epoch": 1.1635934632526435,
      "grad_norm": 3.078125,
      "learning_rate": 0.00048703845398796665,
      "loss": 5.8629,
      "mean_token_accuracy": 0.1489256888628006,
      "num_tokens": 26974695.0,
      "step": 13315
    },
    {
      "entropy": 6.227977085113525,
      "epoch": 1.1640304116053484,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048702802833708466,
      "loss": 5.8553,
      "mean_token_accuracy": 0.15597414076328278,
      "num_tokens": 26984851.0,
      "step": 13320
    },
    {
      "entropy": 6.236116600036621,
      "epoch": 1.1644673599580528,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048701759861939403,
      "loss": 5.9481,
      "mean_token_accuracy": 0.14879257455468178,
      "num_tokens": 26995967.0,
      "step": 13325
    },
    {
      "entropy": 6.125051069259643,
      "epoch": 1.1649043083107578,
      "grad_norm": 4.5,
      "learning_rate": 0.00048700716483509476,
      "loss": 5.8634,
      "mean_token_accuracy": 0.15883732140064238,
      "num_tokens": 27006469.0,
      "step": 13330
    },
    {
      "entropy": 6.1943573474884035,
      "epoch": 1.1653412566634624,
      "grad_norm": 4.5,
      "learning_rate": 0.000486996726984387,
      "loss": 5.9676,
      "mean_token_accuracy": 0.14819352924823762,
      "num_tokens": 27016419.0,
      "step": 13335
    },
    {
      "entropy": 6.263797998428345,
      "epoch": 1.1657782050161671,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004869862850674709,
      "loss": 5.9365,
      "mean_token_accuracy": 0.1433882534503937,
      "num_tokens": 27026322.0,
      "step": 13340
    },
    {
      "entropy": 6.152531290054322,
      "epoch": 1.1662151533688718,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048697583908454675,
      "loss": 5.8949,
      "mean_token_accuracy": 0.14920396804809571,
      "num_tokens": 27036518.0,
      "step": 13345
    },
    {
      "entropy": 6.203283405303955,
      "epoch": 1.1666521017215765,
      "grad_norm": 1.9609375,
      "learning_rate": 0.000486965389035815,
      "loss": 5.9558,
      "mean_token_accuracy": 0.1449318841099739,
      "num_tokens": 27048035.0,
      "step": 13350
    },
    {
      "entropy": 6.19470181465149,
      "epoch": 1.1670890500742812,
      "grad_norm": 3.125,
      "learning_rate": 0.00048695493492147596,
      "loss": 5.9154,
      "mean_token_accuracy": 0.14441130608320235,
      "num_tokens": 27058507.0,
      "step": 13355
    },
    {
      "entropy": 6.133526134490967,
      "epoch": 1.167525998426986,
      "grad_norm": 3.5,
      "learning_rate": 0.00048694447674173025,
      "loss": 5.9015,
      "mean_token_accuracy": 0.1514151781797409,
      "num_tokens": 27068350.0,
      "step": 13360
    },
    {
      "entropy": 6.128726005554199,
      "epoch": 1.1679629467796906,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004869340144967784,
      "loss": 5.8985,
      "mean_token_accuracy": 0.15001028180122375,
      "num_tokens": 27077840.0,
      "step": 13365
    },
    {
      "entropy": 6.146130847930908,
      "epoch": 1.1683998951323953,
      "grad_norm": 4.40625,
      "learning_rate": 0.00048692354818682113,
      "loss": 5.8649,
      "mean_token_accuracy": 0.15206778347492217,
      "num_tokens": 27087687.0,
      "step": 13370
    },
    {
      "entropy": 6.208657455444336,
      "epoch": 1.1688368434851,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004869130778120591,
      "loss": 6.0342,
      "mean_token_accuracy": 0.1396991528570652,
      "num_tokens": 27099057.0,
      "step": 13375
    },
    {
      "entropy": 6.197882795333863,
      "epoch": 1.1692737918378047,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004869026033726933,
      "loss": 5.9813,
      "mean_token_accuracy": 0.1465886875987053,
      "num_tokens": 27109617.0,
      "step": 13380
    },
    {
      "entropy": 6.235219240188599,
      "epoch": 1.1697107401905096,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004868921248689244,
      "loss": 5.9435,
      "mean_token_accuracy": 0.14212179258465768,
      "num_tokens": 27119456.0,
      "step": 13385
    },
    {
      "entropy": 6.155871772766114,
      "epoch": 1.1701476885432143,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004868816423009535,
      "loss": 5.8738,
      "mean_token_accuracy": 0.1531216859817505,
      "num_tokens": 27129333.0,
      "step": 13390
    },
    {
      "entropy": 6.150876522064209,
      "epoch": 1.170584636895919,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004868711556689818,
      "loss": 5.9022,
      "mean_token_accuracy": 0.14866817519068717,
      "num_tokens": 27139205.0,
      "step": 13395
    },
    {
      "entropy": 6.224974918365478,
      "epoch": 1.1710215852486237,
      "grad_norm": 2.25,
      "learning_rate": 0.00048686066497321027,
      "loss": 5.9259,
      "mean_token_accuracy": 0.15095333755016327,
      "num_tokens": 27148530.0,
      "step": 13400
    },
    {
      "entropy": 6.277073955535888,
      "epoch": 1.1714585336013283,
      "grad_norm": 6.5625,
      "learning_rate": 0.00048685017021384014,
      "loss": 6.0015,
      "mean_token_accuracy": 0.14719715267419814,
      "num_tokens": 27159000.0,
      "step": 13405
    },
    {
      "entropy": 6.230797481536865,
      "epoch": 1.171895481954033,
      "grad_norm": 2.875,
      "learning_rate": 0.00048683967139107276,
      "loss": 5.9327,
      "mean_token_accuracy": 0.14801158457994462,
      "num_tokens": 27169035.0,
      "step": 13410
    },
    {
      "entropy": 6.1944574356079105,
      "epoch": 1.1723324303067377,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004868291685051095,
      "loss": 5.9278,
      "mean_token_accuracy": 0.14793018102645875,
      "num_tokens": 27179809.0,
      "step": 13415
    },
    {
      "entropy": 6.174774408340454,
      "epoch": 1.1727693786594424,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004868186615561518,
      "loss": 5.9309,
      "mean_token_accuracy": 0.1465114802122116,
      "num_tokens": 27190445.0,
      "step": 13420
    },
    {
      "entropy": 6.214708614349365,
      "epoch": 1.173206327012147,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048680815054440114,
      "loss": 5.8838,
      "mean_token_accuracy": 0.14549873024225235,
      "num_tokens": 27201097.0,
      "step": 13425
    },
    {
      "entropy": 6.180704498291016,
      "epoch": 1.1736432753648518,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048679763547005925,
      "loss": 5.8619,
      "mean_token_accuracy": 0.15421843975782396,
      "num_tokens": 27210181.0,
      "step": 13430
    },
    {
      "entropy": 6.140958547592163,
      "epoch": 1.1740802237175565,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004867871163333276,
      "loss": 5.9087,
      "mean_token_accuracy": 0.14630078971385957,
      "num_tokens": 27219869.0,
      "step": 13435
    },
    {
      "entropy": 6.179342031478882,
      "epoch": 1.1745171720702614,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004867765931344082,
      "loss": 5.9312,
      "mean_token_accuracy": 0.15077333003282548,
      "num_tokens": 27229730.0,
      "step": 13440
    },
    {
      "entropy": 6.194200658798218,
      "epoch": 1.174954120422966,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004867660658735027,
      "loss": 5.91,
      "mean_token_accuracy": 0.15319260582327843,
      "num_tokens": 27239534.0,
      "step": 13445
    },
    {
      "entropy": 6.176281595230103,
      "epoch": 1.1753910687756708,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004867555345508131,
      "loss": 5.9652,
      "mean_token_accuracy": 0.14543908312916756,
      "num_tokens": 27249017.0,
      "step": 13450
    },
    {
      "entropy": 6.242269849777221,
      "epoch": 1.1758280171283755,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004867449991665414,
      "loss": 6.0208,
      "mean_token_accuracy": 0.14285536855459213,
      "num_tokens": 27261135.0,
      "step": 13455
    },
    {
      "entropy": 6.231696653366089,
      "epoch": 1.1762649654810802,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004867344597208897,
      "loss": 5.9102,
      "mean_token_accuracy": 0.1442875511944294,
      "num_tokens": 27270763.0,
      "step": 13460
    },
    {
      "entropy": 6.1746025562286375,
      "epoch": 1.1767019138337849,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004867239162140601,
      "loss": 5.9359,
      "mean_token_accuracy": 0.14871663004159927,
      "num_tokens": 27280115.0,
      "step": 13465
    },
    {
      "entropy": 6.155801773071289,
      "epoch": 1.1771388621864896,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004867133686462548,
      "loss": 5.9032,
      "mean_token_accuracy": 0.14929405450820923,
      "num_tokens": 27290162.0,
      "step": 13470
    },
    {
      "entropy": 6.1686280250549315,
      "epoch": 1.1775758105391942,
      "grad_norm": 3.171875,
      "learning_rate": 0.00048670281701767616,
      "loss": 5.8689,
      "mean_token_accuracy": 0.1549352988600731,
      "num_tokens": 27300467.0,
      "step": 13475
    },
    {
      "entropy": 6.157352781295776,
      "epoch": 1.178012758891899,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004866922613285266,
      "loss": 5.9389,
      "mean_token_accuracy": 0.14707837551832198,
      "num_tokens": 27310524.0,
      "step": 13480
    },
    {
      "entropy": 6.187600374221802,
      "epoch": 1.1784497072446036,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004866817015790085,
      "loss": 5.9397,
      "mean_token_accuracy": 0.1510642573237419,
      "num_tokens": 27320861.0,
      "step": 13485
    },
    {
      "entropy": 6.2566859245300295,
      "epoch": 1.1788866555973083,
      "grad_norm": 3.78125,
      "learning_rate": 0.00048667113776932453,
      "loss": 5.9566,
      "mean_token_accuracy": 0.1472768247127533,
      "num_tokens": 27331771.0,
      "step": 13490
    },
    {
      "entropy": 6.187179136276245,
      "epoch": 1.1793236039500132,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004866605698996772,
      "loss": 5.9401,
      "mean_token_accuracy": 0.1475620873272419,
      "num_tokens": 27341449.0,
      "step": 13495
    },
    {
      "entropy": 6.1723979949951175,
      "epoch": 1.1797605523027177,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004866499979702691,
      "loss": 6.0136,
      "mean_token_accuracy": 0.13973579555749893,
      "num_tokens": 27353127.0,
      "step": 13500
    },
    {
      "entropy": 6.197310924530029,
      "epoch": 1.1801975006554226,
      "grad_norm": 2.375,
      "learning_rate": 0.0004866394219813032,
      "loss": 5.8818,
      "mean_token_accuracy": 0.15200105756521226,
      "num_tokens": 27362116.0,
      "step": 13505
    },
    {
      "entropy": 6.184662580490112,
      "epoch": 1.1806344490081273,
      "grad_norm": 4.34375,
      "learning_rate": 0.00048662884193298233,
      "loss": 5.8896,
      "mean_token_accuracy": 0.15370760038495063,
      "num_tokens": 27372204.0,
      "step": 13510
    },
    {
      "entropy": 6.152497243881226,
      "epoch": 1.181071397360832,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004866182578255094,
      "loss": 5.9377,
      "mean_token_accuracy": 0.14656676426529885,
      "num_tokens": 27382963.0,
      "step": 13515
    },
    {
      "entropy": 6.13969931602478,
      "epoch": 1.1815083457135367,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00048660766965908736,
      "loss": 5.8866,
      "mean_token_accuracy": 0.15366474986076356,
      "num_tokens": 27392426.0,
      "step": 13520
    },
    {
      "entropy": 6.129809761047364,
      "epoch": 1.1819452940662414,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004865970774339194,
      "loss": 5.8982,
      "mean_token_accuracy": 0.15134273767471312,
      "num_tokens": 27402679.0,
      "step": 13525
    },
    {
      "entropy": 6.20675106048584,
      "epoch": 1.182382242418946,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048658648115020855,
      "loss": 5.9629,
      "mean_token_accuracy": 0.14752191081643104,
      "num_tokens": 27412911.0,
      "step": 13530
    },
    {
      "entropy": 6.236071634292602,
      "epoch": 1.1828191907716508,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004865758808081581,
      "loss": 5.8992,
      "mean_token_accuracy": 0.1471305325627327,
      "num_tokens": 27423394.0,
      "step": 13535
    },
    {
      "entropy": 6.168351316452027,
      "epoch": 1.1832561391243555,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048656527640797144,
      "loss": 5.8875,
      "mean_token_accuracy": 0.15766822248697282,
      "num_tokens": 27433016.0,
      "step": 13540
    },
    {
      "entropy": 6.146002244949341,
      "epoch": 1.1836930874770601,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048655466794985194,
      "loss": 5.9066,
      "mean_token_accuracy": 0.14376786053180696,
      "num_tokens": 27442414.0,
      "step": 13545
    },
    {
      "entropy": 6.139078283309937,
      "epoch": 1.1841300358297648,
      "grad_norm": 2.46875,
      "learning_rate": 0.000486544055434003,
      "loss": 5.9274,
      "mean_token_accuracy": 0.14735411405563353,
      "num_tokens": 27452410.0,
      "step": 13550
    },
    {
      "entropy": 6.209070253372192,
      "epoch": 1.1845669841824695,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004865334388606282,
      "loss": 5.8717,
      "mean_token_accuracy": 0.15575417578220369,
      "num_tokens": 27462196.0,
      "step": 13555
    },
    {
      "entropy": 6.22119550704956,
      "epoch": 1.1850039325351744,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048652281822993127,
      "loss": 5.9685,
      "mean_token_accuracy": 0.1431777723133564,
      "num_tokens": 27473934.0,
      "step": 13560
    },
    {
      "entropy": 6.19681749343872,
      "epoch": 1.1854408808878791,
      "grad_norm": 3.75,
      "learning_rate": 0.00048651219354211585,
      "loss": 6.0147,
      "mean_token_accuracy": 0.1507688894867897,
      "num_tokens": 27484218.0,
      "step": 13565
    },
    {
      "entropy": 6.153265857696534,
      "epoch": 1.1858778292405838,
      "grad_norm": 2.484375,
      "learning_rate": 0.00048650156479738575,
      "loss": 6.0073,
      "mean_token_accuracy": 0.14877953380346298,
      "num_tokens": 27494663.0,
      "step": 13570
    },
    {
      "entropy": 6.203997945785522,
      "epoch": 1.1863147775932885,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048649093199594473,
      "loss": 5.8486,
      "mean_token_accuracy": 0.15308129787445068,
      "num_tokens": 27504233.0,
      "step": 13575
    },
    {
      "entropy": 6.185912227630615,
      "epoch": 1.1867517259459932,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004864802951379969,
      "loss": 5.9159,
      "mean_token_accuracy": 0.14943065941333772,
      "num_tokens": 27514032.0,
      "step": 13580
    },
    {
      "entropy": 6.176620435714722,
      "epoch": 1.187188674298698,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048646965422374615,
      "loss": 5.9781,
      "mean_token_accuracy": 0.14695518538355828,
      "num_tokens": 27524659.0,
      "step": 13585
    },
    {
      "entropy": 6.179401302337647,
      "epoch": 1.1876256226514026,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004864590092533966,
      "loss": 5.8604,
      "mean_token_accuracy": 0.14711184054613113,
      "num_tokens": 27534003.0,
      "step": 13590
    },
    {
      "entropy": 6.138573026657104,
      "epoch": 1.1880625710041073,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004864483602271524,
      "loss": 5.8916,
      "mean_token_accuracy": 0.14601757675409316,
      "num_tokens": 27543586.0,
      "step": 13595
    },
    {
      "entropy": 6.123298597335816,
      "epoch": 1.188499519356812,
      "grad_norm": 2.578125,
      "learning_rate": 0.000486437707145218,
      "loss": 5.8622,
      "mean_token_accuracy": 0.15870054811239243,
      "num_tokens": 27554210.0,
      "step": 13600
    },
    {
      "entropy": 6.230636405944824,
      "epoch": 1.1889364677095167,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048642705000779754,
      "loss": 5.9765,
      "mean_token_accuracy": 0.14838793352246285,
      "num_tokens": 27565324.0,
      "step": 13605
    },
    {
      "entropy": 6.166018772125244,
      "epoch": 1.1893734160622214,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004864163888150955,
      "loss": 5.8915,
      "mean_token_accuracy": 0.15250848829746247,
      "num_tokens": 27575192.0,
      "step": 13610
    },
    {
      "entropy": 6.148905420303345,
      "epoch": 1.1898103644149263,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048640572356731623,
      "loss": 5.9068,
      "mean_token_accuracy": 0.14767027348279954,
      "num_tokens": 27585294.0,
      "step": 13615
    },
    {
      "entropy": 6.18740873336792,
      "epoch": 1.190247312767631,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048639505426466447,
      "loss": 5.925,
      "mean_token_accuracy": 0.15437621027231216,
      "num_tokens": 27595667.0,
      "step": 13620
    },
    {
      "entropy": 6.194943046569824,
      "epoch": 1.1906842611203357,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004863843809073448,
      "loss": 5.8869,
      "mean_token_accuracy": 0.15669466108083724,
      "num_tokens": 27605816.0,
      "step": 13625
    },
    {
      "entropy": 6.0837657928466795,
      "epoch": 1.1911212094730403,
      "grad_norm": 6.90625,
      "learning_rate": 0.000486373703495562,
      "loss": 5.9017,
      "mean_token_accuracy": 0.150448040664196,
      "num_tokens": 27614874.0,
      "step": 13630
    },
    {
      "entropy": 6.157348489761352,
      "epoch": 1.191558157825745,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004863630220295207,
      "loss": 5.9021,
      "mean_token_accuracy": 0.14978635609149932,
      "num_tokens": 27625610.0,
      "step": 13635
    },
    {
      "entropy": 6.209554290771484,
      "epoch": 1.1919951061784497,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048635233650942596,
      "loss": 5.9245,
      "mean_token_accuracy": 0.14628103449940683,
      "num_tokens": 27636646.0,
      "step": 13640
    },
    {
      "entropy": 6.155116558074951,
      "epoch": 1.1924320545311544,
      "grad_norm": 2.859375,
      "learning_rate": 0.00048634164693548267,
      "loss": 5.9354,
      "mean_token_accuracy": 0.14759790599346162,
      "num_tokens": 27647311.0,
      "step": 13645
    },
    {
      "entropy": 6.159931707382202,
      "epoch": 1.192869002883859,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004863309533078957,
      "loss": 5.935,
      "mean_token_accuracy": 0.14424525052309037,
      "num_tokens": 27658347.0,
      "step": 13650
    },
    {
      "entropy": 6.1673603534698485,
      "epoch": 1.1933059512365638,
      "grad_norm": 2.734375,
      "learning_rate": 0.00048632025562687047,
      "loss": 5.8853,
      "mean_token_accuracy": 0.15577892661094667,
      "num_tokens": 27669211.0,
      "step": 13655
    },
    {
      "entropy": 6.192775058746338,
      "epoch": 1.1937428995892685,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048630955389261193,
      "loss": 5.8798,
      "mean_token_accuracy": 0.15026595443487167,
      "num_tokens": 27678904.0,
      "step": 13660
    },
    {
      "entropy": 6.157775449752807,
      "epoch": 1.1941798479419732,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048629884810532547,
      "loss": 5.8649,
      "mean_token_accuracy": 0.14732719361782073,
      "num_tokens": 27688928.0,
      "step": 13665
    },
    {
      "entropy": 6.167954349517823,
      "epoch": 1.1946167962946779,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004862881382652164,
      "loss": 5.9483,
      "mean_token_accuracy": 0.14584649205207825,
      "num_tokens": 27699855.0,
      "step": 13670
    },
    {
      "entropy": 6.195680952072143,
      "epoch": 1.1950537446473826,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004862774243724901,
      "loss": 5.935,
      "mean_token_accuracy": 0.15434314161539078,
      "num_tokens": 27709857.0,
      "step": 13675
    },
    {
      "entropy": 6.197531509399414,
      "epoch": 1.1954906930000875,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004862667064273521,
      "loss": 5.9395,
      "mean_token_accuracy": 0.1450137674808502,
      "num_tokens": 27719612.0,
      "step": 13680
    },
    {
      "entropy": 6.145286178588867,
      "epoch": 1.1959276413527922,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004862559844300079,
      "loss": 5.9466,
      "mean_token_accuracy": 0.1506308361887932,
      "num_tokens": 27730052.0,
      "step": 13685
    },
    {
      "entropy": 6.152642488479614,
      "epoch": 1.1963645897054969,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004862452583806633,
      "loss": 5.8167,
      "mean_token_accuracy": 0.1600808709859848,
      "num_tokens": 27739980.0,
      "step": 13690
    },
    {
      "entropy": 6.171810626983643,
      "epoch": 1.1968015380582016,
      "grad_norm": 4.0625,
      "learning_rate": 0.00048623452827952385,
      "loss": 5.9246,
      "mean_token_accuracy": 0.1522218957543373,
      "num_tokens": 27748863.0,
      "step": 13695
    },
    {
      "entropy": 6.127953767776489,
      "epoch": 1.1972384864109062,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004862237941267956,
      "loss": 5.89,
      "mean_token_accuracy": 0.15447674840688705,
      "num_tokens": 27758672.0,
      "step": 13700
    },
    {
      "entropy": 6.164484596252441,
      "epoch": 1.197675434763611,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004862130559226842,
      "loss": 5.9148,
      "mean_token_accuracy": 0.14712174534797667,
      "num_tokens": 27768195.0,
      "step": 13705
    },
    {
      "entropy": 6.153045701980591,
      "epoch": 1.1981123831163156,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004862023136673958,
      "loss": 5.8601,
      "mean_token_accuracy": 0.15653797835111619,
      "num_tokens": 27778135.0,
      "step": 13710
    },
    {
      "entropy": 6.255939865112305,
      "epoch": 1.1985493314690203,
      "grad_norm": 4.8125,
      "learning_rate": 0.00048619156736113634,
      "loss": 5.9518,
      "mean_token_accuracy": 0.14238950833678246,
      "num_tokens": 27788723.0,
      "step": 13715
    },
    {
      "entropy": 6.21520037651062,
      "epoch": 1.198986279821725,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004861808170041119,
      "loss": 5.9571,
      "mean_token_accuracy": 0.14887107238173486,
      "num_tokens": 27799060.0,
      "step": 13720
    },
    {
      "entropy": 6.1764750480651855,
      "epoch": 1.1994232281744297,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004861700625965287,
      "loss": 5.9634,
      "mean_token_accuracy": 0.14113897532224656,
      "num_tokens": 27809542.0,
      "step": 13725
    },
    {
      "entropy": 6.108205413818359,
      "epoch": 1.1998601765271344,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048615930413859317,
      "loss": 5.9451,
      "mean_token_accuracy": 0.14324765130877495,
      "num_tokens": 27819738.0,
      "step": 13730
    },
    {
      "entropy": 6.273303651809693,
      "epoch": 1.2002971248798393,
      "grad_norm": 4.8125,
      "learning_rate": 0.00048614854163051143,
      "loss": 5.877,
      "mean_token_accuracy": 0.1496002718806267,
      "num_tokens": 27829052.0,
      "step": 13735
    },
    {
      "entropy": 6.236565160751343,
      "epoch": 1.200734073232544,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004861377750724901,
      "loss": 5.8685,
      "mean_token_accuracy": 0.15066418796777725,
      "num_tokens": 27839277.0,
      "step": 13740
    },
    {
      "entropy": 6.149129629135132,
      "epoch": 1.2011710215852487,
      "grad_norm": 2.625,
      "learning_rate": 0.0004861270044647356,
      "loss": 5.9291,
      "mean_token_accuracy": 0.1499796360731125,
      "num_tokens": 27848550.0,
      "step": 13745
    },
    {
      "entropy": 6.093921709060669,
      "epoch": 1.2016079699379534,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048611622980745446,
      "loss": 5.952,
      "mean_token_accuracy": 0.1432810753583908,
      "num_tokens": 27858546.0,
      "step": 13750
    },
    {
      "entropy": 6.213047790527344,
      "epoch": 1.202044918290658,
      "grad_norm": 5.28125,
      "learning_rate": 0.00048610545110085347,
      "loss": 5.9667,
      "mean_token_accuracy": 0.1422676384449005,
      "num_tokens": 27869192.0,
      "step": 13755
    },
    {
      "entropy": 6.250215244293213,
      "epoch": 1.2024818666433628,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004860946683451393,
      "loss": 5.9186,
      "mean_token_accuracy": 0.15389849692583085,
      "num_tokens": 27878534.0,
      "step": 13760
    },
    {
      "entropy": 6.122016906738281,
      "epoch": 1.2029188149960675,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048608388154051875,
      "loss": 5.852,
      "mean_token_accuracy": 0.1557212255895138,
      "num_tokens": 27888074.0,
      "step": 13765
    },
    {
      "entropy": 6.118573999404907,
      "epoch": 1.2033557633487721,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048607309068719875,
      "loss": 5.9079,
      "mean_token_accuracy": 0.14594862833619118,
      "num_tokens": 27898210.0,
      "step": 13770
    },
    {
      "entropy": 6.130645847320556,
      "epoch": 1.2037927117014768,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004860622957853863,
      "loss": 5.8463,
      "mean_token_accuracy": 0.15255392268300055,
      "num_tokens": 27907472.0,
      "step": 13775
    },
    {
      "entropy": 6.132734966278076,
      "epoch": 1.2042296600541815,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004860514968352885,
      "loss": 5.926,
      "mean_token_accuracy": 0.15163296610116958,
      "num_tokens": 27917265.0,
      "step": 13780
    },
    {
      "entropy": 6.158398628234863,
      "epoch": 1.2046666084068862,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048604069383711225,
      "loss": 5.8463,
      "mean_token_accuracy": 0.15847632437944412,
      "num_tokens": 27926883.0,
      "step": 13785
    },
    {
      "entropy": 6.217565965652466,
      "epoch": 1.2051035567595911,
      "grad_norm": 3.71875,
      "learning_rate": 0.00048602988679106497,
      "loss": 5.8988,
      "mean_token_accuracy": 0.14781321734189987,
      "num_tokens": 27937167.0,
      "step": 13790
    },
    {
      "entropy": 6.21237154006958,
      "epoch": 1.2055405051122958,
      "grad_norm": 2.5,
      "learning_rate": 0.0004860190756973539,
      "loss": 5.8773,
      "mean_token_accuracy": 0.15900922194123268,
      "num_tokens": 27947422.0,
      "step": 13795
    },
    {
      "entropy": 6.213820266723633,
      "epoch": 1.2059774534650005,
      "grad_norm": 2.75,
      "learning_rate": 0.00048600826055618634,
      "loss": 5.9497,
      "mean_token_accuracy": 0.15147775411605835,
      "num_tokens": 27958407.0,
      "step": 13800
    },
    {
      "entropy": 6.177189016342163,
      "epoch": 1.2064144018177052,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048599744136776986,
      "loss": 5.8194,
      "mean_token_accuracy": 0.15412741899490356,
      "num_tokens": 27968277.0,
      "step": 13805
    },
    {
      "entropy": 6.1591204643249515,
      "epoch": 1.20685135017041,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048598661813231177,
      "loss": 5.8357,
      "mean_token_accuracy": 0.15879237949848174,
      "num_tokens": 27978116.0,
      "step": 13810
    },
    {
      "entropy": 6.143211412429809,
      "epoch": 1.2072882985231146,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048597579085001984,
      "loss": 5.8447,
      "mean_token_accuracy": 0.15078818202018737,
      "num_tokens": 27988383.0,
      "step": 13815
    },
    {
      "entropy": 6.095460319519043,
      "epoch": 1.2077252468758193,
      "grad_norm": 5.875,
      "learning_rate": 0.00048596495952110176,
      "loss": 5.8608,
      "mean_token_accuracy": 0.15658497512340547,
      "num_tokens": 27999118.0,
      "step": 13820
    },
    {
      "entropy": 6.140877056121826,
      "epoch": 1.208162195228524,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004859541241457651,
      "loss": 5.8461,
      "mean_token_accuracy": 0.15611317455768586,
      "num_tokens": 28008849.0,
      "step": 13825
    },
    {
      "entropy": 6.238790130615234,
      "epoch": 1.2085991435812287,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004859432847242179,
      "loss": 5.9392,
      "mean_token_accuracy": 0.14997541755437852,
      "num_tokens": 28019063.0,
      "step": 13830
    },
    {
      "entropy": 6.172117805480957,
      "epoch": 1.2090360919339334,
      "grad_norm": 3.578125,
      "learning_rate": 0.00048593244125666793,
      "loss": 5.8229,
      "mean_token_accuracy": 0.15812181681394577,
      "num_tokens": 28028333.0,
      "step": 13835
    },
    {
      "entropy": 6.1784388542175295,
      "epoch": 1.209473040286638,
      "grad_norm": 2.625,
      "learning_rate": 0.0004859215937433232,
      "loss": 5.9117,
      "mean_token_accuracy": 0.1518342137336731,
      "num_tokens": 28037704.0,
      "step": 13840
    },
    {
      "entropy": 6.186066818237305,
      "epoch": 1.2099099886393427,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004859107421843918,
      "loss": 5.928,
      "mean_token_accuracy": 0.15025399029254913,
      "num_tokens": 28047157.0,
      "step": 13845
    },
    {
      "entropy": 6.176485013961792,
      "epoch": 1.2103469369920474,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048589988658008184,
      "loss": 5.8577,
      "mean_token_accuracy": 0.15271882265806197,
      "num_tokens": 28057582.0,
      "step": 13850
    },
    {
      "entropy": 6.161338520050049,
      "epoch": 1.2107838853447523,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048588902693060156,
      "loss": 5.921,
      "mean_token_accuracy": 0.1398226022720337,
      "num_tokens": 28067458.0,
      "step": 13855
    },
    {
      "entropy": 6.277706146240234,
      "epoch": 1.211220833697457,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004858781632361592,
      "loss": 5.9029,
      "mean_token_accuracy": 0.15111400932073593,
      "num_tokens": 28078967.0,
      "step": 13860
    },
    {
      "entropy": 6.1715000629425045,
      "epoch": 1.2116577820501617,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004858672954969632,
      "loss": 5.8585,
      "mean_token_accuracy": 0.1534536749124527,
      "num_tokens": 28088498.0,
      "step": 13865
    },
    {
      "entropy": 6.064240789413452,
      "epoch": 1.2120947304028664,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048585642371322196,
      "loss": 5.9441,
      "mean_token_accuracy": 0.14823956340551375,
      "num_tokens": 28098884.0,
      "step": 13870
    },
    {
      "entropy": 6.179372549057007,
      "epoch": 1.212531678755571,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004858455478851441,
      "loss": 5.8704,
      "mean_token_accuracy": 0.15619773864746095,
      "num_tokens": 28109629.0,
      "step": 13875
    },
    {
      "entropy": 6.195692491531372,
      "epoch": 1.2129686271082758,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000485834668012938,
      "loss": 5.8296,
      "mean_token_accuracy": 0.15061113387346267,
      "num_tokens": 28118555.0,
      "step": 13880
    },
    {
      "entropy": 6.197376346588134,
      "epoch": 1.2134055754609805,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004858237840968127,
      "loss": 5.9593,
      "mean_token_accuracy": 0.14522878602147102,
      "num_tokens": 28129028.0,
      "step": 13885
    },
    {
      "entropy": 6.117794990539551,
      "epoch": 1.2138425238136852,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048581289613697654,
      "loss": 5.9193,
      "mean_token_accuracy": 0.14912183135747908,
      "num_tokens": 28139188.0,
      "step": 13890
    },
    {
      "entropy": 6.159627056121826,
      "epoch": 1.2142794721663899,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048580200413363866,
      "loss": 5.8633,
      "mean_token_accuracy": 0.15412092655897142,
      "num_tokens": 28148199.0,
      "step": 13895
    },
    {
      "entropy": 6.163512134552002,
      "epoch": 1.2147164205190946,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004857911080870079,
      "loss": 5.8396,
      "mean_token_accuracy": 0.15558690428733826,
      "num_tokens": 28157811.0,
      "step": 13900
    },
    {
      "entropy": 6.101589059829712,
      "epoch": 1.2151533688717993,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048578020799729327,
      "loss": 5.857,
      "mean_token_accuracy": 0.15734661221504212,
      "num_tokens": 28168644.0,
      "step": 13905
    },
    {
      "entropy": 6.158587074279785,
      "epoch": 1.2155903172245042,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004857693038647037,
      "loss": 5.9259,
      "mean_token_accuracy": 0.14993413537740707,
      "num_tokens": 28179846.0,
      "step": 13910
    },
    {
      "entropy": 6.212523889541626,
      "epoch": 1.2160272655772089,
      "grad_norm": 6.34375,
      "learning_rate": 0.0004857583956894485,
      "loss": 5.8776,
      "mean_token_accuracy": 0.15534001141786574,
      "num_tokens": 28189400.0,
      "step": 13915
    },
    {
      "entropy": 6.139555788040161,
      "epoch": 1.2164642139299136,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004857474834717368,
      "loss": 5.951,
      "mean_token_accuracy": 0.14881992489099502,
      "num_tokens": 28199695.0,
      "step": 13920
    },
    {
      "entropy": 6.1129858016967775,
      "epoch": 1.2169011622826182,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048573656721177805,
      "loss": 5.8692,
      "mean_token_accuracy": 0.15997774451971053,
      "num_tokens": 28209977.0,
      "step": 13925
    },
    {
      "entropy": 6.212561178207397,
      "epoch": 1.217338110635323,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004857256469097814,
      "loss": 5.931,
      "mean_token_accuracy": 0.15285259187221528,
      "num_tokens": 28220619.0,
      "step": 13930
    },
    {
      "entropy": 6.156783628463745,
      "epoch": 1.2177750589880276,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004857147225659564,
      "loss": 5.8825,
      "mean_token_accuracy": 0.15748235881328582,
      "num_tokens": 28230065.0,
      "step": 13935
    },
    {
      "entropy": 6.115792846679687,
      "epoch": 1.2182120073407323,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048570379418051273,
      "loss": 5.8678,
      "mean_token_accuracy": 0.15798841714859008,
      "num_tokens": 28239307.0,
      "step": 13940
    },
    {
      "entropy": 6.199590015411377,
      "epoch": 1.218648955693437,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004856928617536598,
      "loss": 5.8483,
      "mean_token_accuracy": 0.15174517184495925,
      "num_tokens": 28249421.0,
      "step": 13945
    },
    {
      "entropy": 6.088633441925049,
      "epoch": 1.2190859040461417,
      "grad_norm": 3.5,
      "learning_rate": 0.0004856819252856073,
      "loss": 5.8415,
      "mean_token_accuracy": 0.15172751992940903,
      "num_tokens": 28258125.0,
      "step": 13950
    },
    {
      "entropy": 6.005021095275879,
      "epoch": 1.2195228523988464,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048567098477656526,
      "loss": 5.8553,
      "mean_token_accuracy": 0.1543092504143715,
      "num_tokens": 28268897.0,
      "step": 13955
    },
    {
      "entropy": 6.118499994277954,
      "epoch": 1.219959800751551,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004856600402267432,
      "loss": 5.8847,
      "mean_token_accuracy": 0.15268064066767692,
      "num_tokens": 28278506.0,
      "step": 13960
    },
    {
      "entropy": 6.2342699527740475,
      "epoch": 1.220396749104256,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004856490916363513,
      "loss": 5.8577,
      "mean_token_accuracy": 0.15139369666576385,
      "num_tokens": 28287905.0,
      "step": 13965
    },
    {
      "entropy": 6.190746784210205,
      "epoch": 1.2208336974569605,
      "grad_norm": 2.0,
      "learning_rate": 0.0004856381390055994,
      "loss": 5.9371,
      "mean_token_accuracy": 0.1514703616499901,
      "num_tokens": 28297123.0,
      "step": 13970
    },
    {
      "entropy": 6.140774345397949,
      "epoch": 1.2212706458096654,
      "grad_norm": 2.71875,
      "learning_rate": 0.00048562718233469756,
      "loss": 5.9523,
      "mean_token_accuracy": 0.14569918811321259,
      "num_tokens": 28306621.0,
      "step": 13975
    },
    {
      "entropy": 6.2415910243988035,
      "epoch": 1.22170759416237,
      "grad_norm": 2.578125,
      "learning_rate": 0.00048561622162385606,
      "loss": 6.0122,
      "mean_token_accuracy": 0.1411128021776676,
      "num_tokens": 28318369.0,
      "step": 13980
    },
    {
      "entropy": 6.2192271709442135,
      "epoch": 1.2221445425150748,
      "grad_norm": 6.09375,
      "learning_rate": 0.00048560525687328497,
      "loss": 5.9434,
      "mean_token_accuracy": 0.14681155383586883,
      "num_tokens": 28328817.0,
      "step": 13985
    },
    {
      "entropy": 6.235672283172607,
      "epoch": 1.2225814908677795,
      "grad_norm": 3.484375,
      "learning_rate": 0.00048559428808319483,
      "loss": 5.9738,
      "mean_token_accuracy": 0.14341663420200348,
      "num_tokens": 28339448.0,
      "step": 13990
    },
    {
      "entropy": 6.175102710723877,
      "epoch": 1.2230184392204841,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004855833152537958,
      "loss": 5.8812,
      "mean_token_accuracy": 0.15200393050909042,
      "num_tokens": 28349243.0,
      "step": 13995
    },
    {
      "entropy": 6.1426825523376465,
      "epoch": 1.2234553875731888,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004855723383852985,
      "loss": 5.9229,
      "mean_token_accuracy": 0.15116727948188782,
      "num_tokens": 28359392.0,
      "step": 14000
    },
    {
      "entropy": 6.176077270507813,
      "epoch": 1.2238923359258935,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004855613574779134,
      "loss": 5.8477,
      "mean_token_accuracy": 0.15116518288850783,
      "num_tokens": 28369160.0,
      "step": 14005
    },
    {
      "entropy": 6.183259916305542,
      "epoch": 1.2243292842785982,
      "grad_norm": 5.34375,
      "learning_rate": 0.00048555037253185114,
      "loss": 5.9235,
      "mean_token_accuracy": 0.15362579822540284,
      "num_tokens": 28379107.0,
      "step": 14010
    },
    {
      "entropy": 6.18773832321167,
      "epoch": 1.224766232631303,
      "grad_norm": 2.78125,
      "learning_rate": 0.00048553938354732243,
      "loss": 5.9218,
      "mean_token_accuracy": 0.14881558418273927,
      "num_tokens": 28388956.0,
      "step": 14015
    },
    {
      "entropy": 6.153997135162354,
      "epoch": 1.2252031809840076,
      "grad_norm": 4.5625,
      "learning_rate": 0.00048552839052453794,
      "loss": 5.8899,
      "mean_token_accuracy": 0.14805519729852676,
      "num_tokens": 28398925.0,
      "step": 14020
    },
    {
      "entropy": 6.20629563331604,
      "epoch": 1.2256401293367123,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048551739346370864,
      "loss": 5.9681,
      "mean_token_accuracy": 0.13878847360610963,
      "num_tokens": 28408714.0,
      "step": 14025
    },
    {
      "entropy": 6.247835731506347,
      "epoch": 1.2260770776894172,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004855063923650455,
      "loss": 5.941,
      "mean_token_accuracy": 0.15068021565675735,
      "num_tokens": 28419408.0,
      "step": 14030
    },
    {
      "entropy": 6.126561450958252,
      "epoch": 1.226514026042122,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004854953872287593,
      "loss": 5.8724,
      "mean_token_accuracy": 0.1522832691669464,
      "num_tokens": 28430044.0,
      "step": 14035
    },
    {
      "entropy": 6.119768047332764,
      "epoch": 1.2269509743948266,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004854843780550614,
      "loss": 5.8326,
      "mean_token_accuracy": 0.15552353262901306,
      "num_tokens": 28440100.0,
      "step": 14040
    },
    {
      "entropy": 6.137379693984985,
      "epoch": 1.2273879227475313,
      "grad_norm": 4.75,
      "learning_rate": 0.0004854733648441628,
      "loss": 5.848,
      "mean_token_accuracy": 0.1561823680996895,
      "num_tokens": 28449497.0,
      "step": 14045
    },
    {
      "entropy": 6.177059316635132,
      "epoch": 1.227824871100236,
      "grad_norm": 4.40625,
      "learning_rate": 0.00048546234759627477,
      "loss": 5.8933,
      "mean_token_accuracy": 0.15604597926139832,
      "num_tokens": 28459789.0,
      "step": 14050
    },
    {
      "entropy": 6.245036840438843,
      "epoch": 1.2282618194529407,
      "grad_norm": 3.296875,
      "learning_rate": 0.00048545132631160856,
      "loss": 5.957,
      "mean_token_accuracy": 0.1454416736960411,
      "num_tokens": 28471728.0,
      "step": 14055
    },
    {
      "entropy": 6.240487480163575,
      "epoch": 1.2286987678056454,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048544030099037576,
      "loss": 5.94,
      "mean_token_accuracy": 0.15105270743370056,
      "num_tokens": 28481876.0,
      "step": 14060
    },
    {
      "entropy": 6.210722970962524,
      "epoch": 1.22913571615835,
      "grad_norm": 2.125,
      "learning_rate": 0.00048542927163278766,
      "loss": 5.9094,
      "mean_token_accuracy": 0.150232669711113,
      "num_tokens": 28491800.0,
      "step": 14065
    },
    {
      "entropy": 6.079072761535644,
      "epoch": 1.2295726645110547,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004854182382390559,
      "loss": 5.8662,
      "mean_token_accuracy": 0.1539934977889061,
      "num_tokens": 28502033.0,
      "step": 14070
    },
    {
      "entropy": 6.143470764160156,
      "epoch": 1.2300096128637594,
      "grad_norm": 2.859375,
      "learning_rate": 0.000485407200809392,
      "loss": 5.937,
      "mean_token_accuracy": 0.14482716843485832,
      "num_tokens": 28511921.0,
      "step": 14075
    },
    {
      "entropy": 6.253262233734131,
      "epoch": 1.2304465612164641,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004853961593440078,
      "loss": 5.8849,
      "mean_token_accuracy": 0.15441647917032242,
      "num_tokens": 28521449.0,
      "step": 14080
    },
    {
      "entropy": 6.197099256515503,
      "epoch": 1.230883509569169,
      "grad_norm": 2.265625,
      "learning_rate": 0.000485385113843115,
      "loss": 5.8801,
      "mean_token_accuracy": 0.15310391336679458,
      "num_tokens": 28532088.0,
      "step": 14085
    },
    {
      "entropy": 6.2283447265625,
      "epoch": 1.2313204579218737,
      "grad_norm": 3.75,
      "learning_rate": 0.0004853740643069255,
      "loss": 5.9268,
      "mean_token_accuracy": 0.14941720366477967,
      "num_tokens": 28541835.0,
      "step": 14090
    },
    {
      "entropy": 6.197736835479736,
      "epoch": 1.2317574062745784,
      "grad_norm": 5.6875,
      "learning_rate": 0.00048536301073565113,
      "loss": 5.9061,
      "mean_token_accuracy": 0.14509299993515015,
      "num_tokens": 28551601.0,
      "step": 14095
    },
    {
      "entropy": 6.148233699798584,
      "epoch": 1.232194354627283,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004853519531295041,
      "loss": 5.8651,
      "mean_token_accuracy": 0.1543732464313507,
      "num_tokens": 28561012.0,
      "step": 14100
    },
    {
      "entropy": 6.177007007598877,
      "epoch": 1.2326313029799878,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004853408914886963,
      "loss": 5.9669,
      "mean_token_accuracy": 0.1462666854262352,
      "num_tokens": 28571431.0,
      "step": 14105
    },
    {
      "entropy": 6.159669589996338,
      "epoch": 1.2330682513326925,
      "grad_norm": 3.4375,
      "learning_rate": 0.00048532982581344004,
      "loss": 5.8383,
      "mean_token_accuracy": 0.15932746529579161,
      "num_tokens": 28581185.0,
      "step": 14110
    },
    {
      "entropy": 6.150724983215332,
      "epoch": 1.2335051996853972,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004853187561039475,
      "loss": 5.8828,
      "mean_token_accuracy": 0.15079872608184813,
      "num_tokens": 28591164.0,
      "step": 14115
    },
    {
      "entropy": 6.100163269042969,
      "epoch": 1.2339421480381019,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048530768236043094,
      "loss": 5.7967,
      "mean_token_accuracy": 0.1576806828379631,
      "num_tokens": 28600931.0,
      "step": 14120
    },
    {
      "entropy": 6.145612096786499,
      "epoch": 1.2343790963908066,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004852966045831029,
      "loss": 5.9577,
      "mean_token_accuracy": 0.14835426062345505,
      "num_tokens": 28611908.0,
      "step": 14125
    },
    {
      "entropy": 6.226451396942139,
      "epoch": 1.2348160447435113,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004852855227721758,
      "loss": 5.9716,
      "mean_token_accuracy": 0.14340434670448304,
      "num_tokens": 28622106.0,
      "step": 14130
    },
    {
      "entropy": 6.262969732284546,
      "epoch": 1.235252993096216,
      "grad_norm": 3.328125,
      "learning_rate": 0.00048527443692786217,
      "loss": 5.9967,
      "mean_token_accuracy": 0.14171316847205162,
      "num_tokens": 28632634.0,
      "step": 14135
    },
    {
      "entropy": 6.206275415420532,
      "epoch": 1.2356899414489209,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004852633470503746,
      "loss": 5.9305,
      "mean_token_accuracy": 0.14711518213152885,
      "num_tokens": 28642936.0,
      "step": 14140
    },
    {
      "entropy": 6.25238356590271,
      "epoch": 1.2361268898016253,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048525225313992594,
      "loss": 5.8888,
      "mean_token_accuracy": 0.15033729523420333,
      "num_tokens": 28653539.0,
      "step": 14145
    },
    {
      "entropy": 6.1946642875671385,
      "epoch": 1.2365638381543302,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004852411551967289,
      "loss": 5.9603,
      "mean_token_accuracy": 0.14472445994615554,
      "num_tokens": 28664212.0,
      "step": 14150
    },
    {
      "entropy": 6.2647181987762455,
      "epoch": 1.237000786507035,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048523005322099626,
      "loss": 5.9567,
      "mean_token_accuracy": 0.14432160556316376,
      "num_tokens": 28676028.0,
      "step": 14155
    },
    {
      "entropy": 6.234403800964356,
      "epoch": 1.2374377348597396,
      "grad_norm": 3.96875,
      "learning_rate": 0.00048521894721294115,
      "loss": 5.8825,
      "mean_token_accuracy": 0.14716139659285546,
      "num_tokens": 28685370.0,
      "step": 14160
    },
    {
      "entropy": 6.154047298431396,
      "epoch": 1.2378746832124443,
      "grad_norm": 2.75,
      "learning_rate": 0.00048520783717277645,
      "loss": 5.8782,
      "mean_token_accuracy": 0.1565360024571419,
      "num_tokens": 28695300.0,
      "step": 14165
    },
    {
      "entropy": 6.1391188621521,
      "epoch": 1.238311631565149,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048519672310071523,
      "loss": 5.9016,
      "mean_token_accuracy": 0.15393647998571397,
      "num_tokens": 28705485.0,
      "step": 14170
    },
    {
      "entropy": 6.192353963851929,
      "epoch": 1.2387485799178537,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048518560499697066,
      "loss": 5.9153,
      "mean_token_accuracy": 0.14622731655836105,
      "num_tokens": 28715407.0,
      "step": 14175
    },
    {
      "entropy": 6.211352300643921,
      "epoch": 1.2391855282705584,
      "grad_norm": 3.875,
      "learning_rate": 0.00048517448286175617,
      "loss": 5.9353,
      "mean_token_accuracy": 0.15140830874443054,
      "num_tokens": 28725283.0,
      "step": 14180
    },
    {
      "entropy": 6.13202428817749,
      "epoch": 1.239622476623263,
      "grad_norm": 4.65625,
      "learning_rate": 0.00048516335669528495,
      "loss": 5.8496,
      "mean_token_accuracy": 0.15868661999702455,
      "num_tokens": 28736131.0,
      "step": 14185
    },
    {
      "entropy": 6.110800886154175,
      "epoch": 1.2400594249759678,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004851522264977704,
      "loss": 5.8742,
      "mean_token_accuracy": 0.15677210539579392,
      "num_tokens": 28745446.0,
      "step": 14190
    },
    {
      "entropy": 6.157938051223755,
      "epoch": 1.2404963733286725,
      "grad_norm": 2.484375,
      "learning_rate": 0.000485141092269426,
      "loss": 5.8145,
      "mean_token_accuracy": 0.15657073333859445,
      "num_tokens": 28754639.0,
      "step": 14195
    },
    {
      "entropy": 6.164085483551025,
      "epoch": 1.2409333216813772,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004851299540104654,
      "loss": 5.9198,
      "mean_token_accuracy": 0.15186750441789626,
      "num_tokens": 28764397.0,
      "step": 14200
    },
    {
      "entropy": 6.126307487487793,
      "epoch": 1.241370270034082,
      "grad_norm": 3.4375,
      "learning_rate": 0.00048511881172110216,
      "loss": 5.8665,
      "mean_token_accuracy": 0.15348217338323594,
      "num_tokens": 28773680.0,
      "step": 14205
    },
    {
      "entropy": 6.2114870071411135,
      "epoch": 1.2418072183867868,
      "grad_norm": 2.375,
      "learning_rate": 0.00048510766540154997,
      "loss": 5.938,
      "mean_token_accuracy": 0.1543569818139076,
      "num_tokens": 28784034.0,
      "step": 14210
    },
    {
      "entropy": 6.178019046783447,
      "epoch": 1.2422441667394915,
      "grad_norm": 9.6875,
      "learning_rate": 0.00048509651505202275,
      "loss": 5.9514,
      "mean_token_accuracy": 0.14725581854581832,
      "num_tokens": 28794487.0,
      "step": 14215
    },
    {
      "entropy": 6.205572175979614,
      "epoch": 1.2426811150921961,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004850853606727342,
      "loss": 5.8605,
      "mean_token_accuracy": 0.1567555233836174,
      "num_tokens": 28803942.0,
      "step": 14220
    },
    {
      "entropy": 6.22735013961792,
      "epoch": 1.2431180634449008,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004850742022638984,
      "loss": 5.8838,
      "mean_token_accuracy": 0.14974181205034257,
      "num_tokens": 28814498.0,
      "step": 14225
    },
    {
      "entropy": 6.124869060516358,
      "epoch": 1.2435550117976055,
      "grad_norm": 2.484375,
      "learning_rate": 0.00048506303982572924,
      "loss": 5.8581,
      "mean_token_accuracy": 0.15606651157140733,
      "num_tokens": 28824513.0,
      "step": 14230
    },
    {
      "entropy": 6.208728075027466,
      "epoch": 1.2439919601503102,
      "grad_norm": 2.75,
      "learning_rate": 0.000485051873358441,
      "loss": 5.9939,
      "mean_token_accuracy": 0.13807140439748763,
      "num_tokens": 28835643.0,
      "step": 14235
    },
    {
      "entropy": 6.300460147857666,
      "epoch": 1.244428908503015,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004850407028622477,
      "loss": 6.0243,
      "mean_token_accuracy": 0.1439963012933731,
      "num_tokens": 28845615.0,
      "step": 14240
    },
    {
      "entropy": 6.223985242843628,
      "epoch": 1.2448658568557196,
      "grad_norm": 2.375,
      "learning_rate": 0.00048502952833736365,
      "loss": 5.8938,
      "mean_token_accuracy": 0.1521884098649025,
      "num_tokens": 28856152.0,
      "step": 14245
    },
    {
      "entropy": 6.114186573028564,
      "epoch": 1.2453028052084243,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048501834978400323,
      "loss": 5.8207,
      "mean_token_accuracy": 0.1594656988978386,
      "num_tokens": 28866926.0,
      "step": 14250
    },
    {
      "entropy": 6.1715654850006105,
      "epoch": 1.245739753561129,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004850071672023809,
      "loss": 5.9053,
      "mean_token_accuracy": 0.14316325187683104,
      "num_tokens": 28877188.0,
      "step": 14255
    },
    {
      "entropy": 6.161950588226318,
      "epoch": 1.246176701913834,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00048499598059271087,
      "loss": 5.9003,
      "mean_token_accuracy": 0.15031933039426804,
      "num_tokens": 28887710.0,
      "step": 14260
    },
    {
      "entropy": 6.179935646057129,
      "epoch": 1.2466136502665386,
      "grad_norm": 3.25,
      "learning_rate": 0.000484984789955208,
      "loss": 5.9564,
      "mean_token_accuracy": 0.14734355956315995,
      "num_tokens": 28897064.0,
      "step": 14265
    },
    {
      "entropy": 6.162792730331421,
      "epoch": 1.2470505986192433,
      "grad_norm": 2.25,
      "learning_rate": 0.00048497359529008684,
      "loss": 5.9389,
      "mean_token_accuracy": 0.14970822632312775,
      "num_tokens": 28907103.0,
      "step": 14270
    },
    {
      "entropy": 6.175220203399658,
      "epoch": 1.247487546971948,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004849623965975621,
      "loss": 5.9142,
      "mean_token_accuracy": 0.15144832134246827,
      "num_tokens": 28916851.0,
      "step": 14275
    },
    {
      "entropy": 6.184738969802856,
      "epoch": 1.2479244953246527,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004849511938778485,
      "loss": 5.939,
      "mean_token_accuracy": 0.14545366317033767,
      "num_tokens": 28926638.0,
      "step": 14280
    },
    {
      "entropy": 6.281378841400146,
      "epoch": 1.2483614436773574,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048493998713116103,
      "loss": 5.9008,
      "mean_token_accuracy": 0.1458335377275944,
      "num_tokens": 28936764.0,
      "step": 14285
    },
    {
      "entropy": 6.153763341903686,
      "epoch": 1.248798392030062,
      "grad_norm": 6.28125,
      "learning_rate": 0.00048492877635771464,
      "loss": 5.9097,
      "mean_token_accuracy": 0.14964258149266244,
      "num_tokens": 28946494.0,
      "step": 14290
    },
    {
      "entropy": 6.078196859359741,
      "epoch": 1.2492353403827667,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048491756155772433,
      "loss": 5.8633,
      "mean_token_accuracy": 0.1571328654885292,
      "num_tokens": 28955626.0,
      "step": 14295
    },
    {
      "entropy": 6.212883758544922,
      "epoch": 1.2496722887354714,
      "grad_norm": 2.546875,
      "learning_rate": 0.00048490634273140517,
      "loss": 5.9133,
      "mean_token_accuracy": 0.1500519648194313,
      "num_tokens": 28966317.0,
      "step": 14300
    },
    {
      "entropy": 6.1862303733825685,
      "epoch": 1.2501092370881761,
      "grad_norm": 2.828125,
      "learning_rate": 0.00048489511987897236,
      "loss": 5.9262,
      "mean_token_accuracy": 0.14952007979154586,
      "num_tokens": 28976063.0,
      "step": 14305
    },
    {
      "entropy": 6.153039216995239,
      "epoch": 1.2505461854408808,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004848838930006411,
      "loss": 5.9168,
      "mean_token_accuracy": 0.14818838983774185,
      "num_tokens": 28985789.0,
      "step": 14310
    },
    {
      "entropy": 6.097548723220825,
      "epoch": 1.2509831337935857,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004848726620966269,
      "loss": 5.9753,
      "mean_token_accuracy": 0.14401112347841263,
      "num_tokens": 28996458.0,
      "step": 14315
    },
    {
      "entropy": 6.223761940002442,
      "epoch": 1.2514200821462902,
      "grad_norm": 2.71875,
      "learning_rate": 0.000484861427167145,
      "loss": 5.9032,
      "mean_token_accuracy": 0.15155038833618165,
      "num_tokens": 29007641.0,
      "step": 14320
    },
    {
      "entropy": 6.279244327545166,
      "epoch": 1.251857030498995,
      "grad_norm": 3.78125,
      "learning_rate": 0.00048485018821241104,
      "loss": 5.9027,
      "mean_token_accuracy": 0.14320011883974076,
      "num_tokens": 29017351.0,
      "step": 14325
    },
    {
      "entropy": 6.154199171066284,
      "epoch": 1.2522939788516998,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048483894523264045,
      "loss": 5.9518,
      "mean_token_accuracy": 0.1451245054602623,
      "num_tokens": 29026951.0,
      "step": 14330
    },
    {
      "entropy": 6.141295862197876,
      "epoch": 1.2527309272044045,
      "grad_norm": 1.875,
      "learning_rate": 0.0004848276982280489,
      "loss": 5.9114,
      "mean_token_accuracy": 0.15095143765211105,
      "num_tokens": 29037596.0,
      "step": 14335
    },
    {
      "entropy": 6.18510365486145,
      "epoch": 1.2531678755571092,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004848164471988522,
      "loss": 5.9092,
      "mean_token_accuracy": 0.15030819922685623,
      "num_tokens": 29049010.0,
      "step": 14340
    },
    {
      "entropy": 6.204849195480347,
      "epoch": 1.2536048239098139,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048480519214526616,
      "loss": 5.9164,
      "mean_token_accuracy": 0.15004382357001306,
      "num_tokens": 29059028.0,
      "step": 14345
    },
    {
      "entropy": 6.18299822807312,
      "epoch": 1.2540417722625186,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048479393306750647,
      "loss": 5.91,
      "mean_token_accuracy": 0.14778466001152993,
      "num_tokens": 29069661.0,
      "step": 14350
    },
    {
      "entropy": 6.192969179153442,
      "epoch": 1.2544787206152233,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004847826699657893,
      "loss": 5.9081,
      "mean_token_accuracy": 0.15246857404708863,
      "num_tokens": 29079971.0,
      "step": 14355
    },
    {
      "entropy": 6.177361106872558,
      "epoch": 1.254915668967928,
      "grad_norm": 2.25,
      "learning_rate": 0.00048477140284033054,
      "loss": 5.9444,
      "mean_token_accuracy": 0.14768177792429923,
      "num_tokens": 29091378.0,
      "step": 14360
    },
    {
      "entropy": 6.168608331680298,
      "epoch": 1.2553526173206326,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004847601316913464,
      "loss": 5.874,
      "mean_token_accuracy": 0.1504699133336544,
      "num_tokens": 29101516.0,
      "step": 14365
    },
    {
      "entropy": 6.149482154846192,
      "epoch": 1.2557895656733375,
      "grad_norm": 2.765625,
      "learning_rate": 0.000484748856519053,
      "loss": 5.9651,
      "mean_token_accuracy": 0.14562420397996903,
      "num_tokens": 29112147.0,
      "step": 14370
    },
    {
      "entropy": 6.199831819534301,
      "epoch": 1.256226514026042,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004847375773236667,
      "loss": 5.8537,
      "mean_token_accuracy": 0.1556869275867939,
      "num_tokens": 29122084.0,
      "step": 14375
    },
    {
      "entropy": 6.222476291656494,
      "epoch": 1.256663462378747,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004847262941054037,
      "loss": 5.9288,
      "mean_token_accuracy": 0.1485342875123024,
      "num_tokens": 29131988.0,
      "step": 14380
    },
    {
      "entropy": 6.153385162353516,
      "epoch": 1.2571004107314516,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004847150068644805,
      "loss": 5.9683,
      "mean_token_accuracy": 0.14586437940597535,
      "num_tokens": 29141731.0,
      "step": 14385
    },
    {
      "entropy": 6.234918069839478,
      "epoch": 1.2575373590841563,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048470371560111353,
      "loss": 5.9764,
      "mean_token_accuracy": 0.14204879105091095,
      "num_tokens": 29152185.0,
      "step": 14390
    },
    {
      "entropy": 6.17518162727356,
      "epoch": 1.257974307436861,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004846924203155194,
      "loss": 5.8084,
      "mean_token_accuracy": 0.15808162093162537,
      "num_tokens": 29162337.0,
      "step": 14395
    },
    {
      "entropy": 6.174739646911621,
      "epoch": 1.2584112557895657,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048468112100791483,
      "loss": 5.9379,
      "mean_token_accuracy": 0.14998275637626649,
      "num_tokens": 29173246.0,
      "step": 14400
    },
    {
      "entropy": 6.193538475036621,
      "epoch": 1.2588482041422704,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004846698176785165,
      "loss": 5.8916,
      "mean_token_accuracy": 0.15311593413352967,
      "num_tokens": 29183700.0,
      "step": 14405
    },
    {
      "entropy": 6.180676078796386,
      "epoch": 1.259285152494975,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004846585103275412,
      "loss": 5.8798,
      "mean_token_accuracy": 0.1494651809334755,
      "num_tokens": 29193686.0,
      "step": 14410
    },
    {
      "entropy": 6.096537685394287,
      "epoch": 1.2597221008476798,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048464719895520577,
      "loss": 5.9186,
      "mean_token_accuracy": 0.15009548962116243,
      "num_tokens": 29203575.0,
      "step": 14415
    },
    {
      "entropy": 6.14689998626709,
      "epoch": 1.2601590492003845,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004846358835617273,
      "loss": 5.9124,
      "mean_token_accuracy": 0.14776561483740808,
      "num_tokens": 29213591.0,
      "step": 14420
    },
    {
      "entropy": 6.245035982131958,
      "epoch": 1.2605959975530892,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048462456414732257,
      "loss": 5.8942,
      "mean_token_accuracy": 0.15431948006153107,
      "num_tokens": 29222799.0,
      "step": 14425
    },
    {
      "entropy": 6.156334686279297,
      "epoch": 1.2610329459057938,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048461324071220896,
      "loss": 5.8885,
      "mean_token_accuracy": 0.15020613521337509,
      "num_tokens": 29232477.0,
      "step": 14430
    },
    {
      "entropy": 6.164317464828491,
      "epoch": 1.2614698942584988,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048460191325660356,
      "loss": 5.8765,
      "mean_token_accuracy": 0.14821490719914437,
      "num_tokens": 29243477.0,
      "step": 14435
    },
    {
      "entropy": 6.179251861572266,
      "epoch": 1.2619068426112032,
      "grad_norm": 4.15625,
      "learning_rate": 0.00048459058178072366,
      "loss": 5.9527,
      "mean_token_accuracy": 0.14439554736018181,
      "num_tokens": 29253426.0,
      "step": 14440
    },
    {
      "entropy": 6.253608751296997,
      "epoch": 1.2623437909639081,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004845792462847866,
      "loss": 5.9044,
      "mean_token_accuracy": 0.14989247918128967,
      "num_tokens": 29263639.0,
      "step": 14445
    },
    {
      "entropy": 6.1858583927154545,
      "epoch": 1.2627807393166128,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048456790676900963,
      "loss": 5.9368,
      "mean_token_accuracy": 0.14528666511178018,
      "num_tokens": 29273980.0,
      "step": 14450
    },
    {
      "entropy": 6.170363903045654,
      "epoch": 1.2632176876693175,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048455656323361055,
      "loss": 5.8503,
      "mean_token_accuracy": 0.15631430894136428,
      "num_tokens": 29283983.0,
      "step": 14455
    },
    {
      "entropy": 6.245016288757324,
      "epoch": 1.2636546360220222,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004845452156788067,
      "loss": 5.9828,
      "mean_token_accuracy": 0.1483379125595093,
      "num_tokens": 29294460.0,
      "step": 14460
    },
    {
      "entropy": 6.129631805419922,
      "epoch": 1.264091584374727,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004845338641048158,
      "loss": 5.9632,
      "mean_token_accuracy": 0.14798096418380738,
      "num_tokens": 29305788.0,
      "step": 14465
    },
    {
      "entropy": 6.250259923934936,
      "epoch": 1.2645285327274316,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048452250851185565,
      "loss": 5.9422,
      "mean_token_accuracy": 0.14614320248365403,
      "num_tokens": 29315457.0,
      "step": 14470
    },
    {
      "entropy": 6.224140596389771,
      "epoch": 1.2649654810801363,
      "grad_norm": 2.515625,
      "learning_rate": 0.000484511148900144,
      "loss": 5.8709,
      "mean_token_accuracy": 0.1495636910200119,
      "num_tokens": 29325713.0,
      "step": 14475
    },
    {
      "entropy": 6.12435245513916,
      "epoch": 1.265402429432841,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004844997852698987,
      "loss": 5.9023,
      "mean_token_accuracy": 0.14698525816202163,
      "num_tokens": 29336489.0,
      "step": 14480
    },
    {
      "entropy": 6.146484518051148,
      "epoch": 1.2658393777855457,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004844884176213377,
      "loss": 5.8808,
      "mean_token_accuracy": 0.15530710518360138,
      "num_tokens": 29345567.0,
      "step": 14485
    },
    {
      "entropy": 6.2139365673065186,
      "epoch": 1.2662763261382506,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048447704595467917,
      "loss": 5.8864,
      "mean_token_accuracy": 0.15221369564533233,
      "num_tokens": 29355890.0,
      "step": 14490
    },
    {
      "entropy": 6.194058895111084,
      "epoch": 1.266713274490955,
      "grad_norm": 3.125,
      "learning_rate": 0.00048446567027014104,
      "loss": 5.9328,
      "mean_token_accuracy": 0.1531245470046997,
      "num_tokens": 29365948.0,
      "step": 14495
    },
    {
      "entropy": 6.153910350799561,
      "epoch": 1.26715022284366,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004844542905679417,
      "loss": 5.8788,
      "mean_token_accuracy": 0.14753377586603164,
      "num_tokens": 29375673.0,
      "step": 14500
    },
    {
      "entropy": 6.104307842254639,
      "epoch": 1.2675871711963647,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004844429068482993,
      "loss": 5.87,
      "mean_token_accuracy": 0.14662039428949356,
      "num_tokens": 29384567.0,
      "step": 14505
    },
    {
      "entropy": 6.120322132110596,
      "epoch": 1.2680241195490694,
      "grad_norm": 2.125,
      "learning_rate": 0.00048443151911143213,
      "loss": 5.7971,
      "mean_token_accuracy": 0.1598436340689659,
      "num_tokens": 29393747.0,
      "step": 14510
    },
    {
      "entropy": 6.15862078666687,
      "epoch": 1.268461067901774,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004844201273575587,
      "loss": 5.9218,
      "mean_token_accuracy": 0.1484014794230461,
      "num_tokens": 29404095.0,
      "step": 14515
    },
    {
      "entropy": 6.155391836166382,
      "epoch": 1.2688980162544787,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048440873158689754,
      "loss": 5.8901,
      "mean_token_accuracy": 0.15144707411527633,
      "num_tokens": 29414450.0,
      "step": 14520
    },
    {
      "entropy": 6.144805431365967,
      "epoch": 1.2693349646071834,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004843973317996672,
      "loss": 5.8761,
      "mean_token_accuracy": 0.14961023330688478,
      "num_tokens": 29424054.0,
      "step": 14525
    },
    {
      "entropy": 6.23119421005249,
      "epoch": 1.2697719129598881,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004843859279960862,
      "loss": 5.9654,
      "mean_token_accuracy": 0.14579635262489318,
      "num_tokens": 29434585.0,
      "step": 14530
    },
    {
      "entropy": 6.171369028091431,
      "epoch": 1.2702088613125928,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004843745201763735,
      "loss": 5.8982,
      "mean_token_accuracy": 0.15390822887420655,
      "num_tokens": 29443863.0,
      "step": 14535
    },
    {
      "entropy": 6.1620285987854,
      "epoch": 1.2706458096652975,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048436310834074764,
      "loss": 5.8545,
      "mean_token_accuracy": 0.15127672255039215,
      "num_tokens": 29453851.0,
      "step": 14540
    },
    {
      "entropy": 6.121674203872681,
      "epoch": 1.2710827580180022,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004843516924894278,
      "loss": 5.8042,
      "mean_token_accuracy": 0.16123372465372085,
      "num_tokens": 29465152.0,
      "step": 14545
    },
    {
      "entropy": 6.130862045288086,
      "epoch": 1.2715197063707069,
      "grad_norm": 2.96875,
      "learning_rate": 0.00048434027262263274,
      "loss": 5.8334,
      "mean_token_accuracy": 0.1620439700782299,
      "num_tokens": 29474749.0,
      "step": 14550
    },
    {
      "entropy": 6.169938659667968,
      "epoch": 1.2719566547234118,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004843288487405816,
      "loss": 5.8832,
      "mean_token_accuracy": 0.15470795184373856,
      "num_tokens": 29483554.0,
      "step": 14555
    },
    {
      "entropy": 6.127190589904785,
      "epoch": 1.2723936030761165,
      "grad_norm": 2.984375,
      "learning_rate": 0.00048431742084349347,
      "loss": 5.8714,
      "mean_token_accuracy": 0.1568259835243225,
      "num_tokens": 29493718.0,
      "step": 14560
    },
    {
      "entropy": 6.147765874862671,
      "epoch": 1.2728305514288212,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004843059889315874,
      "loss": 5.9602,
      "mean_token_accuracy": 0.14497726261615754,
      "num_tokens": 29504407.0,
      "step": 14565
    },
    {
      "entropy": 6.289806747436524,
      "epoch": 1.2732674997815259,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004842945530050829,
      "loss": 5.9383,
      "mean_token_accuracy": 0.143418525159359,
      "num_tokens": 29514992.0,
      "step": 14570
    },
    {
      "entropy": 6.278067493438721,
      "epoch": 1.2737044481342306,
      "grad_norm": 1.921875,
      "learning_rate": 0.00048428311306419913,
      "loss": 5.9552,
      "mean_token_accuracy": 0.14940451830625534,
      "num_tokens": 29525109.0,
      "step": 14575
    },
    {
      "entropy": 6.171770906448364,
      "epoch": 1.2741413964869353,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048427166910915565,
      "loss": 5.9297,
      "mean_token_accuracy": 0.15071330070495606,
      "num_tokens": 29536131.0,
      "step": 14580
    },
    {
      "entropy": 6.155459690093994,
      "epoch": 1.27457834483964,
      "grad_norm": 5.875,
      "learning_rate": 0.0004842602211401719,
      "loss": 5.88,
      "mean_token_accuracy": 0.1590773269534111,
      "num_tokens": 29547614.0,
      "step": 14585
    },
    {
      "entropy": 6.171518802642822,
      "epoch": 1.2750152931923446,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004842487691574675,
      "loss": 5.9044,
      "mean_token_accuracy": 0.15063830241560935,
      "num_tokens": 29557499.0,
      "step": 14590
    },
    {
      "entropy": 6.212518548965454,
      "epoch": 1.2754522415450493,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048423731316126194,
      "loss": 5.9303,
      "mean_token_accuracy": 0.14245495349168777,
      "num_tokens": 29569909.0,
      "step": 14595
    },
    {
      "entropy": 6.2845064163208,
      "epoch": 1.275889189897754,
      "grad_norm": 4.125,
      "learning_rate": 0.0004842258531517752,
      "loss": 6.0474,
      "mean_token_accuracy": 0.1399131067097187,
      "num_tokens": 29581261.0,
      "step": 14600
    },
    {
      "entropy": 6.194215154647827,
      "epoch": 1.2763261382504587,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004842143891292269,
      "loss": 5.8545,
      "mean_token_accuracy": 0.1560487687587738,
      "num_tokens": 29590604.0,
      "step": 14605
    },
    {
      "entropy": 6.134926891326904,
      "epoch": 1.2767630866031636,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048420292109383704,
      "loss": 5.9543,
      "mean_token_accuracy": 0.14538030028343202,
      "num_tokens": 29600129.0,
      "step": 14610
    },
    {
      "entropy": 6.161518955230713,
      "epoch": 1.277200034955868,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048419144904582545,
      "loss": 5.8293,
      "mean_token_accuracy": 0.15877978950738908,
      "num_tokens": 29609608.0,
      "step": 14615
    },
    {
      "entropy": 6.120277404785156,
      "epoch": 1.277636983308573,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048417997298541235,
      "loss": 5.8078,
      "mean_token_accuracy": 0.15769645273685456,
      "num_tokens": 29618764.0,
      "step": 14620
    },
    {
      "entropy": 6.20949501991272,
      "epoch": 1.2780739316612777,
      "grad_norm": 2.484375,
      "learning_rate": 0.00048416849291281765,
      "loss": 5.9983,
      "mean_token_accuracy": 0.1464912362396717,
      "num_tokens": 29628703.0,
      "step": 14625
    },
    {
      "entropy": 6.209291124343872,
      "epoch": 1.2785108800139824,
      "grad_norm": 7.125,
      "learning_rate": 0.00048415700882826167,
      "loss": 5.9119,
      "mean_token_accuracy": 0.1516139253973961,
      "num_tokens": 29639291.0,
      "step": 14630
    },
    {
      "entropy": 6.132389068603516,
      "epoch": 1.278947828366687,
      "grad_norm": 2.71875,
      "learning_rate": 0.00048414552073196477,
      "loss": 5.895,
      "mean_token_accuracy": 0.15139073729515076,
      "num_tokens": 29649058.0,
      "step": 14635
    },
    {
      "entropy": 6.1872399806976315,
      "epoch": 1.2793847767193918,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004841340286241471,
      "loss": 5.9169,
      "mean_token_accuracy": 0.150474514067173,
      "num_tokens": 29658629.0,
      "step": 14640
    },
    {
      "entropy": 6.2158318042755125,
      "epoch": 1.2798217250720965,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004841225325050292,
      "loss": 5.9749,
      "mean_token_accuracy": 0.14868110716342925,
      "num_tokens": 29668875.0,
      "step": 14645
    },
    {
      "entropy": 6.182331609725952,
      "epoch": 1.2802586734248012,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048411103237483156,
      "loss": 5.9508,
      "mean_token_accuracy": 0.14346095845103263,
      "num_tokens": 29678427.0,
      "step": 14650
    },
    {
      "entropy": 6.166947937011718,
      "epoch": 1.2806956217775058,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048409952823377474,
      "loss": 5.8331,
      "mean_token_accuracy": 0.155053573846817,
      "num_tokens": 29688975.0,
      "step": 14655
    },
    {
      "entropy": 6.2332848072052,
      "epoch": 1.2811325701302105,
      "grad_norm": 3.3125,
      "learning_rate": 0.00048408802008207937,
      "loss": 5.9506,
      "mean_token_accuracy": 0.14689844325184823,
      "num_tokens": 29699550.0,
      "step": 14660
    },
    {
      "entropy": 6.167322015762329,
      "epoch": 1.2815695184829154,
      "grad_norm": 3.828125,
      "learning_rate": 0.00048407650791996624,
      "loss": 5.8898,
      "mean_token_accuracy": 0.15218482315540313,
      "num_tokens": 29709597.0,
      "step": 14665
    },
    {
      "entropy": 6.196347856521607,
      "epoch": 1.28200646683562,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004840649917476561,
      "loss": 5.9377,
      "mean_token_accuracy": 0.14479824751615525,
      "num_tokens": 29719250.0,
      "step": 14670
    },
    {
      "entropy": 6.226090764999389,
      "epoch": 1.2824434151883248,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048405347156536995,
      "loss": 5.9619,
      "mean_token_accuracy": 0.14350830093026162,
      "num_tokens": 29728809.0,
      "step": 14675
    },
    {
      "entropy": 6.182662153244019,
      "epoch": 1.2828803635410295,
      "grad_norm": 9.3125,
      "learning_rate": 0.0004840419473733287,
      "loss": 5.8683,
      "mean_token_accuracy": 0.15773474425077438,
      "num_tokens": 29739240.0,
      "step": 14680
    },
    {
      "entropy": 6.257904052734375,
      "epoch": 1.2833173118937342,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048403041917175323,
      "loss": 5.9348,
      "mean_token_accuracy": 0.15196478217840195,
      "num_tokens": 29749284.0,
      "step": 14685
    },
    {
      "entropy": 6.19420804977417,
      "epoch": 1.283754260246439,
      "grad_norm": 3.59375,
      "learning_rate": 0.00048401888696086485,
      "loss": 5.9377,
      "mean_token_accuracy": 0.1439095050096512,
      "num_tokens": 29758770.0,
      "step": 14690
    },
    {
      "entropy": 6.133532524108887,
      "epoch": 1.2841912085991436,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004840073507408847,
      "loss": 5.8919,
      "mean_token_accuracy": 0.14940415620803832,
      "num_tokens": 29770111.0,
      "step": 14695
    },
    {
      "entropy": 6.124604606628418,
      "epoch": 1.2846281569518483,
      "grad_norm": 3.421875,
      "learning_rate": 0.000483995810512034,
      "loss": 5.866,
      "mean_token_accuracy": 0.1501758873462677,
      "num_tokens": 29780210.0,
      "step": 14700
    },
    {
      "entropy": 6.143610954284668,
      "epoch": 1.285065105304553,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048398426627453423,
      "loss": 5.8915,
      "mean_token_accuracy": 0.1509410858154297,
      "num_tokens": 29789903.0,
      "step": 14705
    },
    {
      "entropy": 6.23986930847168,
      "epoch": 1.2855020536572577,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048397271802860667,
      "loss": 5.9421,
      "mean_token_accuracy": 0.15100280642509462,
      "num_tokens": 29799377.0,
      "step": 14710
    },
    {
      "entropy": 6.223856353759766,
      "epoch": 1.2859390020099624,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004839611657744729,
      "loss": 5.8629,
      "mean_token_accuracy": 0.15203122347593306,
      "num_tokens": 29809173.0,
      "step": 14715
    },
    {
      "entropy": 6.100538730621338,
      "epoch": 1.286375950362667,
      "grad_norm": 4.25,
      "learning_rate": 0.00048394960951235444,
      "loss": 5.9484,
      "mean_token_accuracy": 0.15066004991531373,
      "num_tokens": 29819475.0,
      "step": 14720
    },
    {
      "entropy": 6.101089286804199,
      "epoch": 1.2868128987153717,
      "grad_norm": 3.71875,
      "learning_rate": 0.000483938049242473,
      "loss": 5.8497,
      "mean_token_accuracy": 0.14844564348459244,
      "num_tokens": 29829109.0,
      "step": 14725
    },
    {
      "entropy": 6.2474274158477785,
      "epoch": 1.2872498470680767,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048392648496505024,
      "loss": 5.865,
      "mean_token_accuracy": 0.15116432979702948,
      "num_tokens": 29839735.0,
      "step": 14730
    },
    {
      "entropy": 6.202255153656006,
      "epoch": 1.2876867954207811,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004839149166803081,
      "loss": 5.8653,
      "mean_token_accuracy": 0.1497869923710823,
      "num_tokens": 29850039.0,
      "step": 14735
    },
    {
      "entropy": 6.1789017677307125,
      "epoch": 1.288123743773486,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004839033443884684,
      "loss": 5.8675,
      "mean_token_accuracy": 0.1542421191930771,
      "num_tokens": 29859411.0,
      "step": 14740
    },
    {
      "entropy": 6.182638645172119,
      "epoch": 1.2885606921261907,
      "grad_norm": 4.625,
      "learning_rate": 0.0004838917680897531,
      "loss": 5.977,
      "mean_token_accuracy": 0.14336021691560746,
      "num_tokens": 29870426.0,
      "step": 14745
    },
    {
      "entropy": 6.197987842559814,
      "epoch": 1.2889976404788954,
      "grad_norm": 2.734375,
      "learning_rate": 0.00048388018778438413,
      "loss": 5.8966,
      "mean_token_accuracy": 0.14952237159013748,
      "num_tokens": 29881992.0,
      "step": 14750
    },
    {
      "entropy": 6.1714198112487795,
      "epoch": 1.2894345888316001,
      "grad_norm": 2.125,
      "learning_rate": 0.0004838686034725838,
      "loss": 5.8821,
      "mean_token_accuracy": 0.1571191132068634,
      "num_tokens": 29891496.0,
      "step": 14755
    },
    {
      "entropy": 6.186280012130737,
      "epoch": 1.2898715371843048,
      "grad_norm": 2.96875,
      "learning_rate": 0.00048385701515457413,
      "loss": 5.8917,
      "mean_token_accuracy": 0.14586877152323724,
      "num_tokens": 29901529.0,
      "step": 14760
    },
    {
      "entropy": 6.172721338272095,
      "epoch": 1.2903084855370095,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004838454228305775,
      "loss": 5.922,
      "mean_token_accuracy": 0.15179523825645447,
      "num_tokens": 29910773.0,
      "step": 14765
    },
    {
      "entropy": 6.1468113422393795,
      "epoch": 1.2907454338897142,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048383382650081633,
      "loss": 5.9087,
      "mean_token_accuracy": 0.14850216880440711,
      "num_tokens": 29919381.0,
      "step": 14770
    },
    {
      "entropy": 6.1847312450408936,
      "epoch": 1.2911823822424189,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004838222261655129,
      "loss": 5.9324,
      "mean_token_accuracy": 0.1445184215903282,
      "num_tokens": 29929179.0,
      "step": 14775
    },
    {
      "entropy": 6.20357494354248,
      "epoch": 1.2916193305951236,
      "grad_norm": 3.25,
      "learning_rate": 0.00048381062182488964,
      "loss": 5.8986,
      "mean_token_accuracy": 0.15342671424150467,
      "num_tokens": 29938252.0,
      "step": 14780
    },
    {
      "entropy": 6.130641937255859,
      "epoch": 1.2920562789478285,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048379901347916926,
      "loss": 5.9049,
      "mean_token_accuracy": 0.15069559663534166,
      "num_tokens": 29947489.0,
      "step": 14785
    },
    {
      "entropy": 6.114811038970947,
      "epoch": 1.292493227300533,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048378740112857443,
      "loss": 5.8311,
      "mean_token_accuracy": 0.15836550891399384,
      "num_tokens": 29956788.0,
      "step": 14790
    },
    {
      "entropy": 6.205475473403931,
      "epoch": 1.2929301756532379,
      "grad_norm": 6.25,
      "learning_rate": 0.00048377578477332785,
      "loss": 5.94,
      "mean_token_accuracy": 0.1439168445765972,
      "num_tokens": 29967161.0,
      "step": 14795
    },
    {
      "entropy": 6.186266565322876,
      "epoch": 1.2933671240059426,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004837641644136523,
      "loss": 5.7726,
      "mean_token_accuracy": 0.16408921927213668,
      "num_tokens": 29976739.0,
      "step": 14800
    },
    {
      "entropy": 6.1210907936096195,
      "epoch": 1.2938040723586473,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048375254004977074,
      "loss": 5.8829,
      "mean_token_accuracy": 0.15104206800460815,
      "num_tokens": 29986739.0,
      "step": 14805
    },
    {
      "entropy": 6.1187256336212155,
      "epoch": 1.294241020711352,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004837409116819061,
      "loss": 5.8638,
      "mean_token_accuracy": 0.1541775420308113,
      "num_tokens": 29996687.0,
      "step": 14810
    },
    {
      "entropy": 6.1896298885345455,
      "epoch": 1.2946779690640566,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004837292793102813,
      "loss": 5.9114,
      "mean_token_accuracy": 0.15141934752464295,
      "num_tokens": 30007757.0,
      "step": 14815
    },
    {
      "entropy": 6.14250807762146,
      "epoch": 1.2951149174167613,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048371764293511963,
      "loss": 5.874,
      "mean_token_accuracy": 0.1616487592458725,
      "num_tokens": 30018129.0,
      "step": 14820
    },
    {
      "entropy": 6.200066232681275,
      "epoch": 1.295551865769466,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004837060025566441,
      "loss": 5.8952,
      "mean_token_accuracy": 0.14735157638788224,
      "num_tokens": 30028194.0,
      "step": 14825
    },
    {
      "entropy": 6.165783643722534,
      "epoch": 1.2959888141221707,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004836943581750782,
      "loss": 5.9972,
      "mean_token_accuracy": 0.1404227539896965,
      "num_tokens": 30038182.0,
      "step": 14830
    },
    {
      "entropy": 6.258273410797119,
      "epoch": 1.2964257624748754,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004836827097906451,
      "loss": 5.9005,
      "mean_token_accuracy": 0.15336589217185975,
      "num_tokens": 30047848.0,
      "step": 14835
    },
    {
      "entropy": 6.2141883850097654,
      "epoch": 1.2968627108275803,
      "grad_norm": 2.484375,
      "learning_rate": 0.00048367105740356825,
      "loss": 5.993,
      "mean_token_accuracy": 0.15072886943817138,
      "num_tokens": 30058697.0,
      "step": 14840
    },
    {
      "entropy": 6.160948419570923,
      "epoch": 1.2972996591802848,
      "grad_norm": 1.75,
      "learning_rate": 0.00048365940101407123,
      "loss": 5.8247,
      "mean_token_accuracy": 0.15410750061273576,
      "num_tokens": 30068611.0,
      "step": 14845
    },
    {
      "entropy": 6.098754644393921,
      "epoch": 1.2977366075329897,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004836477406223775,
      "loss": 5.8587,
      "mean_token_accuracy": 0.15362597554922103,
      "num_tokens": 30077992.0,
      "step": 14850
    },
    {
      "entropy": 6.339209318161011,
      "epoch": 1.2981735558856944,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048363607622871086,
      "loss": 6.0329,
      "mean_token_accuracy": 0.1438070684671402,
      "num_tokens": 30088940.0,
      "step": 14855
    },
    {
      "entropy": 6.261860942840576,
      "epoch": 1.298610504238399,
      "grad_norm": 9.625,
      "learning_rate": 0.00048362440783329485,
      "loss": 5.9268,
      "mean_token_accuracy": 0.1498449146747589,
      "num_tokens": 30099249.0,
      "step": 14860
    },
    {
      "entropy": 6.087208557128906,
      "epoch": 1.2990474525911038,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004836127354363535,
      "loss": 5.9297,
      "mean_token_accuracy": 0.15286509990692138,
      "num_tokens": 30110548.0,
      "step": 14865
    },
    {
      "entropy": 6.082006931304932,
      "epoch": 1.2994844009438085,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048360105903811045,
      "loss": 5.8974,
      "mean_token_accuracy": 0.1580217644572258,
      "num_tokens": 30120842.0,
      "step": 14870
    },
    {
      "entropy": 6.224267482757568,
      "epoch": 1.2999213492965132,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048358937863878984,
      "loss": 5.9568,
      "mean_token_accuracy": 0.14608903005719184,
      "num_tokens": 30130439.0,
      "step": 14875
    },
    {
      "entropy": 6.233326101303101,
      "epoch": 1.3003582976492178,
      "grad_norm": 3.375,
      "learning_rate": 0.0004835776942386157,
      "loss": 5.937,
      "mean_token_accuracy": 0.14689819365739823,
      "num_tokens": 30140976.0,
      "step": 14880
    },
    {
      "entropy": 6.159067153930664,
      "epoch": 1.3007952460019225,
      "grad_norm": 3.265625,
      "learning_rate": 0.000483566005837812,
      "loss": 5.9035,
      "mean_token_accuracy": 0.15265558809041976,
      "num_tokens": 30150989.0,
      "step": 14885
    },
    {
      "entropy": 6.169952821731568,
      "epoch": 1.3012321943546272,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048355431343660303,
      "loss": 5.8907,
      "mean_token_accuracy": 0.14985596537590026,
      "num_tokens": 30161605.0,
      "step": 14890
    },
    {
      "entropy": 6.190279483795166,
      "epoch": 1.301669142707332,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004835426170352131,
      "loss": 5.8733,
      "mean_token_accuracy": 0.1517595186829567,
      "num_tokens": 30171361.0,
      "step": 14895
    },
    {
      "entropy": 6.1301452159881595,
      "epoch": 1.3021060910600366,
      "grad_norm": 2.375,
      "learning_rate": 0.00048353091663386655,
      "loss": 5.8008,
      "mean_token_accuracy": 0.15861595720052718,
      "num_tokens": 30180352.0,
      "step": 14900
    },
    {
      "entropy": 6.201026821136475,
      "epoch": 1.3025430394127415,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004835192122327877,
      "loss": 5.8525,
      "mean_token_accuracy": 0.15493385791778563,
      "num_tokens": 30190624.0,
      "step": 14905
    },
    {
      "entropy": 6.1833131313323975,
      "epoch": 1.302979987765446,
      "grad_norm": 3.09375,
      "learning_rate": 0.00048350750383220105,
      "loss": 5.881,
      "mean_token_accuracy": 0.1573862612247467,
      "num_tokens": 30200203.0,
      "step": 14910
    },
    {
      "entropy": 6.120649671554565,
      "epoch": 1.303416936118151,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048349579143233126,
      "loss": 5.8638,
      "mean_token_accuracy": 0.15807482749223709,
      "num_tokens": 30210546.0,
      "step": 14915
    },
    {
      "entropy": 6.216634082794189,
      "epoch": 1.3038538844708556,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004834840750334029,
      "loss": 6.0038,
      "mean_token_accuracy": 0.14518765062093736,
      "num_tokens": 30221316.0,
      "step": 14920
    },
    {
      "entropy": 6.279817199707031,
      "epoch": 1.3042908328235603,
      "grad_norm": 2.375,
      "learning_rate": 0.0004834723546356408,
      "loss": 5.9442,
      "mean_token_accuracy": 0.14440446943044663,
      "num_tokens": 30232580.0,
      "step": 14925
    },
    {
      "entropy": 6.15865535736084,
      "epoch": 1.304727781176265,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048346063023926964,
      "loss": 5.874,
      "mean_token_accuracy": 0.1546608865261078,
      "num_tokens": 30241976.0,
      "step": 14930
    },
    {
      "entropy": 6.127216720581055,
      "epoch": 1.3051647295289697,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048344890184451436,
      "loss": 5.8836,
      "mean_token_accuracy": 0.14819984436035155,
      "num_tokens": 30252723.0,
      "step": 14935
    },
    {
      "entropy": 6.210349750518799,
      "epoch": 1.3056016778816744,
      "grad_norm": 2.546875,
      "learning_rate": 0.00048343716945159997,
      "loss": 5.9174,
      "mean_token_accuracy": 0.15172467082738877,
      "num_tokens": 30263125.0,
      "step": 14940
    },
    {
      "entropy": 6.158228874206543,
      "epoch": 1.306038626234379,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004834254330607514,
      "loss": 5.8315,
      "mean_token_accuracy": 0.16196292042732238,
      "num_tokens": 30273113.0,
      "step": 14945
    },
    {
      "entropy": 6.129610347747803,
      "epoch": 1.3064755745870837,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004834136926721939,
      "loss": 5.8029,
      "mean_token_accuracy": 0.1589888870716095,
      "num_tokens": 30282550.0,
      "step": 14950
    },
    {
      "entropy": 6.12127275466919,
      "epoch": 1.3069125229397884,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048340194828615246,
      "loss": 5.8541,
      "mean_token_accuracy": 0.14687294214963914,
      "num_tokens": 30293377.0,
      "step": 14955
    },
    {
      "entropy": 6.179308366775513,
      "epoch": 1.3073494712924933,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004833901999028525,
      "loss": 5.8712,
      "mean_token_accuracy": 0.15503755509853362,
      "num_tokens": 30303147.0,
      "step": 14960
    },
    {
      "entropy": 6.183325147628784,
      "epoch": 1.3077864196451978,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004833784475225192,
      "loss": 5.9668,
      "mean_token_accuracy": 0.14612608253955842,
      "num_tokens": 30313426.0,
      "step": 14965
    },
    {
      "entropy": 6.19778151512146,
      "epoch": 1.3082233679979027,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004833666911453783,
      "loss": 5.9343,
      "mean_token_accuracy": 0.14557006433606148,
      "num_tokens": 30323896.0,
      "step": 14970
    },
    {
      "entropy": 6.130338335037232,
      "epoch": 1.3086603163506074,
      "grad_norm": 1.90625,
      "learning_rate": 0.000483354930771655,
      "loss": 5.8943,
      "mean_token_accuracy": 0.14434198811650276,
      "num_tokens": 30333523.0,
      "step": 14975
    },
    {
      "entropy": 6.204380941390991,
      "epoch": 1.3090972647033121,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048334316640157493,
      "loss": 5.8877,
      "mean_token_accuracy": 0.15440258681774138,
      "num_tokens": 30344035.0,
      "step": 14980
    },
    {
      "entropy": 6.2058931350708,
      "epoch": 1.3095342130560168,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004833313980353637,
      "loss": 5.9288,
      "mean_token_accuracy": 0.15022298619151114,
      "num_tokens": 30354385.0,
      "step": 14985
    },
    {
      "entropy": 6.104279375076294,
      "epoch": 1.3099711614087215,
      "grad_norm": 2.625,
      "learning_rate": 0.00048331962567324717,
      "loss": 5.9863,
      "mean_token_accuracy": 0.1413296028971672,
      "num_tokens": 30364941.0,
      "step": 14990
    },
    {
      "entropy": 6.1678324222564695,
      "epoch": 1.3104081097614262,
      "grad_norm": 2.40625,
      "learning_rate": 0.000483307849315451,
      "loss": 5.838,
      "mean_token_accuracy": 0.15294638648629189,
      "num_tokens": 30376160.0,
      "step": 14995
    },
    {
      "entropy": 6.274331426620483,
      "epoch": 1.3108450581141309,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048329606896220116,
      "loss": 5.9276,
      "mean_token_accuracy": 0.14750345796346664,
      "num_tokens": 30385018.0,
      "step": 15000
    },
    {
      "epoch": 1.3108450581141309,
      "eval_entropy": 5.9782779657328025,
      "eval_loss": 5.957468032836914,
      "eval_mean_token_accuracy": 0.15653674234963208,
      "eval_num_tokens": 30385018.0,
      "eval_runtime": 28.2657,
      "eval_samples_per_second": 1309.502,
      "eval_steps_per_second": 163.697,
      "step": 15000
    },
    {
      "entropy": 6.148051929473877,
      "epoch": 1.3112820064668356,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004832842846137236,
      "loss": 5.8828,
      "mean_token_accuracy": 0.15537035912275315,
      "num_tokens": 30394618.0,
      "step": 15005
    },
    {
      "entropy": 6.176465845108032,
      "epoch": 1.3117189548195403,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048327249627024433,
      "loss": 5.8699,
      "mean_token_accuracy": 0.15127553790807724,
      "num_tokens": 30404081.0,
      "step": 15010
    },
    {
      "entropy": 6.134244537353515,
      "epoch": 1.3121559031722452,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004832607039319894,
      "loss": 5.879,
      "mean_token_accuracy": 0.14448730796575546,
      "num_tokens": 30413892.0,
      "step": 15015
    },
    {
      "entropy": 6.214825677871704,
      "epoch": 1.3125928515249496,
      "grad_norm": 2.40625,
      "learning_rate": 0.00048324890759918513,
      "loss": 5.9509,
      "mean_token_accuracy": 0.14472972452640534,
      "num_tokens": 30424523.0,
      "step": 15020
    },
    {
      "entropy": 6.194734239578247,
      "epoch": 1.3130297998776546,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004832371072720576,
      "loss": 5.8769,
      "mean_token_accuracy": 0.15406875610351561,
      "num_tokens": 30435092.0,
      "step": 15025
    },
    {
      "entropy": 6.084768152236938,
      "epoch": 1.3134667482303592,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004832253029508332,
      "loss": 5.8419,
      "mean_token_accuracy": 0.15242524743080138,
      "num_tokens": 30445628.0,
      "step": 15030
    },
    {
      "entropy": 6.1286914348602295,
      "epoch": 1.313903696583064,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004832134946357384,
      "loss": 5.9161,
      "mean_token_accuracy": 0.14976181983947753,
      "num_tokens": 30457433.0,
      "step": 15035
    },
    {
      "entropy": 6.206806182861328,
      "epoch": 1.3143406449357686,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004832016823269998,
      "loss": 5.883,
      "mean_token_accuracy": 0.15214013457298278,
      "num_tokens": 30467987.0,
      "step": 15040
    },
    {
      "entropy": 6.178217506408691,
      "epoch": 1.3147775932884733,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004831898660248437,
      "loss": 5.8849,
      "mean_token_accuracy": 0.14360788315534592,
      "num_tokens": 30476577.0,
      "step": 15045
    },
    {
      "entropy": 6.153936910629272,
      "epoch": 1.315214541641178,
      "grad_norm": 1.859375,
      "learning_rate": 0.00048317804572949697,
      "loss": 5.8517,
      "mean_token_accuracy": 0.15068671256303787,
      "num_tokens": 30487410.0,
      "step": 15050
    },
    {
      "entropy": 6.202679681777954,
      "epoch": 1.3156514899938827,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004831662214411862,
      "loss": 5.9379,
      "mean_token_accuracy": 0.1465588890016079,
      "num_tokens": 30497778.0,
      "step": 15055
    },
    {
      "entropy": 6.178036642074585,
      "epoch": 1.3160884383465874,
      "grad_norm": 2.125,
      "learning_rate": 0.0004831543931601382,
      "loss": 5.8754,
      "mean_token_accuracy": 0.15059689432382584,
      "num_tokens": 30507369.0,
      "step": 15060
    },
    {
      "entropy": 6.166036605834961,
      "epoch": 1.316525386699292,
      "grad_norm": 4.65625,
      "learning_rate": 0.00048314256088657985,
      "loss": 5.9122,
      "mean_token_accuracy": 0.14978303164243698,
      "num_tokens": 30518126.0,
      "step": 15065
    },
    {
      "entropy": 6.145087623596192,
      "epoch": 1.3169623350519968,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004831307246207382,
      "loss": 5.8763,
      "mean_token_accuracy": 0.15576560273766518,
      "num_tokens": 30528167.0,
      "step": 15070
    },
    {
      "entropy": 6.1718322277069095,
      "epoch": 1.3173992834047015,
      "grad_norm": 2.75,
      "learning_rate": 0.00048311888436284017,
      "loss": 5.8401,
      "mean_token_accuracy": 0.15360784232616426,
      "num_tokens": 30538111.0,
      "step": 15075
    },
    {
      "entropy": 6.129813718795776,
      "epoch": 1.3178362317574064,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004831070401131129,
      "loss": 5.9456,
      "mean_token_accuracy": 0.1528167486190796,
      "num_tokens": 30547773.0,
      "step": 15080
    },
    {
      "entropy": 6.119461917877198,
      "epoch": 1.3182731801101109,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004830951918717834,
      "loss": 5.8782,
      "mean_token_accuracy": 0.1502302333712578,
      "num_tokens": 30558058.0,
      "step": 15085
    },
    {
      "entropy": 6.133006429672241,
      "epoch": 1.3187101284628158,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048308333963907927,
      "loss": 5.8117,
      "mean_token_accuracy": 0.15927943140268325,
      "num_tokens": 30567297.0,
      "step": 15090
    },
    {
      "entropy": 6.184013843536377,
      "epoch": 1.3191470768155205,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048307148341522756,
      "loss": 5.9317,
      "mean_token_accuracy": 0.15289607793092727,
      "num_tokens": 30578159.0,
      "step": 15095
    },
    {
      "entropy": 6.1799774169921875,
      "epoch": 1.3195840251682251,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048305962320045574,
      "loss": 5.9017,
      "mean_token_accuracy": 0.15140537172555923,
      "num_tokens": 30588734.0,
      "step": 15100
    },
    {
      "entropy": 6.105605602264404,
      "epoch": 1.3200209735209298,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048304775899499143,
      "loss": 5.8031,
      "mean_token_accuracy": 0.16111132502555847,
      "num_tokens": 30598496.0,
      "step": 15105
    },
    {
      "entropy": 6.117049646377564,
      "epoch": 1.3204579218736345,
      "grad_norm": 2.234375,
      "learning_rate": 0.000483035890799062,
      "loss": 5.8634,
      "mean_token_accuracy": 0.15213498026132583,
      "num_tokens": 30608330.0,
      "step": 15110
    },
    {
      "entropy": 6.183338308334351,
      "epoch": 1.3208948702263392,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004830240186128952,
      "loss": 5.8802,
      "mean_token_accuracy": 0.15419178307056428,
      "num_tokens": 30618887.0,
      "step": 15115
    },
    {
      "entropy": 6.22232494354248,
      "epoch": 1.321331818579044,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048301214243671884,
      "loss": 5.8504,
      "mean_token_accuracy": 0.15129734724760055,
      "num_tokens": 30629479.0,
      "step": 15120
    },
    {
      "entropy": 6.147677993774414,
      "epoch": 1.3217687669317486,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004830002622707604,
      "loss": 5.817,
      "mean_token_accuracy": 0.154408822953701,
      "num_tokens": 30639270.0,
      "step": 15125
    },
    {
      "entropy": 6.130454587936401,
      "epoch": 1.3222057152844533,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004829883781152481,
      "loss": 5.9363,
      "mean_token_accuracy": 0.15437458902597428,
      "num_tokens": 30649156.0,
      "step": 15130
    },
    {
      "entropy": 6.149850511550904,
      "epoch": 1.3226426636371582,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048297648997040966,
      "loss": 5.9678,
      "mean_token_accuracy": 0.14519903287291527,
      "num_tokens": 30659919.0,
      "step": 15135
    },
    {
      "entropy": 6.226070213317871,
      "epoch": 1.3230796119898627,
      "grad_norm": 2.625,
      "learning_rate": 0.00048296459783647304,
      "loss": 5.9235,
      "mean_token_accuracy": 0.15368696600198745,
      "num_tokens": 30670523.0,
      "step": 15140
    },
    {
      "entropy": 6.204253149032593,
      "epoch": 1.3235165603425676,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048295270171366665,
      "loss": 5.9256,
      "mean_token_accuracy": 0.15158659368753433,
      "num_tokens": 30680803.0,
      "step": 15145
    },
    {
      "entropy": 6.1911430835723875,
      "epoch": 1.3239535086952723,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048294080160221837,
      "loss": 5.8146,
      "mean_token_accuracy": 0.1503637984395027,
      "num_tokens": 30690401.0,
      "step": 15150
    },
    {
      "entropy": 6.188399791717529,
      "epoch": 1.324390457047977,
      "grad_norm": 2.484375,
      "learning_rate": 0.00048292889750235665,
      "loss": 5.9368,
      "mean_token_accuracy": 0.14866487085819244,
      "num_tokens": 30700779.0,
      "step": 15155
    },
    {
      "entropy": 6.187041854858398,
      "epoch": 1.3248274054006817,
      "grad_norm": 2.828125,
      "learning_rate": 0.00048291698941430955,
      "loss": 5.8849,
      "mean_token_accuracy": 0.14539026319980622,
      "num_tokens": 30711556.0,
      "step": 15160
    },
    {
      "entropy": 6.15223240852356,
      "epoch": 1.3252643537533864,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004829050773383058,
      "loss": 5.7862,
      "mean_token_accuracy": 0.15828098356723785,
      "num_tokens": 30721929.0,
      "step": 15165
    },
    {
      "entropy": 6.176116609573365,
      "epoch": 1.325701302106091,
      "grad_norm": 2.125,
      "learning_rate": 0.0004828931612745737,
      "loss": 5.9348,
      "mean_token_accuracy": 0.1483965814113617,
      "num_tokens": 30732848.0,
      "step": 15170
    },
    {
      "entropy": 6.141019487380982,
      "epoch": 1.3261382504587957,
      "grad_norm": 1.9375,
      "learning_rate": 0.00048288124122334176,
      "loss": 5.8738,
      "mean_token_accuracy": 0.14580026268959045,
      "num_tokens": 30742628.0,
      "step": 15175
    },
    {
      "entropy": 6.191814947128296,
      "epoch": 1.3265751988115004,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004828693171848388,
      "loss": 5.9259,
      "mean_token_accuracy": 0.14872915670275688,
      "num_tokens": 30752866.0,
      "step": 15180
    },
    {
      "entropy": 6.157923841476441,
      "epoch": 1.3270121471642051,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004828573891592933,
      "loss": 5.8502,
      "mean_token_accuracy": 0.1581645578145981,
      "num_tokens": 30762861.0,
      "step": 15185
    },
    {
      "entropy": 6.089473295211792,
      "epoch": 1.32744909551691,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004828454571469342,
      "loss": 5.9181,
      "mean_token_accuracy": 0.15590844303369522,
      "num_tokens": 30772555.0,
      "step": 15190
    },
    {
      "entropy": 6.173651218414307,
      "epoch": 1.3278860438696145,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004828335211479903,
      "loss": 5.8846,
      "mean_token_accuracy": 0.15066574960947038,
      "num_tokens": 30782739.0,
      "step": 15195
    },
    {
      "entropy": 6.259365129470825,
      "epoch": 1.3283229922223194,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048282158116269065,
      "loss": 5.8882,
      "mean_token_accuracy": 0.1502900466322899,
      "num_tokens": 30792600.0,
      "step": 15200
    },
    {
      "entropy": 6.158281898498535,
      "epoch": 1.3287599405750241,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004828096371912641,
      "loss": 5.8832,
      "mean_token_accuracy": 0.144914473593235,
      "num_tokens": 30801433.0,
      "step": 15205
    },
    {
      "entropy": 6.101165246963501,
      "epoch": 1.3291968889277288,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048279768923393987,
      "loss": 5.9775,
      "mean_token_accuracy": 0.14661043509840965,
      "num_tokens": 30811277.0,
      "step": 15210
    },
    {
      "entropy": 6.1981529712677,
      "epoch": 1.3296338372804335,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048278573729094705,
      "loss": 5.9108,
      "mean_token_accuracy": 0.15046690851449968,
      "num_tokens": 30821120.0,
      "step": 15215
    },
    {
      "entropy": 6.171682071685791,
      "epoch": 1.3300707856331382,
      "grad_norm": 2.625,
      "learning_rate": 0.00048277378136251487,
      "loss": 5.8477,
      "mean_token_accuracy": 0.15861936211585997,
      "num_tokens": 30831660.0,
      "step": 15220
    },
    {
      "entropy": 6.217426538467407,
      "epoch": 1.3305077339858429,
      "grad_norm": 2.578125,
      "learning_rate": 0.00048276182144887274,
      "loss": 5.9028,
      "mean_token_accuracy": 0.14908593744039536,
      "num_tokens": 30842530.0,
      "step": 15225
    },
    {
      "entropy": 6.231807518005371,
      "epoch": 1.3309446823385476,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048274985755025003,
      "loss": 5.905,
      "mean_token_accuracy": 0.14666102826595306,
      "num_tokens": 30852954.0,
      "step": 15230
    },
    {
      "entropy": 6.13426661491394,
      "epoch": 1.3313816306912523,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048273788966687615,
      "loss": 5.8191,
      "mean_token_accuracy": 0.153940911591053,
      "num_tokens": 30862864.0,
      "step": 15235
    },
    {
      "entropy": 6.167418193817139,
      "epoch": 1.331818579043957,
      "grad_norm": 4.25,
      "learning_rate": 0.0004827259177989807,
      "loss": 5.8957,
      "mean_token_accuracy": 0.14726108610630034,
      "num_tokens": 30873533.0,
      "step": 15240
    },
    {
      "entropy": 6.175304317474366,
      "epoch": 1.3322555273966616,
      "grad_norm": 2.875,
      "learning_rate": 0.00048271394194679336,
      "loss": 5.8363,
      "mean_token_accuracy": 0.15829232484102249,
      "num_tokens": 30883564.0,
      "step": 15245
    },
    {
      "entropy": 6.18950309753418,
      "epoch": 1.3326924757493663,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004827019621105436,
      "loss": 5.861,
      "mean_token_accuracy": 0.15210068076848984,
      "num_tokens": 30893516.0,
      "step": 15250
    },
    {
      "entropy": 6.149360847473145,
      "epoch": 1.3331294241020712,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048268997829046156,
      "loss": 5.8475,
      "mean_token_accuracy": 0.15454588681459427,
      "num_tokens": 30903104.0,
      "step": 15255
    },
    {
      "entropy": 6.150010442733764,
      "epoch": 1.3335663724547757,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048267799048677684,
      "loss": 5.8517,
      "mean_token_accuracy": 0.15118254125118255,
      "num_tokens": 30912759.0,
      "step": 15260
    },
    {
      "entropy": 6.1635654926300045,
      "epoch": 1.3340033208074806,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004826659986997194,
      "loss": 5.8662,
      "mean_token_accuracy": 0.15174566358327865,
      "num_tokens": 30921361.0,
      "step": 15265
    },
    {
      "entropy": 6.174410438537597,
      "epoch": 1.3344402691601853,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004826540029295193,
      "loss": 5.868,
      "mean_token_accuracy": 0.14787096679210662,
      "num_tokens": 30931006.0,
      "step": 15270
    },
    {
      "entropy": 6.157722854614258,
      "epoch": 1.33487721751289,
      "grad_norm": 6.25,
      "learning_rate": 0.00048264200317640665,
      "loss": 5.871,
      "mean_token_accuracy": 0.1552622899413109,
      "num_tokens": 30940647.0,
      "step": 15275
    },
    {
      "entropy": 6.197625541687012,
      "epoch": 1.3353141658655947,
      "grad_norm": 2.625,
      "learning_rate": 0.0004826299994406116,
      "loss": 5.9313,
      "mean_token_accuracy": 0.14712630659341813,
      "num_tokens": 30951028.0,
      "step": 15280
    },
    {
      "entropy": 6.146320819854736,
      "epoch": 1.3357511142182994,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048261799172236435,
      "loss": 5.8707,
      "mean_token_accuracy": 0.15156200900673866,
      "num_tokens": 30960726.0,
      "step": 15285
    },
    {
      "entropy": 6.188878679275513,
      "epoch": 1.336188062571004,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004826059800218952,
      "loss": 5.9599,
      "mean_token_accuracy": 0.14741793274879456,
      "num_tokens": 30971061.0,
      "step": 15290
    },
    {
      "entropy": 6.15547456741333,
      "epoch": 1.3366250109237088,
      "grad_norm": 3.25,
      "learning_rate": 0.0004825939643394346,
      "loss": 5.7647,
      "mean_token_accuracy": 0.16355321407318116,
      "num_tokens": 30980518.0,
      "step": 15295
    },
    {
      "entropy": 6.189762926101684,
      "epoch": 1.3370619592764135,
      "grad_norm": 4.59375,
      "learning_rate": 0.00048258194467521294,
      "loss": 5.9525,
      "mean_token_accuracy": 0.14899965077638627,
      "num_tokens": 30990759.0,
      "step": 15300
    },
    {
      "entropy": 6.232774972915649,
      "epoch": 1.3374989076291182,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048256992102946084,
      "loss": 5.9453,
      "mean_token_accuracy": 0.14278285801410676,
      "num_tokens": 31002218.0,
      "step": 15305
    },
    {
      "entropy": 6.13329291343689,
      "epoch": 1.337935855981823,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004825578934024089,
      "loss": 5.8844,
      "mean_token_accuracy": 0.1570137172937393,
      "num_tokens": 31012434.0,
      "step": 15310
    },
    {
      "entropy": 6.19418683052063,
      "epoch": 1.3383728043345275,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048254586179428776,
      "loss": 5.9642,
      "mean_token_accuracy": 0.14452686607837678,
      "num_tokens": 31022805.0,
      "step": 15315
    },
    {
      "entropy": 6.179207801818848,
      "epoch": 1.3388097526872325,
      "grad_norm": 3.0,
      "learning_rate": 0.0004825338262053283,
      "loss": 5.862,
      "mean_token_accuracy": 0.15455538034439087,
      "num_tokens": 31032563.0,
      "step": 15320
    },
    {
      "entropy": 6.09782919883728,
      "epoch": 1.3392467010399371,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004825217866357613,
      "loss": 5.8146,
      "mean_token_accuracy": 0.1602785497903824,
      "num_tokens": 31041445.0,
      "step": 15325
    },
    {
      "entropy": 6.1660455703735355,
      "epoch": 1.3396836493926418,
      "grad_norm": 3.078125,
      "learning_rate": 0.00048250974308581765,
      "loss": 5.8604,
      "mean_token_accuracy": 0.15620155781507492,
      "num_tokens": 31051413.0,
      "step": 15330
    },
    {
      "entropy": 6.218658828735352,
      "epoch": 1.3401205977453465,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048249769555572844,
      "loss": 5.8446,
      "mean_token_accuracy": 0.15629526674747468,
      "num_tokens": 31061822.0,
      "step": 15335
    },
    {
      "entropy": 6.133849668502807,
      "epoch": 1.3405575460980512,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048248564404572463,
      "loss": 5.8672,
      "mean_token_accuracy": 0.15279928594827652,
      "num_tokens": 31071889.0,
      "step": 15340
    },
    {
      "entropy": 6.089661073684693,
      "epoch": 1.340994494450756,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048247358855603754,
      "loss": 5.8109,
      "mean_token_accuracy": 0.16055045872926713,
      "num_tokens": 31080933.0,
      "step": 15345
    },
    {
      "entropy": 6.156580591201783,
      "epoch": 1.3414314428034606,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004824615290868982,
      "loss": 5.9418,
      "mean_token_accuracy": 0.14569876790046693,
      "num_tokens": 31091349.0,
      "step": 15350
    },
    {
      "entropy": 6.312840843200684,
      "epoch": 1.3418683911561653,
      "grad_norm": 2.125,
      "learning_rate": 0.0004824494656385381,
      "loss": 5.9142,
      "mean_token_accuracy": 0.1483866587281227,
      "num_tokens": 31101485.0,
      "step": 15355
    },
    {
      "entropy": 6.1742030620574955,
      "epoch": 1.34230533950887,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004824373982111886,
      "loss": 5.8678,
      "mean_token_accuracy": 0.1494912676513195,
      "num_tokens": 31111965.0,
      "step": 15360
    },
    {
      "entropy": 6.140328216552734,
      "epoch": 1.3427422878615747,
      "grad_norm": 3.375,
      "learning_rate": 0.000482425326805081,
      "loss": 5.853,
      "mean_token_accuracy": 0.15451739430427552,
      "num_tokens": 31122284.0,
      "step": 15365
    },
    {
      "entropy": 6.207666969299316,
      "epoch": 1.3431792362142794,
      "grad_norm": 4.59375,
      "learning_rate": 0.00048241325142044697,
      "loss": 5.8889,
      "mean_token_accuracy": 0.1462631568312645,
      "num_tokens": 31132156.0,
      "step": 15370
    },
    {
      "entropy": 6.172466468811035,
      "epoch": 1.3436161845669843,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048240117205751817,
      "loss": 5.812,
      "mean_token_accuracy": 0.15407064855098723,
      "num_tokens": 31141609.0,
      "step": 15375
    },
    {
      "entropy": 6.060343790054321,
      "epoch": 1.344053132919689,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004823890887165261,
      "loss": 5.9125,
      "mean_token_accuracy": 0.14953329041600227,
      "num_tokens": 31153159.0,
      "step": 15380
    },
    {
      "entropy": 6.133303213119507,
      "epoch": 1.3444900812723937,
      "grad_norm": 2.984375,
      "learning_rate": 0.00048237700139770273,
      "loss": 5.9338,
      "mean_token_accuracy": 0.1543626755475998,
      "num_tokens": 31163330.0,
      "step": 15385
    },
    {
      "entropy": 6.2770579814910885,
      "epoch": 1.3449270296250984,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004823649101012797,
      "loss": 5.9351,
      "mean_token_accuracy": 0.15404071658849716,
      "num_tokens": 31173874.0,
      "step": 15390
    },
    {
      "entropy": 6.158340740203857,
      "epoch": 1.345363977977803,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004823528148274891,
      "loss": 5.824,
      "mean_token_accuracy": 0.15880247950553894,
      "num_tokens": 31183933.0,
      "step": 15395
    },
    {
      "entropy": 6.081319904327392,
      "epoch": 1.3458009263305077,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048234071557656295,
      "loss": 5.8066,
      "mean_token_accuracy": 0.16156378239393235,
      "num_tokens": 31194308.0,
      "step": 15400
    },
    {
      "entropy": 6.11509256362915,
      "epoch": 1.3462378746832124,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004823286123487331,
      "loss": 5.8618,
      "mean_token_accuracy": 0.15184128731489183,
      "num_tokens": 31203757.0,
      "step": 15405
    },
    {
      "entropy": 6.218196010589599,
      "epoch": 1.3466748230359171,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048231650514423194,
      "loss": 5.945,
      "mean_token_accuracy": 0.14972004294395447,
      "num_tokens": 31213980.0,
      "step": 15410
    },
    {
      "entropy": 6.261203813552856,
      "epoch": 1.3471117713886218,
      "grad_norm": 3.25,
      "learning_rate": 0.00048230439396329143,
      "loss": 5.9059,
      "mean_token_accuracy": 0.14936172515153884,
      "num_tokens": 31225034.0,
      "step": 15415
    },
    {
      "entropy": 6.133521842956543,
      "epoch": 1.3475487197413265,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048229227880614413,
      "loss": 5.8927,
      "mean_token_accuracy": 0.14812134951353073,
      "num_tokens": 31236393.0,
      "step": 15420
    },
    {
      "entropy": 6.150119924545288,
      "epoch": 1.3479856680940312,
      "grad_norm": 2.625,
      "learning_rate": 0.0004822801596730222,
      "loss": 5.8827,
      "mean_token_accuracy": 0.1485971063375473,
      "num_tokens": 31246577.0,
      "step": 15425
    },
    {
      "entropy": 6.1682209968566895,
      "epoch": 1.348422616446736,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048226803656415833,
      "loss": 5.8241,
      "mean_token_accuracy": 0.15531778931617737,
      "num_tokens": 31256004.0,
      "step": 15430
    },
    {
      "entropy": 6.202504634857178,
      "epoch": 1.3488595647994406,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004822559094797849,
      "loss": 5.9346,
      "mean_token_accuracy": 0.14537073522806168,
      "num_tokens": 31267294.0,
      "step": 15435
    },
    {
      "entropy": 6.152498722076416,
      "epoch": 1.3492965131521455,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048224377842013443,
      "loss": 5.8217,
      "mean_token_accuracy": 0.16052476167678834,
      "num_tokens": 31276896.0,
      "step": 15440
    },
    {
      "entropy": 6.140977430343628,
      "epoch": 1.3497334615048502,
      "grad_norm": 2.875,
      "learning_rate": 0.0004822316433854397,
      "loss": 5.9065,
      "mean_token_accuracy": 0.15101585239171983,
      "num_tokens": 31287248.0,
      "step": 15445
    },
    {
      "entropy": 6.11960883140564,
      "epoch": 1.3501704098575549,
      "grad_norm": 5.78125,
      "learning_rate": 0.00048221950437593353,
      "loss": 5.8882,
      "mean_token_accuracy": 0.15626996457576753,
      "num_tokens": 31297422.0,
      "step": 15450
    },
    {
      "entropy": 6.153040027618408,
      "epoch": 1.3506073582102596,
      "grad_norm": 1.8125,
      "learning_rate": 0.00048220736139184864,
      "loss": 5.9142,
      "mean_token_accuracy": 0.15504897236824036,
      "num_tokens": 31307182.0,
      "step": 15455
    },
    {
      "entropy": 6.206346225738526,
      "epoch": 1.3510443065629643,
      "grad_norm": 2.390625,
      "learning_rate": 0.000482195214433418,
      "loss": 5.9918,
      "mean_token_accuracy": 0.1427098862826824,
      "num_tokens": 31317234.0,
      "step": 15460
    },
    {
      "entropy": 6.22294397354126,
      "epoch": 1.351481254915669,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004821830635008746,
      "loss": 5.9764,
      "mean_token_accuracy": 0.14484017714858055,
      "num_tokens": 31328294.0,
      "step": 15465
    },
    {
      "entropy": 6.2555149555206295,
      "epoch": 1.3519182032683736,
      "grad_norm": 2.0,
      "learning_rate": 0.0004821709085944515,
      "loss": 5.854,
      "mean_token_accuracy": 0.15396426022052764,
      "num_tokens": 31337250.0,
      "step": 15470
    },
    {
      "entropy": 6.183167791366577,
      "epoch": 1.3523551516210783,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004821587497143817,
      "loss": 5.9305,
      "mean_token_accuracy": 0.15111508667469026,
      "num_tokens": 31346900.0,
      "step": 15475
    },
    {
      "entropy": 6.159300804138184,
      "epoch": 1.352792099973783,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004821465868608986,
      "loss": 5.9023,
      "mean_token_accuracy": 0.1471735455095768,
      "num_tokens": 31357233.0,
      "step": 15480
    },
    {
      "entropy": 6.18670687675476,
      "epoch": 1.353229048326488,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004821344200342355,
      "loss": 5.8153,
      "mean_token_accuracy": 0.1530952975153923,
      "num_tokens": 31367043.0,
      "step": 15485
    },
    {
      "entropy": 6.1640722274780275,
      "epoch": 1.3536659966791924,
      "grad_norm": 2.625,
      "learning_rate": 0.0004821222492346256,
      "loss": 5.8513,
      "mean_token_accuracy": 0.14524135291576384,
      "num_tokens": 31376770.0,
      "step": 15490
    },
    {
      "entropy": 6.202436351776123,
      "epoch": 1.3541029450318973,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048211007446230243,
      "loss": 5.8779,
      "mean_token_accuracy": 0.15482534021139144,
      "num_tokens": 31386056.0,
      "step": 15495
    },
    {
      "entropy": 6.1484537601470945,
      "epoch": 1.354539893384602,
      "grad_norm": 2.859375,
      "learning_rate": 0.00048209789571749954,
      "loss": 5.9419,
      "mean_token_accuracy": 0.14602854549884797,
      "num_tokens": 31395129.0,
      "step": 15500
    },
    {
      "entropy": 6.220837545394898,
      "epoch": 1.3549768417373067,
      "grad_norm": 2.5,
      "learning_rate": 0.00048208571300045045,
      "loss": 5.9448,
      "mean_token_accuracy": 0.14848102629184723,
      "num_tokens": 31404720.0,
      "step": 15505
    },
    {
      "entropy": 6.230710124969482,
      "epoch": 1.3554137900900114,
      "grad_norm": 3.09375,
      "learning_rate": 0.00048207352631138897,
      "loss": 5.8808,
      "mean_token_accuracy": 0.15658908486366271,
      "num_tokens": 31415588.0,
      "step": 15510
    },
    {
      "entropy": 6.180061388015747,
      "epoch": 1.355850738442716,
      "grad_norm": 3.140625,
      "learning_rate": 0.00048206133565054863,
      "loss": 5.9698,
      "mean_token_accuracy": 0.14740943163633347,
      "num_tokens": 31425904.0,
      "step": 15515
    },
    {
      "entropy": 6.158311080932617,
      "epoch": 1.3562876867954208,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004820491410181634,
      "loss": 5.9264,
      "mean_token_accuracy": 0.14886042326688767,
      "num_tokens": 31435282.0,
      "step": 15520
    },
    {
      "entropy": 6.227653503417969,
      "epoch": 1.3567246351481255,
      "grad_norm": 2.578125,
      "learning_rate": 0.00048203694241446724,
      "loss": 5.9971,
      "mean_token_accuracy": 0.14620602056384085,
      "num_tokens": 31445681.0,
      "step": 15525
    },
    {
      "entropy": 6.146025085449219,
      "epoch": 1.3571615835008302,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048202473983969396,
      "loss": 5.7007,
      "mean_token_accuracy": 0.15896469801664354,
      "num_tokens": 31456934.0,
      "step": 15530
    },
    {
      "entropy": 6.076095151901245,
      "epoch": 1.3575985318535349,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048201253329407766,
      "loss": 5.8692,
      "mean_token_accuracy": 0.1567166641354561,
      "num_tokens": 31466692.0,
      "step": 15535
    },
    {
      "entropy": 6.108785200119018,
      "epoch": 1.3580354802062395,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004820003227778526,
      "loss": 5.8906,
      "mean_token_accuracy": 0.14972759932279586,
      "num_tokens": 31477166.0,
      "step": 15540
    },
    {
      "entropy": 6.241550159454346,
      "epoch": 1.3584724285589442,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004819881082912528,
      "loss": 5.9007,
      "mean_token_accuracy": 0.14770390912890435,
      "num_tokens": 31486998.0,
      "step": 15545
    },
    {
      "entropy": 6.217063617706299,
      "epoch": 1.3589093769116491,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048197588983451265,
      "loss": 5.9147,
      "mean_token_accuracy": 0.1456634059548378,
      "num_tokens": 31496674.0,
      "step": 15550
    },
    {
      "entropy": 6.151744365692139,
      "epoch": 1.3593463252643536,
      "grad_norm": 4.4375,
      "learning_rate": 0.00048196366740786646,
      "loss": 5.9016,
      "mean_token_accuracy": 0.1474331870675087,
      "num_tokens": 31507412.0,
      "step": 15555
    },
    {
      "entropy": 6.1885085105896,
      "epoch": 1.3597832736170585,
      "grad_norm": 2.375,
      "learning_rate": 0.00048195144101154867,
      "loss": 5.8738,
      "mean_token_accuracy": 0.14947630912065507,
      "num_tokens": 31517048.0,
      "step": 15560
    },
    {
      "entropy": 6.142838001251221,
      "epoch": 1.3602202219697632,
      "grad_norm": 3.25,
      "learning_rate": 0.00048193921064579385,
      "loss": 5.9656,
      "mean_token_accuracy": 0.14596446081995965,
      "num_tokens": 31526353.0,
      "step": 15565
    },
    {
      "entropy": 6.214523601531982,
      "epoch": 1.360657170322468,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004819269763108365,
      "loss": 5.9382,
      "mean_token_accuracy": 0.14883836954832078,
      "num_tokens": 31537194.0,
      "step": 15570
    },
    {
      "entropy": 6.268964099884033,
      "epoch": 1.3610941186751726,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048191473800691145,
      "loss": 5.876,
      "mean_token_accuracy": 0.150735567510128,
      "num_tokens": 31547896.0,
      "step": 15575
    },
    {
      "entropy": 6.131948471069336,
      "epoch": 1.3615310670278773,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004819024957342531,
      "loss": 5.8904,
      "mean_token_accuracy": 0.14850337356328963,
      "num_tokens": 31558750.0,
      "step": 15580
    },
    {
      "entropy": 6.130724382400513,
      "epoch": 1.361968015380582,
      "grad_norm": 1.921875,
      "learning_rate": 0.00048189024949309655,
      "loss": 5.8586,
      "mean_token_accuracy": 0.15690528899431228,
      "num_tokens": 31568721.0,
      "step": 15585
    },
    {
      "entropy": 6.1273688793182375,
      "epoch": 1.3624049637332867,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004818779992836766,
      "loss": 5.8465,
      "mean_token_accuracy": 0.1506320394575596,
      "num_tokens": 31579907.0,
      "step": 15590
    },
    {
      "entropy": 6.243776464462281,
      "epoch": 1.3628419120859914,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048186574510622823,
      "loss": 5.9077,
      "mean_token_accuracy": 0.15120431780815125,
      "num_tokens": 31590137.0,
      "step": 15595
    },
    {
      "entropy": 6.202559566497802,
      "epoch": 1.363278860438696,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048185348696098653,
      "loss": 5.9645,
      "mean_token_accuracy": 0.14439767971634865,
      "num_tokens": 31599969.0,
      "step": 15600
    },
    {
      "entropy": 6.144268560409546,
      "epoch": 1.363715808791401,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004818412248481865,
      "loss": 5.9,
      "mean_token_accuracy": 0.1490763917565346,
      "num_tokens": 31609951.0,
      "step": 15605
    },
    {
      "entropy": 6.142098474502563,
      "epoch": 1.3641527571441054,
      "grad_norm": 3.78125,
      "learning_rate": 0.00048182895876806343,
      "loss": 5.8319,
      "mean_token_accuracy": 0.1605251982808113,
      "num_tokens": 31619214.0,
      "step": 15610
    },
    {
      "entropy": 6.203738832473755,
      "epoch": 1.3645897054968104,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004818166887208524,
      "loss": 5.8602,
      "mean_token_accuracy": 0.15474345684051513,
      "num_tokens": 31629562.0,
      "step": 15615
    },
    {
      "entropy": 6.123479223251342,
      "epoch": 1.365026653849515,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004818044147067891,
      "loss": 5.8561,
      "mean_token_accuracy": 0.1578165262937546,
      "num_tokens": 31639801.0,
      "step": 15620
    },
    {
      "entropy": 6.197249507904052,
      "epoch": 1.3654636022022197,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048179213672610866,
      "loss": 5.9213,
      "mean_token_accuracy": 0.14585366025567054,
      "num_tokens": 31650593.0,
      "step": 15625
    },
    {
      "entropy": 6.189125061035156,
      "epoch": 1.3659005505549244,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048177985477904675,
      "loss": 5.8872,
      "mean_token_accuracy": 0.1538390077650547,
      "num_tokens": 31660942.0,
      "step": 15630
    },
    {
      "entropy": 6.2054060935974125,
      "epoch": 1.3663374989076291,
      "grad_norm": 1.96875,
      "learning_rate": 0.00048176756886583884,
      "loss": 5.9436,
      "mean_token_accuracy": 0.1442718930542469,
      "num_tokens": 31671105.0,
      "step": 15635
    },
    {
      "entropy": 6.199615240097046,
      "epoch": 1.3667744472603338,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004817552789867206,
      "loss": 5.8721,
      "mean_token_accuracy": 0.1513895183801651,
      "num_tokens": 31680696.0,
      "step": 15640
    },
    {
      "entropy": 6.077017545700073,
      "epoch": 1.3672113956130385,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048174298514192784,
      "loss": 5.8006,
      "mean_token_accuracy": 0.15794742107391357,
      "num_tokens": 31689755.0,
      "step": 15645
    },
    {
      "entropy": 6.0574463367462155,
      "epoch": 1.3676483439657432,
      "grad_norm": 3.515625,
      "learning_rate": 0.00048173068733169615,
      "loss": 5.8491,
      "mean_token_accuracy": 0.15717778652906417,
      "num_tokens": 31699614.0,
      "step": 15650
    },
    {
      "entropy": 6.192130374908447,
      "epoch": 1.3680852923184479,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004817183855562617,
      "loss": 5.8485,
      "mean_token_accuracy": 0.15516546070575715,
      "num_tokens": 31709106.0,
      "step": 15655
    },
    {
      "entropy": 6.214150381088257,
      "epoch": 1.3685222406711528,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048170607981586023,
      "loss": 5.9604,
      "mean_token_accuracy": 0.1518714964389801,
      "num_tokens": 31719542.0,
      "step": 15660
    },
    {
      "entropy": 6.217283868789673,
      "epoch": 1.3689591890238573,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048169377011072784,
      "loss": 5.887,
      "mean_token_accuracy": 0.15334418565034866,
      "num_tokens": 31730477.0,
      "step": 15665
    },
    {
      "entropy": 6.132981204986573,
      "epoch": 1.3693961373765622,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004816814564411006,
      "loss": 5.8786,
      "mean_token_accuracy": 0.15643389523029327,
      "num_tokens": 31740269.0,
      "step": 15670
    },
    {
      "entropy": 6.173278713226319,
      "epoch": 1.3698330857292669,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004816691388072148,
      "loss": 5.8596,
      "mean_token_accuracy": 0.1569000393152237,
      "num_tokens": 31750153.0,
      "step": 15675
    },
    {
      "entropy": 6.2268585681915285,
      "epoch": 1.3702700340819716,
      "grad_norm": 1.90625,
      "learning_rate": 0.00048165681720930653,
      "loss": 5.9525,
      "mean_token_accuracy": 0.14221562445163727,
      "num_tokens": 31760171.0,
      "step": 15680
    },
    {
      "entropy": 6.195370864868164,
      "epoch": 1.3707069824346763,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004816444916476123,
      "loss": 5.9355,
      "mean_token_accuracy": 0.14723035991191863,
      "num_tokens": 31771601.0,
      "step": 15685
    },
    {
      "entropy": 6.2029111862182615,
      "epoch": 1.371143930787381,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048163216212236844,
      "loss": 5.875,
      "mean_token_accuracy": 0.15174343883991243,
      "num_tokens": 31781423.0,
      "step": 15690
    },
    {
      "entropy": 6.125151777267456,
      "epoch": 1.3715808791400856,
      "grad_norm": 2.75,
      "learning_rate": 0.0004816198286338115,
      "loss": 5.7959,
      "mean_token_accuracy": 0.15884669944643975,
      "num_tokens": 31791025.0,
      "step": 15695
    },
    {
      "entropy": 6.069617414474488,
      "epoch": 1.3720178274927903,
      "grad_norm": 2.875,
      "learning_rate": 0.00048160749118217783,
      "loss": 5.8965,
      "mean_token_accuracy": 0.15135340988636017,
      "num_tokens": 31801125.0,
      "step": 15700
    },
    {
      "entropy": 6.214752244949341,
      "epoch": 1.372454775845495,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004815951497677043,
      "loss": 5.889,
      "mean_token_accuracy": 0.14464661851525307,
      "num_tokens": 31811100.0,
      "step": 15705
    },
    {
      "entropy": 6.267079305648804,
      "epoch": 1.3728917241981997,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004815828043906275,
      "loss": 5.9266,
      "mean_token_accuracy": 0.14064408764243125,
      "num_tokens": 31821667.0,
      "step": 15710
    },
    {
      "entropy": 6.178414678573608,
      "epoch": 1.3733286725509044,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004815704550511843,
      "loss": 5.8316,
      "mean_token_accuracy": 0.15786113291978837,
      "num_tokens": 31831859.0,
      "step": 15715
    },
    {
      "entropy": 6.1278722286224365,
      "epoch": 1.373765620903609,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004815581017496115,
      "loss": 5.8342,
      "mean_token_accuracy": 0.15364986658096313,
      "num_tokens": 31842952.0,
      "step": 15720
    },
    {
      "entropy": 6.0776642799377445,
      "epoch": 1.374202569256314,
      "grad_norm": 2.125,
      "learning_rate": 0.00048154574448614607,
      "loss": 5.9285,
      "mean_token_accuracy": 0.15256015807390214,
      "num_tokens": 31852869.0,
      "step": 15725
    },
    {
      "entropy": 6.202900838851929,
      "epoch": 1.3746395176090185,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004815333832610251,
      "loss": 5.8515,
      "mean_token_accuracy": 0.157974199205637,
      "num_tokens": 31862590.0,
      "step": 15730
    },
    {
      "entropy": 6.1410387516021725,
      "epoch": 1.3750764659617234,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048152101807448554,
      "loss": 5.8943,
      "mean_token_accuracy": 0.14932089447975158,
      "num_tokens": 31873071.0,
      "step": 15735
    },
    {
      "entropy": 6.168727540969849,
      "epoch": 1.375513414314428,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004815086489267647,
      "loss": 5.8891,
      "mean_token_accuracy": 0.15680812895298005,
      "num_tokens": 31883325.0,
      "step": 15740
    },
    {
      "entropy": 6.200935983657837,
      "epoch": 1.3759503626671328,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048149627581809974,
      "loss": 5.8847,
      "mean_token_accuracy": 0.15281272530555726,
      "num_tokens": 31893071.0,
      "step": 15745
    },
    {
      "entropy": 6.141010189056397,
      "epoch": 1.3763873110198375,
      "grad_norm": 2.984375,
      "learning_rate": 0.000481483898748728,
      "loss": 5.8618,
      "mean_token_accuracy": 0.15925009846687316,
      "num_tokens": 31902948.0,
      "step": 15750
    },
    {
      "entropy": 6.140580844879151,
      "epoch": 1.3768242593725422,
      "grad_norm": 3.75,
      "learning_rate": 0.0004814715177188869,
      "loss": 5.8714,
      "mean_token_accuracy": 0.15180886685848236,
      "num_tokens": 31913425.0,
      "step": 15755
    },
    {
      "entropy": 6.158436346054077,
      "epoch": 1.3772612077252468,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004814591327288139,
      "loss": 5.8722,
      "mean_token_accuracy": 0.1527306467294693,
      "num_tokens": 31923608.0,
      "step": 15760
    },
    {
      "entropy": 6.113433170318603,
      "epoch": 1.3776981560779515,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048144674377874656,
      "loss": 5.8151,
      "mean_token_accuracy": 0.15847359746694564,
      "num_tokens": 31932863.0,
      "step": 15765
    },
    {
      "entropy": 6.157820081710815,
      "epoch": 1.3781351044306562,
      "grad_norm": 2.40625,
      "learning_rate": 0.00048143435086892244,
      "loss": 5.8862,
      "mean_token_accuracy": 0.14980321377515793,
      "num_tokens": 31942425.0,
      "step": 15770
    },
    {
      "entropy": 6.172450399398803,
      "epoch": 1.378572052783361,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004814219539995793,
      "loss": 5.888,
      "mean_token_accuracy": 0.15043785572052001,
      "num_tokens": 31952152.0,
      "step": 15775
    },
    {
      "entropy": 6.202452230453491,
      "epoch": 1.3790090011360658,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048140955317095503,
      "loss": 5.917,
      "mean_token_accuracy": 0.14953644722700118,
      "num_tokens": 31962224.0,
      "step": 15780
    },
    {
      "entropy": 6.17941370010376,
      "epoch": 1.3794459494887703,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004813971483832873,
      "loss": 5.8366,
      "mean_token_accuracy": 0.16343390345573425,
      "num_tokens": 31971517.0,
      "step": 15785
    },
    {
      "entropy": 6.140935802459717,
      "epoch": 1.3798828978414752,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004813847396368141,
      "loss": 5.8988,
      "mean_token_accuracy": 0.1522977203130722,
      "num_tokens": 31982461.0,
      "step": 15790
    },
    {
      "entropy": 6.127985000610352,
      "epoch": 1.38031984619418,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004813723269317734,
      "loss": 5.8314,
      "mean_token_accuracy": 0.15892155915498735,
      "num_tokens": 31992202.0,
      "step": 15795
    },
    {
      "entropy": 6.197576189041138,
      "epoch": 1.3807567945468846,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004813599102684034,
      "loss": 5.8771,
      "mean_token_accuracy": 0.15693169012665747,
      "num_tokens": 32002283.0,
      "step": 15800
    },
    {
      "entropy": 6.225122261047363,
      "epoch": 1.3811937428995893,
      "grad_norm": 3.953125,
      "learning_rate": 0.00048134748964694227,
      "loss": 5.876,
      "mean_token_accuracy": 0.15252324640750886,
      "num_tokens": 32013135.0,
      "step": 15805
    },
    {
      "entropy": 6.193136405944824,
      "epoch": 1.381630691252294,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004813350650676281,
      "loss": 5.8541,
      "mean_token_accuracy": 0.15140448957681657,
      "num_tokens": 32022631.0,
      "step": 15810
    },
    {
      "entropy": 6.133950710296631,
      "epoch": 1.3820676396049987,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004813226365306992,
      "loss": 5.9092,
      "mean_token_accuracy": 0.14851686209440232,
      "num_tokens": 32032216.0,
      "step": 15815
    },
    {
      "entropy": 6.141397094726562,
      "epoch": 1.3825045879577034,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048131020403639405,
      "loss": 5.8846,
      "mean_token_accuracy": 0.16149332374334335,
      "num_tokens": 32042382.0,
      "step": 15820
    },
    {
      "entropy": 6.180974531173706,
      "epoch": 1.382941536310408,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048129776758495116,
      "loss": 5.8646,
      "mean_token_accuracy": 0.15454696267843246,
      "num_tokens": 32052592.0,
      "step": 15825
    },
    {
      "entropy": 6.103881072998047,
      "epoch": 1.3833784846631128,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004812853271766089,
      "loss": 5.7835,
      "mean_token_accuracy": 0.15395764410495758,
      "num_tokens": 32062260.0,
      "step": 15830
    },
    {
      "entropy": 6.156110858917236,
      "epoch": 1.3838154330158177,
      "grad_norm": 3.15625,
      "learning_rate": 0.000481272882811606,
      "loss": 5.8878,
      "mean_token_accuracy": 0.15162370949983597,
      "num_tokens": 32072079.0,
      "step": 15835
    },
    {
      "entropy": 6.158740663528443,
      "epoch": 1.3842523813685221,
      "grad_norm": 3.703125,
      "learning_rate": 0.00048126043449018115,
      "loss": 5.8899,
      "mean_token_accuracy": 0.15353164523839952,
      "num_tokens": 32082151.0,
      "step": 15840
    },
    {
      "entropy": 6.227280569076538,
      "epoch": 1.384689329721227,
      "grad_norm": 2.140625,
      "learning_rate": 0.000481247982212573,
      "loss": 5.9265,
      "mean_token_accuracy": 0.1456128641963005,
      "num_tokens": 32091893.0,
      "step": 15845
    },
    {
      "entropy": 6.185877513885498,
      "epoch": 1.3851262780739317,
      "grad_norm": 5.15625,
      "learning_rate": 0.00048123552597902053,
      "loss": 5.832,
      "mean_token_accuracy": 0.15836296677589418,
      "num_tokens": 32101550.0,
      "step": 15850
    },
    {
      "entropy": 6.039649963378906,
      "epoch": 1.3855632264266364,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048122306578976266,
      "loss": 5.7481,
      "mean_token_accuracy": 0.16527795940637588,
      "num_tokens": 32112278.0,
      "step": 15855
    },
    {
      "entropy": 6.148902320861817,
      "epoch": 1.3860001747793411,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048121060164503835,
      "loss": 5.9183,
      "mean_token_accuracy": 0.15111266374588012,
      "num_tokens": 32122284.0,
      "step": 15860
    },
    {
      "entropy": 6.235367679595948,
      "epoch": 1.3864371231320458,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004811981335450865,
      "loss": 5.9132,
      "mean_token_accuracy": 0.1495277166366577,
      "num_tokens": 32131552.0,
      "step": 15865
    },
    {
      "entropy": 6.262882804870605,
      "epoch": 1.3868740714847505,
      "grad_norm": 2.78125,
      "learning_rate": 0.00048118566149014653,
      "loss": 5.9636,
      "mean_token_accuracy": 0.14120246171951295,
      "num_tokens": 32141758.0,
      "step": 15870
    },
    {
      "entropy": 6.135331535339356,
      "epoch": 1.3873110198374552,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004811731854804575,
      "loss": 5.8394,
      "mean_token_accuracy": 0.1489864706993103,
      "num_tokens": 32150679.0,
      "step": 15875
    },
    {
      "entropy": 6.072045612335205,
      "epoch": 1.3877479681901599,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004811607055162587,
      "loss": 5.8252,
      "mean_token_accuracy": 0.1568121463060379,
      "num_tokens": 32160967.0,
      "step": 15880
    },
    {
      "entropy": 6.209223699569702,
      "epoch": 1.3881849165428646,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004811482215977895,
      "loss": 5.8908,
      "mean_token_accuracy": 0.1478721097111702,
      "num_tokens": 32170516.0,
      "step": 15885
    },
    {
      "entropy": 6.191197443008423,
      "epoch": 1.3886218648955693,
      "grad_norm": 2.25,
      "learning_rate": 0.00048113573372528956,
      "loss": 5.9,
      "mean_token_accuracy": 0.14994119107723236,
      "num_tokens": 32180741.0,
      "step": 15890
    },
    {
      "entropy": 6.231997156143189,
      "epoch": 1.389058813248274,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004811232418989981,
      "loss": 5.9283,
      "mean_token_accuracy": 0.15348664447665214,
      "num_tokens": 32190646.0,
      "step": 15895
    },
    {
      "entropy": 6.1127461910247805,
      "epoch": 1.3894957616009789,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048111074611915483,
      "loss": 5.861,
      "mean_token_accuracy": 0.15147869288921356,
      "num_tokens": 32201361.0,
      "step": 15900
    },
    {
      "entropy": 6.161264371871948,
      "epoch": 1.3899327099536833,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004810982463859996,
      "loss": 5.9823,
      "mean_token_accuracy": 0.13942893147468566,
      "num_tokens": 32211759.0,
      "step": 15905
    },
    {
      "entropy": 6.228479719161987,
      "epoch": 1.3903696583063883,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004810857426997719,
      "loss": 5.9377,
      "mean_token_accuracy": 0.14883668646216391,
      "num_tokens": 32222582.0,
      "step": 15910
    },
    {
      "entropy": 6.170031929016114,
      "epoch": 1.390806606659093,
      "grad_norm": 1.75,
      "learning_rate": 0.00048107323506071167,
      "loss": 5.89,
      "mean_token_accuracy": 0.15693831741809844,
      "num_tokens": 32232936.0,
      "step": 15915
    },
    {
      "entropy": 6.172857189178467,
      "epoch": 1.3912435550117976,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048106072346905875,
      "loss": 5.9068,
      "mean_token_accuracy": 0.15017607063055038,
      "num_tokens": 32242392.0,
      "step": 15920
    },
    {
      "entropy": 6.227355003356934,
      "epoch": 1.3916805033645023,
      "grad_norm": 2.734375,
      "learning_rate": 0.00048104820792505335,
      "loss": 5.9655,
      "mean_token_accuracy": 0.14521736949682235,
      "num_tokens": 32252385.0,
      "step": 15925
    },
    {
      "entropy": 6.209153461456299,
      "epoch": 1.392117451717207,
      "grad_norm": 1.859375,
      "learning_rate": 0.00048103568842893527,
      "loss": 5.9163,
      "mean_token_accuracy": 0.145188207924366,
      "num_tokens": 32263537.0,
      "step": 15930
    },
    {
      "entropy": 6.183942461013794,
      "epoch": 1.3925544000699117,
      "grad_norm": 1.90625,
      "learning_rate": 0.00048102316498094467,
      "loss": 5.9264,
      "mean_token_accuracy": 0.1478283680975437,
      "num_tokens": 32273450.0,
      "step": 15935
    },
    {
      "entropy": 6.1553215980529785,
      "epoch": 1.3929913484226164,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048101063758132186,
      "loss": 5.8802,
      "mean_token_accuracy": 0.1517511449754238,
      "num_tokens": 32283243.0,
      "step": 15940
    },
    {
      "entropy": 6.11683611869812,
      "epoch": 1.393428296775321,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004809981062303071,
      "loss": 5.8512,
      "mean_token_accuracy": 0.1503870740532875,
      "num_tokens": 32292458.0,
      "step": 15945
    },
    {
      "entropy": 6.18965163230896,
      "epoch": 1.3938652451280258,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048098557092814066,
      "loss": 5.8614,
      "mean_token_accuracy": 0.14753165915608407,
      "num_tokens": 32302574.0,
      "step": 15950
    },
    {
      "entropy": 6.1169227123260494,
      "epoch": 1.3943021934807307,
      "grad_norm": 2.0,
      "learning_rate": 0.000480973031675063,
      "loss": 5.8437,
      "mean_token_accuracy": 0.15693144202232362,
      "num_tokens": 32312658.0,
      "step": 15955
    },
    {
      "entropy": 6.168728446960449,
      "epoch": 1.3947391418334352,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004809604884713148,
      "loss": 5.9579,
      "mean_token_accuracy": 0.15188750624656677,
      "num_tokens": 32322717.0,
      "step": 15960
    },
    {
      "entropy": 6.131944751739502,
      "epoch": 1.39517609018614,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048094794131713645,
      "loss": 5.861,
      "mean_token_accuracy": 0.1567105695605278,
      "num_tokens": 32333316.0,
      "step": 15965
    },
    {
      "entropy": 6.188786745071411,
      "epoch": 1.3956130385388448,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004809353902127686,
      "loss": 5.8929,
      "mean_token_accuracy": 0.15127069801092147,
      "num_tokens": 32343857.0,
      "step": 15970
    },
    {
      "entropy": 6.116759538650513,
      "epoch": 1.3960499868915495,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048092283515845216,
      "loss": 5.7933,
      "mean_token_accuracy": 0.157977394759655,
      "num_tokens": 32353883.0,
      "step": 15975
    },
    {
      "entropy": 6.045762014389038,
      "epoch": 1.3964869352442542,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004809102761544278,
      "loss": 5.7867,
      "mean_token_accuracy": 0.15422953963279723,
      "num_tokens": 32363355.0,
      "step": 15980
    },
    {
      "entropy": 6.151582336425781,
      "epoch": 1.3969238835969588,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004808977132009364,
      "loss": 5.8707,
      "mean_token_accuracy": 0.15060843676328659,
      "num_tokens": 32373066.0,
      "step": 15985
    },
    {
      "entropy": 6.129072904586792,
      "epoch": 1.3973608319496635,
      "grad_norm": 1.8671875,
      "learning_rate": 0.000480885146298219,
      "loss": 5.8415,
      "mean_token_accuracy": 0.15550466179847716,
      "num_tokens": 32383720.0,
      "step": 15990
    },
    {
      "entropy": 6.18328824043274,
      "epoch": 1.3977977803023682,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004808725754465167,
      "loss": 5.8839,
      "mean_token_accuracy": 0.14687421619892121,
      "num_tokens": 32393823.0,
      "step": 15995
    },
    {
      "entropy": 6.090304279327393,
      "epoch": 1.398234728655073,
      "grad_norm": 2.5,
      "learning_rate": 0.00048086000064607044,
      "loss": 5.7918,
      "mean_token_accuracy": 0.16058974340558052,
      "num_tokens": 32404745.0,
      "step": 16000
    },
    {
      "entropy": 6.121276426315307,
      "epoch": 1.3986716770077776,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048084742189712147,
      "loss": 5.8583,
      "mean_token_accuracy": 0.1604558691382408,
      "num_tokens": 32415325.0,
      "step": 16005
    },
    {
      "entropy": 6.180326843261719,
      "epoch": 1.3991086253604825,
      "grad_norm": 2.375,
      "learning_rate": 0.0004808348391999112,
      "loss": 5.9299,
      "mean_token_accuracy": 0.14896190762519837,
      "num_tokens": 32425810.0,
      "step": 16010
    },
    {
      "entropy": 6.181202220916748,
      "epoch": 1.399545573713187,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004808222525546808,
      "loss": 5.898,
      "mean_token_accuracy": 0.15360227674245835,
      "num_tokens": 32436799.0,
      "step": 16015
    },
    {
      "entropy": 6.1136504173278805,
      "epoch": 1.399982522065892,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048080966196167163,
      "loss": 5.8196,
      "mean_token_accuracy": 0.1552768036723137,
      "num_tokens": 32446409.0,
      "step": 16020
    },
    {
      "entropy": 6.143660068511963,
      "epoch": 1.4004194704185966,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004807970674211255,
      "loss": 5.871,
      "mean_token_accuracy": 0.1488344267010689,
      "num_tokens": 32456268.0,
      "step": 16025
    },
    {
      "entropy": 6.109568881988525,
      "epoch": 1.4008564187713013,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048078446893328364,
      "loss": 5.9595,
      "mean_token_accuracy": 0.14640575498342515,
      "num_tokens": 32467622.0,
      "step": 16030
    },
    {
      "entropy": 6.19821720123291,
      "epoch": 1.401293367124006,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004807718664983879,
      "loss": 5.9063,
      "mean_token_accuracy": 0.15022818595170975,
      "num_tokens": 32477753.0,
      "step": 16035
    },
    {
      "entropy": 6.182455921173096,
      "epoch": 1.4017303154767107,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048075926011667997,
      "loss": 5.8057,
      "mean_token_accuracy": 0.15594885796308516,
      "num_tokens": 32487758.0,
      "step": 16040
    },
    {
      "entropy": 6.08267765045166,
      "epoch": 1.4021672638294154,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048074664978840147,
      "loss": 5.8551,
      "mean_token_accuracy": 0.15419219434261322,
      "num_tokens": 32497725.0,
      "step": 16045
    },
    {
      "entropy": 6.134919738769531,
      "epoch": 1.40260421218212,
      "grad_norm": 1.75,
      "learning_rate": 0.0004807340355137945,
      "loss": 5.7749,
      "mean_token_accuracy": 0.1592742070555687,
      "num_tokens": 32507515.0,
      "step": 16050
    },
    {
      "entropy": 6.138142251968384,
      "epoch": 1.4030411605348247,
      "grad_norm": 3.078125,
      "learning_rate": 0.000480721417293101,
      "loss": 5.8096,
      "mean_token_accuracy": 0.15780906230211258,
      "num_tokens": 32516943.0,
      "step": 16055
    },
    {
      "entropy": 6.06535496711731,
      "epoch": 1.4034781088875294,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004807087951265628,
      "loss": 5.7503,
      "mean_token_accuracy": 0.16902215480804444,
      "num_tokens": 32527223.0,
      "step": 16060
    },
    {
      "entropy": 6.12424521446228,
      "epoch": 1.4039150572402341,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004806961690144222,
      "loss": 5.836,
      "mean_token_accuracy": 0.15572343319654464,
      "num_tokens": 32536668.0,
      "step": 16065
    },
    {
      "entropy": 6.160105180740357,
      "epoch": 1.4043520055929388,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004806835389569213,
      "loss": 5.875,
      "mean_token_accuracy": 0.15538810044527054,
      "num_tokens": 32546566.0,
      "step": 16070
    },
    {
      "entropy": 6.0662766456604,
      "epoch": 1.4047889539456437,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048067090495430235,
      "loss": 5.7922,
      "mean_token_accuracy": 0.1557777628302574,
      "num_tokens": 32555927.0,
      "step": 16075
    },
    {
      "entropy": 6.173278856277466,
      "epoch": 1.4052259022983482,
      "grad_norm": 2.375,
      "learning_rate": 0.00048065826700680763,
      "loss": 5.8285,
      "mean_token_accuracy": 0.1510176494717598,
      "num_tokens": 32566569.0,
      "step": 16080
    },
    {
      "entropy": 6.1212835788726805,
      "epoch": 1.4056628506510531,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048064562511467956,
      "loss": 5.8047,
      "mean_token_accuracy": 0.15893922448158265,
      "num_tokens": 32577076.0,
      "step": 16085
    },
    {
      "entropy": 6.1175226211547855,
      "epoch": 1.4060997990037578,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048063297927816067,
      "loss": 5.8931,
      "mean_token_accuracy": 0.1452830120921135,
      "num_tokens": 32587926.0,
      "step": 16090
    },
    {
      "entropy": 6.199253511428833,
      "epoch": 1.4065367473564625,
      "grad_norm": 1.875,
      "learning_rate": 0.0004806203294974935,
      "loss": 5.9456,
      "mean_token_accuracy": 0.15250675678253173,
      "num_tokens": 32599012.0,
      "step": 16095
    },
    {
      "entropy": 6.211430263519287,
      "epoch": 1.4069736957091672,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048060767577292067,
      "loss": 5.8373,
      "mean_token_accuracy": 0.15296147912740707,
      "num_tokens": 32609094.0,
      "step": 16100
    },
    {
      "entropy": 6.109265851974487,
      "epoch": 1.4074106440618719,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004805950181046849,
      "loss": 5.9055,
      "mean_token_accuracy": 0.1492737740278244,
      "num_tokens": 32620356.0,
      "step": 16105
    },
    {
      "entropy": 6.159668636322022,
      "epoch": 1.4078475924145766,
      "grad_norm": 2.4375,
      "learning_rate": 0.000480582356493029,
      "loss": 5.8821,
      "mean_token_accuracy": 0.15144226998090743,
      "num_tokens": 32629841.0,
      "step": 16110
    },
    {
      "entropy": 6.156450986862183,
      "epoch": 1.4082845407672813,
      "grad_norm": 1.875,
      "learning_rate": 0.0004805696909381957,
      "loss": 5.9017,
      "mean_token_accuracy": 0.15453390926122665,
      "num_tokens": 32639006.0,
      "step": 16115
    },
    {
      "entropy": 6.1538920402526855,
      "epoch": 1.408721489119986,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048055702144042794,
      "loss": 5.8764,
      "mean_token_accuracy": 0.15112707912921905,
      "num_tokens": 32648419.0,
      "step": 16120
    },
    {
      "entropy": 6.151839399337769,
      "epoch": 1.4091584374726907,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004805443479999689,
      "loss": 5.8913,
      "mean_token_accuracy": 0.1580185577273369,
      "num_tokens": 32658050.0,
      "step": 16125
    },
    {
      "entropy": 6.197067975997925,
      "epoch": 1.4095953858253956,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004805316706170615,
      "loss": 5.8735,
      "mean_token_accuracy": 0.14884357899427414,
      "num_tokens": 32668176.0,
      "step": 16130
    },
    {
      "entropy": 6.071049928665161,
      "epoch": 1.4100323341781,
      "grad_norm": 4.59375,
      "learning_rate": 0.00048051898929194907,
      "loss": 5.7842,
      "mean_token_accuracy": 0.1572985127568245,
      "num_tokens": 32678517.0,
      "step": 16135
    },
    {
      "entropy": 6.122809267044067,
      "epoch": 1.410469282530805,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004805063040248747,
      "loss": 5.8322,
      "mean_token_accuracy": 0.15304223895072938,
      "num_tokens": 32688459.0,
      "step": 16140
    },
    {
      "entropy": 6.1579183578491214,
      "epoch": 1.4109062308835096,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004804936148160817,
      "loss": 5.8627,
      "mean_token_accuracy": 0.15743836015462875,
      "num_tokens": 32698481.0,
      "step": 16145
    },
    {
      "entropy": 6.121659803390503,
      "epoch": 1.4113431792362143,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004804809216658135,
      "loss": 5.8826,
      "mean_token_accuracy": 0.14823081195354462,
      "num_tokens": 32709031.0,
      "step": 16150
    },
    {
      "entropy": 6.130120420455933,
      "epoch": 1.411780127588919,
      "grad_norm": 2.125,
      "learning_rate": 0.00048046822457431347,
      "loss": 5.8815,
      "mean_token_accuracy": 0.15891651660203934,
      "num_tokens": 32720293.0,
      "step": 16155
    },
    {
      "entropy": 6.1630120277404785,
      "epoch": 1.4122170759416237,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048045552354182536,
      "loss": 5.9028,
      "mean_token_accuracy": 0.15743406265974044,
      "num_tokens": 32730108.0,
      "step": 16160
    },
    {
      "entropy": 6.163303184509277,
      "epoch": 1.4126540242943284,
      "grad_norm": 2.25,
      "learning_rate": 0.00048044281856859255,
      "loss": 5.8795,
      "mean_token_accuracy": 0.15551360696554184,
      "num_tokens": 32740614.0,
      "step": 16165
    },
    {
      "entropy": 6.1726726531982425,
      "epoch": 1.413090972647033,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004804301096548589,
      "loss": 5.8176,
      "mean_token_accuracy": 0.15367681086063384,
      "num_tokens": 32751155.0,
      "step": 16170
    },
    {
      "entropy": 6.102839374542237,
      "epoch": 1.4135279209997378,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004804173968008681,
      "loss": 5.7854,
      "mean_token_accuracy": 0.15775360763072968,
      "num_tokens": 32761067.0,
      "step": 16175
    },
    {
      "entropy": 6.067156505584717,
      "epoch": 1.4139648693524425,
      "grad_norm": 1.984375,
      "learning_rate": 0.00048040468000686395,
      "loss": 5.7565,
      "mean_token_accuracy": 0.1634780600667,
      "num_tokens": 32771357.0,
      "step": 16180
    },
    {
      "entropy": 6.154537391662598,
      "epoch": 1.4144018177051472,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004803919592730904,
      "loss": 5.9657,
      "mean_token_accuracy": 0.15175942927598954,
      "num_tokens": 32781116.0,
      "step": 16185
    },
    {
      "entropy": 6.17523889541626,
      "epoch": 1.4148387660578519,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004803792345997915,
      "loss": 5.9262,
      "mean_token_accuracy": 0.14918609112501144,
      "num_tokens": 32791660.0,
      "step": 16190
    },
    {
      "entropy": 6.178508949279785,
      "epoch": 1.4152757144105568,
      "grad_norm": 3.46875,
      "learning_rate": 0.00048036650598721125,
      "loss": 5.8589,
      "mean_token_accuracy": 0.15216931849718093,
      "num_tokens": 32801413.0,
      "step": 16195
    },
    {
      "entropy": 6.168720483779907,
      "epoch": 1.4157126627632615,
      "grad_norm": 2.125,
      "learning_rate": 0.00048035377343559377,
      "loss": 5.9225,
      "mean_token_accuracy": 0.15241920799016953,
      "num_tokens": 32811293.0,
      "step": 16200
    },
    {
      "entropy": 6.173145055770874,
      "epoch": 1.4161496111159662,
      "grad_norm": 3.671875,
      "learning_rate": 0.00048034103694518336,
      "loss": 5.912,
      "mean_token_accuracy": 0.15666632652282714,
      "num_tokens": 32821609.0,
      "step": 16205
    },
    {
      "entropy": 6.213352060317993,
      "epoch": 1.4165865594686708,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048032829651622425,
      "loss": 5.863,
      "mean_token_accuracy": 0.15040107369422911,
      "num_tokens": 32831219.0,
      "step": 16210
    },
    {
      "entropy": 6.163787889480591,
      "epoch": 1.4170235078213755,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048031555214896076,
      "loss": 5.8897,
      "mean_token_accuracy": 0.14951475113630294,
      "num_tokens": 32841085.0,
      "step": 16215
    },
    {
      "entropy": 6.106598234176635,
      "epoch": 1.4174604561740802,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004803028038436375,
      "loss": 5.8497,
      "mean_token_accuracy": 0.1578148603439331,
      "num_tokens": 32851592.0,
      "step": 16220
    },
    {
      "entropy": 6.174910926818848,
      "epoch": 1.417897404526785,
      "grad_norm": 3.03125,
      "learning_rate": 0.00048029005160049886,
      "loss": 5.8388,
      "mean_token_accuracy": 0.15424165725708008,
      "num_tokens": 32862124.0,
      "step": 16225
    },
    {
      "entropy": 6.105254411697388,
      "epoch": 1.4183343528794896,
      "grad_norm": 5.375,
      "learning_rate": 0.0004802772954197894,
      "loss": 5.825,
      "mean_token_accuracy": 0.16686519086360932,
      "num_tokens": 32872760.0,
      "step": 16230
    },
    {
      "entropy": 6.1387444019317625,
      "epoch": 1.4187713012321943,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048026453530175393,
      "loss": 5.9392,
      "mean_token_accuracy": 0.15198744982481002,
      "num_tokens": 32883542.0,
      "step": 16235
    },
    {
      "entropy": 6.160576486587525,
      "epoch": 1.419208249584899,
      "grad_norm": 3.53125,
      "learning_rate": 0.00048025177124663704,
      "loss": 5.8266,
      "mean_token_accuracy": 0.15290183424949647,
      "num_tokens": 32893922.0,
      "step": 16240
    },
    {
      "entropy": 6.176000118255615,
      "epoch": 1.4196451979376037,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048023900325468367,
      "loss": 5.8765,
      "mean_token_accuracy": 0.15981116741895676,
      "num_tokens": 32903907.0,
      "step": 16245
    },
    {
      "entropy": 6.179763984680176,
      "epoch": 1.4200821462903086,
      "grad_norm": 3.203125,
      "learning_rate": 0.00048022623132613863,
      "loss": 5.8848,
      "mean_token_accuracy": 0.15064170509576796,
      "num_tokens": 32914656.0,
      "step": 16250
    },
    {
      "entropy": 6.138255262374878,
      "epoch": 1.420519094643013,
      "grad_norm": 2.34375,
      "learning_rate": 0.000480213455461247,
      "loss": 5.902,
      "mean_token_accuracy": 0.1474309653043747,
      "num_tokens": 32925250.0,
      "step": 16255
    },
    {
      "entropy": 6.138006639480591,
      "epoch": 1.420956042995718,
      "grad_norm": 3.796875,
      "learning_rate": 0.00048020067566025373,
      "loss": 5.8617,
      "mean_token_accuracy": 0.15242878198623658,
      "num_tokens": 32935435.0,
      "step": 16260
    },
    {
      "entropy": 6.19958004951477,
      "epoch": 1.4213929913484227,
      "grad_norm": 2.203125,
      "learning_rate": 0.000480187891923404,
      "loss": 5.8708,
      "mean_token_accuracy": 0.15308659821748732,
      "num_tokens": 32946166.0,
      "step": 16265
    },
    {
      "entropy": 6.146883201599121,
      "epoch": 1.4218299397011274,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004801751042509429,
      "loss": 5.8234,
      "mean_token_accuracy": 0.15771649926900863,
      "num_tokens": 32955751.0,
      "step": 16270
    },
    {
      "entropy": 6.084358215332031,
      "epoch": 1.422266888053832,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004801623126431159,
      "loss": 5.9056,
      "mean_token_accuracy": 0.1521984726190567,
      "num_tokens": 32966121.0,
      "step": 16275
    },
    {
      "entropy": 6.160526180267334,
      "epoch": 1.4227038364065367,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004801495171001681,
      "loss": 5.8701,
      "mean_token_accuracy": 0.1625432178378105,
      "num_tokens": 32976950.0,
      "step": 16280
    },
    {
      "entropy": 6.181528854370117,
      "epoch": 1.4231407847592414,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004801367176223452,
      "loss": 5.8395,
      "mean_token_accuracy": 0.16046028435230256,
      "num_tokens": 32987278.0,
      "step": 16285
    },
    {
      "entropy": 6.141334152221679,
      "epoch": 1.4235777331119461,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004801239142098926,
      "loss": 5.9374,
      "mean_token_accuracy": 0.14678418934345244,
      "num_tokens": 32997050.0,
      "step": 16290
    },
    {
      "entropy": 6.168014812469482,
      "epoch": 1.4240146814646508,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004801111068630557,
      "loss": 5.8598,
      "mean_token_accuracy": 0.157503941655159,
      "num_tokens": 33007799.0,
      "step": 16295
    },
    {
      "entropy": 6.19782829284668,
      "epoch": 1.4244516298173555,
      "grad_norm": 8.8125,
      "learning_rate": 0.00048009829558208044,
      "loss": 5.8573,
      "mean_token_accuracy": 0.15878252387046815,
      "num_tokens": 33017476.0,
      "step": 16300
    },
    {
      "entropy": 6.101291751861572,
      "epoch": 1.4248885781700604,
      "grad_norm": 1.8125,
      "learning_rate": 0.00048008548036721226,
      "loss": 5.8389,
      "mean_token_accuracy": 0.15810855329036713,
      "num_tokens": 33027617.0,
      "step": 16305
    },
    {
      "entropy": 6.143493032455444,
      "epoch": 1.425325526522765,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004800726612186972,
      "loss": 5.8965,
      "mean_token_accuracy": 0.15185931473970413,
      "num_tokens": 33037257.0,
      "step": 16310
    },
    {
      "entropy": 6.13328537940979,
      "epoch": 1.4257624748754698,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004800598381367811,
      "loss": 5.8202,
      "mean_token_accuracy": 0.1564851865172386,
      "num_tokens": 33047406.0,
      "step": 16315
    },
    {
      "entropy": 6.109671020507813,
      "epoch": 1.4261994232281745,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004800470111217096,
      "loss": 5.8809,
      "mean_token_accuracy": 0.161655480414629,
      "num_tokens": 33057975.0,
      "step": 16320
    },
    {
      "entropy": 6.200307559967041,
      "epoch": 1.4266363715808792,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048003418017372924,
      "loss": 5.9035,
      "mean_token_accuracy": 0.14856939017772675,
      "num_tokens": 33068645.0,
      "step": 16325
    },
    {
      "entropy": 6.20980920791626,
      "epoch": 1.4270733199335839,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004800213452930857,
      "loss": 5.8856,
      "mean_token_accuracy": 0.1545938655734062,
      "num_tokens": 33079147.0,
      "step": 16330
    },
    {
      "entropy": 6.1469982147216795,
      "epoch": 1.4275102682862886,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004800085064800255,
      "loss": 5.8323,
      "mean_token_accuracy": 0.149704609811306,
      "num_tokens": 33089229.0,
      "step": 16335
    },
    {
      "entropy": 6.134866523742676,
      "epoch": 1.4279472166389933,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004799956637347946,
      "loss": 5.852,
      "mean_token_accuracy": 0.15586744323372842,
      "num_tokens": 33099688.0,
      "step": 16340
    },
    {
      "entropy": 6.1218828678131105,
      "epoch": 1.428384164991698,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004799828170576395,
      "loss": 5.8657,
      "mean_token_accuracy": 0.15585654079914094,
      "num_tokens": 33110019.0,
      "step": 16345
    },
    {
      "entropy": 6.020287752151489,
      "epoch": 1.4288211133444026,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004799699664488065,
      "loss": 5.7511,
      "mean_token_accuracy": 0.16514252573251725,
      "num_tokens": 33120141.0,
      "step": 16350
    },
    {
      "entropy": 6.134222984313965,
      "epoch": 1.4292580616971073,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004799571119085422,
      "loss": 5.8259,
      "mean_token_accuracy": 0.1553936094045639,
      "num_tokens": 33130047.0,
      "step": 16355
    },
    {
      "entropy": 6.186822462081909,
      "epoch": 1.429695010049812,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004799442534370931,
      "loss": 5.9241,
      "mean_token_accuracy": 0.1481953263282776,
      "num_tokens": 33141234.0,
      "step": 16360
    },
    {
      "entropy": 6.117707538604736,
      "epoch": 1.4301319584025167,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004799313910347057,
      "loss": 5.8664,
      "mean_token_accuracy": 0.16003192961215973,
      "num_tokens": 33150788.0,
      "step": 16365
    },
    {
      "entropy": 6.095549154281616,
      "epoch": 1.4305689067552216,
      "grad_norm": 2.96875,
      "learning_rate": 0.000479918524701627,
      "loss": 5.8634,
      "mean_token_accuracy": 0.15991369783878326,
      "num_tokens": 33160928.0,
      "step": 16370
    },
    {
      "entropy": 6.177237462997437,
      "epoch": 1.431005855107926,
      "grad_norm": 2.25,
      "learning_rate": 0.00047990565443810357,
      "loss": 5.9644,
      "mean_token_accuracy": 0.14254036098718642,
      "num_tokens": 33171333.0,
      "step": 16375
    },
    {
      "entropy": 6.048695850372314,
      "epoch": 1.431442803460631,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047989278024438233,
      "loss": 5.7536,
      "mean_token_accuracy": 0.16269160956144332,
      "num_tokens": 33180396.0,
      "step": 16380
    },
    {
      "entropy": 6.073717784881592,
      "epoch": 1.4318797518133357,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004798799021207102,
      "loss": 5.8351,
      "mean_token_accuracy": 0.15462781935930253,
      "num_tokens": 33189571.0,
      "step": 16385
    },
    {
      "entropy": 6.07733678817749,
      "epoch": 1.4323167001660404,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047986702006733417,
      "loss": 5.7978,
      "mean_token_accuracy": 0.15696240216493607,
      "num_tokens": 33198979.0,
      "step": 16390
    },
    {
      "entropy": 6.107527589797973,
      "epoch": 1.432753648518745,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047985413408450136,
      "loss": 5.846,
      "mean_token_accuracy": 0.16013509184122085,
      "num_tokens": 33208862.0,
      "step": 16395
    },
    {
      "entropy": 6.1095493793487545,
      "epoch": 1.4331905968714498,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004798412441724589,
      "loss": 5.8985,
      "mean_token_accuracy": 0.14833230525255203,
      "num_tokens": 33218804.0,
      "step": 16400
    },
    {
      "entropy": 6.23686408996582,
      "epoch": 1.4336275452241545,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047982835033145413,
      "loss": 5.8954,
      "mean_token_accuracy": 0.14775953441858292,
      "num_tokens": 33228783.0,
      "step": 16405
    },
    {
      "entropy": 6.243742513656616,
      "epoch": 1.4340644935768592,
      "grad_norm": 3.203125,
      "learning_rate": 0.00047981545256173415,
      "loss": 5.9379,
      "mean_token_accuracy": 0.1458210825920105,
      "num_tokens": 33241029.0,
      "step": 16410
    },
    {
      "entropy": 6.181990623474121,
      "epoch": 1.4345014419295639,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004798025508635465,
      "loss": 5.9464,
      "mean_token_accuracy": 0.14418817758560182,
      "num_tokens": 33250383.0,
      "step": 16415
    },
    {
      "entropy": 6.099666118621826,
      "epoch": 1.4349383902822685,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004797896452371386,
      "loss": 5.7188,
      "mean_token_accuracy": 0.1675995633006096,
      "num_tokens": 33260424.0,
      "step": 16420
    },
    {
      "entropy": 6.167949771881103,
      "epoch": 1.4353753386349735,
      "grad_norm": 2.5,
      "learning_rate": 0.00047977673568275805,
      "loss": 5.8136,
      "mean_token_accuracy": 0.14974709004163742,
      "num_tokens": 33270200.0,
      "step": 16425
    },
    {
      "entropy": 6.1727800369262695,
      "epoch": 1.435812286987678,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004797638222006524,
      "loss": 5.9702,
      "mean_token_accuracy": 0.1439051777124405,
      "num_tokens": 33279850.0,
      "step": 16430
    },
    {
      "entropy": 6.143870735168457,
      "epoch": 1.4362492353403828,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004797509047910693,
      "loss": 5.8284,
      "mean_token_accuracy": 0.15714142099022865,
      "num_tokens": 33289053.0,
      "step": 16435
    },
    {
      "entropy": 6.18264331817627,
      "epoch": 1.4366861836930875,
      "grad_norm": 3.75,
      "learning_rate": 0.00047973798345425655,
      "loss": 5.853,
      "mean_token_accuracy": 0.15542616695165634,
      "num_tokens": 33298988.0,
      "step": 16440
    },
    {
      "entropy": 6.090325212478637,
      "epoch": 1.4371231320457922,
      "grad_norm": 2.6875,
      "learning_rate": 0.00047972505819046207,
      "loss": 5.8474,
      "mean_token_accuracy": 0.15734313577413558,
      "num_tokens": 33308683.0,
      "step": 16445
    },
    {
      "entropy": 6.0857462882995605,
      "epoch": 1.437560080398497,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004797121289999336,
      "loss": 5.8438,
      "mean_token_accuracy": 0.14859300330281258,
      "num_tokens": 33319230.0,
      "step": 16450
    },
    {
      "entropy": 6.212685823440552,
      "epoch": 1.4379970287512016,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004796991958829193,
      "loss": 5.9095,
      "mean_token_accuracy": 0.14558419734239578,
      "num_tokens": 33329277.0,
      "step": 16455
    },
    {
      "entropy": 6.1804320335388185,
      "epoch": 1.4384339771039063,
      "grad_norm": 2.875,
      "learning_rate": 0.0004796862588396672,
      "loss": 5.8331,
      "mean_token_accuracy": 0.1521575629711151,
      "num_tokens": 33338644.0,
      "step": 16460
    },
    {
      "entropy": 6.154590892791748,
      "epoch": 1.438870925456611,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047967331787042536,
      "loss": 5.7726,
      "mean_token_accuracy": 0.1656111016869545,
      "num_tokens": 33348996.0,
      "step": 16465
    },
    {
      "entropy": 6.159158754348755,
      "epoch": 1.4393078738093157,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047966037297544195,
      "loss": 5.8598,
      "mean_token_accuracy": 0.15389342457056046,
      "num_tokens": 33358581.0,
      "step": 16470
    },
    {
      "entropy": 6.130748081207275,
      "epoch": 1.4397448221620204,
      "grad_norm": 4.3125,
      "learning_rate": 0.00047964742415496553,
      "loss": 5.9035,
      "mean_token_accuracy": 0.15477489084005355,
      "num_tokens": 33368925.0,
      "step": 16475
    },
    {
      "entropy": 6.131583547592163,
      "epoch": 1.4401817705147253,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004796344714092441,
      "loss": 5.8988,
      "mean_token_accuracy": 0.15620177909731864,
      "num_tokens": 33378918.0,
      "step": 16480
    },
    {
      "entropy": 6.107097434997558,
      "epoch": 1.4406187188674298,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004796215147385264,
      "loss": 5.881,
      "mean_token_accuracy": 0.15190946012735368,
      "num_tokens": 33389039.0,
      "step": 16485
    },
    {
      "entropy": 6.183122301101685,
      "epoch": 1.4410556672201347,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004796085541430608,
      "loss": 5.8142,
      "mean_token_accuracy": 0.16158209294080733,
      "num_tokens": 33398893.0,
      "step": 16490
    },
    {
      "entropy": 6.185396146774292,
      "epoch": 1.4414926155728394,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004795955896230959,
      "loss": 5.8967,
      "mean_token_accuracy": 0.15102121606469154,
      "num_tokens": 33408188.0,
      "step": 16495
    },
    {
      "entropy": 6.116655921936035,
      "epoch": 1.441929563925544,
      "grad_norm": 2.5,
      "learning_rate": 0.00047958262117888035,
      "loss": 5.8601,
      "mean_token_accuracy": 0.15978309661149978,
      "num_tokens": 33418926.0,
      "step": 16500
    },
    {
      "entropy": 6.137224912643433,
      "epoch": 1.4423665122782487,
      "grad_norm": 4.0,
      "learning_rate": 0.00047956964881066297,
      "loss": 5.892,
      "mean_token_accuracy": 0.15085428953170776,
      "num_tokens": 33429790.0,
      "step": 16505
    },
    {
      "entropy": 6.1617601871490475,
      "epoch": 1.4428034606309534,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004795566725186925,
      "loss": 5.8679,
      "mean_token_accuracy": 0.15546119213104248,
      "num_tokens": 33440470.0,
      "step": 16510
    },
    {
      "entropy": 6.132394504547119,
      "epoch": 1.4432404089836581,
      "grad_norm": 2.5,
      "learning_rate": 0.0004795436923032179,
      "loss": 5.8521,
      "mean_token_accuracy": 0.15429432392120362,
      "num_tokens": 33449875.0,
      "step": 16515
    },
    {
      "entropy": 6.145984840393067,
      "epoch": 1.4436773573363628,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004795307081644881,
      "loss": 5.8337,
      "mean_token_accuracy": 0.1543783277273178,
      "num_tokens": 33458841.0,
      "step": 16520
    },
    {
      "entropy": 6.109430551528931,
      "epoch": 1.4441143056890675,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004795177201027521,
      "loss": 5.7869,
      "mean_token_accuracy": 0.15995844304561616,
      "num_tokens": 33468656.0,
      "step": 16525
    },
    {
      "entropy": 6.125645494461059,
      "epoch": 1.4445512540417722,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047950472811825916,
      "loss": 5.8677,
      "mean_token_accuracy": 0.1534265995025635,
      "num_tokens": 33478977.0,
      "step": 16530
    },
    {
      "entropy": 6.140962409973144,
      "epoch": 1.444988202394477,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004794917322112583,
      "loss": 5.83,
      "mean_token_accuracy": 0.15178718864917756,
      "num_tokens": 33488418.0,
      "step": 16535
    },
    {
      "entropy": 6.101782512664795,
      "epoch": 1.4454251507471816,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004794787323819989,
      "loss": 5.8268,
      "mean_token_accuracy": 0.15940861105918885,
      "num_tokens": 33499125.0,
      "step": 16540
    },
    {
      "entropy": 6.119931888580322,
      "epoch": 1.4458620990998865,
      "grad_norm": 2.875,
      "learning_rate": 0.00047946572863073023,
      "loss": 5.8593,
      "mean_token_accuracy": 0.1551864489912987,
      "num_tokens": 33508826.0,
      "step": 16545
    },
    {
      "entropy": 6.058589887619019,
      "epoch": 1.446299047452591,
      "grad_norm": 3.40625,
      "learning_rate": 0.00047945272095770174,
      "loss": 5.7741,
      "mean_token_accuracy": 0.1704589009284973,
      "num_tokens": 33518398.0,
      "step": 16550
    },
    {
      "entropy": 6.140607261657715,
      "epoch": 1.4467359958052959,
      "grad_norm": 3.75,
      "learning_rate": 0.000479439709363163,
      "loss": 5.8728,
      "mean_token_accuracy": 0.15202014595270158,
      "num_tokens": 33528289.0,
      "step": 16555
    },
    {
      "entropy": 6.199735116958618,
      "epoch": 1.4471729441580006,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004794266938473635,
      "loss": 5.9483,
      "mean_token_accuracy": 0.14671660512685775,
      "num_tokens": 33539228.0,
      "step": 16560
    },
    {
      "entropy": 6.148595476150513,
      "epoch": 1.4476098925107053,
      "grad_norm": 2.65625,
      "learning_rate": 0.00047941367441055283,
      "loss": 5.8986,
      "mean_token_accuracy": 0.15249387919902802,
      "num_tokens": 33551015.0,
      "step": 16565
    },
    {
      "entropy": 6.174115991592407,
      "epoch": 1.44804684086341,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047940065105298073,
      "loss": 5.8368,
      "mean_token_accuracy": 0.15895399302244187,
      "num_tokens": 33560374.0,
      "step": 16570
    },
    {
      "entropy": 6.054957389831543,
      "epoch": 1.4484837892161146,
      "grad_norm": 2.125,
      "learning_rate": 0.0004793876237748972,
      "loss": 5.7117,
      "mean_token_accuracy": 0.16946170032024382,
      "num_tokens": 33569916.0,
      "step": 16575
    },
    {
      "entropy": 6.032973527908325,
      "epoch": 1.4489207375688193,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004793745925765518,
      "loss": 5.7836,
      "mean_token_accuracy": 0.16140772253274918,
      "num_tokens": 33579930.0,
      "step": 16580
    },
    {
      "entropy": 6.098497581481934,
      "epoch": 1.449357685921524,
      "grad_norm": 2.375,
      "learning_rate": 0.0004793615574581947,
      "loss": 5.8759,
      "mean_token_accuracy": 0.152655665576458,
      "num_tokens": 33588612.0,
      "step": 16585
    },
    {
      "entropy": 6.191740942001343,
      "epoch": 1.4497946342742287,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047934851842007585,
      "loss": 5.8825,
      "mean_token_accuracy": 0.14861786514520645,
      "num_tokens": 33599253.0,
      "step": 16590
    },
    {
      "entropy": 6.181076622009277,
      "epoch": 1.4502315826269334,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004793354754624453,
      "loss": 5.8152,
      "mean_token_accuracy": 0.1626364678144455,
      "num_tokens": 33609484.0,
      "step": 16595
    },
    {
      "entropy": 6.172141408920288,
      "epoch": 1.4506685309796383,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004793224285855533,
      "loss": 5.8219,
      "mean_token_accuracy": 0.1544149249792099,
      "num_tokens": 33619660.0,
      "step": 16600
    },
    {
      "entropy": 6.133356523513794,
      "epoch": 1.4511054793323428,
      "grad_norm": 2.421875,
      "learning_rate": 0.00047930937778965,
      "loss": 5.9138,
      "mean_token_accuracy": 0.15033407211303712,
      "num_tokens": 33629505.0,
      "step": 16605
    },
    {
      "entropy": 6.140414619445801,
      "epoch": 1.4515424276850477,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004792963230749858,
      "loss": 5.9248,
      "mean_token_accuracy": 0.15426726564764975,
      "num_tokens": 33639680.0,
      "step": 16610
    },
    {
      "entropy": 6.164324617385864,
      "epoch": 1.4519793760377524,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047928326444181117,
      "loss": 5.9262,
      "mean_token_accuracy": 0.1457628309726715,
      "num_tokens": 33649200.0,
      "step": 16615
    },
    {
      "entropy": 6.167906141281128,
      "epoch": 1.452416324390457,
      "grad_norm": 3.546875,
      "learning_rate": 0.00047927020189037634,
      "loss": 5.8441,
      "mean_token_accuracy": 0.1539966955780983,
      "num_tokens": 33658386.0,
      "step": 16620
    },
    {
      "entropy": 6.125351095199585,
      "epoch": 1.4528532727431618,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004792571354209321,
      "loss": 5.8061,
      "mean_token_accuracy": 0.156576569378376,
      "num_tokens": 33669139.0,
      "step": 16625
    },
    {
      "entropy": 6.141805696487427,
      "epoch": 1.4532902210958665,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004792440650337289,
      "loss": 5.9229,
      "mean_token_accuracy": 0.15396281853318214,
      "num_tokens": 33679629.0,
      "step": 16630
    },
    {
      "entropy": 6.184770774841309,
      "epoch": 1.4537271694485712,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004792309907290175,
      "loss": 5.9029,
      "mean_token_accuracy": 0.14761248975992203,
      "num_tokens": 33689633.0,
      "step": 16635
    },
    {
      "entropy": 6.150835847854614,
      "epoch": 1.4541641178012759,
      "grad_norm": 2.9375,
      "learning_rate": 0.00047921791250704875,
      "loss": 5.8601,
      "mean_token_accuracy": 0.15278035700321196,
      "num_tokens": 33699719.0,
      "step": 16640
    },
    {
      "entropy": 6.223496007919311,
      "epoch": 1.4546010661539805,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047920483036807346,
      "loss": 5.8938,
      "mean_token_accuracy": 0.1495446264743805,
      "num_tokens": 33709389.0,
      "step": 16645
    },
    {
      "entropy": 6.098867893218994,
      "epoch": 1.4550380145066852,
      "grad_norm": 3.0,
      "learning_rate": 0.00047919174431234246,
      "loss": 5.888,
      "mean_token_accuracy": 0.14994095414876937,
      "num_tokens": 33719063.0,
      "step": 16650
    },
    {
      "entropy": 6.185135364532471,
      "epoch": 1.4554749628593902,
      "grad_norm": 2.375,
      "learning_rate": 0.0004791786543401068,
      "loss": 5.967,
      "mean_token_accuracy": 0.144273891299963,
      "num_tokens": 33729798.0,
      "step": 16655
    },
    {
      "entropy": 6.22833399772644,
      "epoch": 1.4559119112120946,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004791655604516176,
      "loss": 5.894,
      "mean_token_accuracy": 0.1472241647541523,
      "num_tokens": 33740012.0,
      "step": 16660
    },
    {
      "entropy": 6.107581281661988,
      "epoch": 1.4563488595647995,
      "grad_norm": 4.28125,
      "learning_rate": 0.00047915246264712595,
      "loss": 5.8403,
      "mean_token_accuracy": 0.158633478730917,
      "num_tokens": 33749478.0,
      "step": 16665
    },
    {
      "entropy": 6.059699249267578,
      "epoch": 1.4567858079175042,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004791393609268831,
      "loss": 5.716,
      "mean_token_accuracy": 0.1739884614944458,
      "num_tokens": 33758935.0,
      "step": 16670
    },
    {
      "entropy": 6.125878190994262,
      "epoch": 1.457222756270209,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004791262552911404,
      "loss": 5.8118,
      "mean_token_accuracy": 0.15728891044855117,
      "num_tokens": 33769035.0,
      "step": 16675
    },
    {
      "entropy": 6.098521614074707,
      "epoch": 1.4576597046229136,
      "grad_norm": 3.1875,
      "learning_rate": 0.00047911314574014904,
      "loss": 5.834,
      "mean_token_accuracy": 0.15414561033248902,
      "num_tokens": 33779288.0,
      "step": 16680
    },
    {
      "entropy": 6.161070251464844,
      "epoch": 1.4580966529756183,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004791000322741606,
      "loss": 5.8601,
      "mean_token_accuracy": 0.1579616993665695,
      "num_tokens": 33789138.0,
      "step": 16685
    },
    {
      "entropy": 6.193655776977539,
      "epoch": 1.458533601328323,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004790869148934267,
      "loss": 5.8629,
      "mean_token_accuracy": 0.14765916317701339,
      "num_tokens": 33798649.0,
      "step": 16690
    },
    {
      "entropy": 6.007314872741699,
      "epoch": 1.4589705496810277,
      "grad_norm": 2.125,
      "learning_rate": 0.00047907379359819877,
      "loss": 5.7806,
      "mean_token_accuracy": 0.16793847233057022,
      "num_tokens": 33808742.0,
      "step": 16695
    },
    {
      "entropy": 6.1155003070831295,
      "epoch": 1.4594074980337324,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004790606683887286,
      "loss": 5.825,
      "mean_token_accuracy": 0.15763843283057213,
      "num_tokens": 33819507.0,
      "step": 16700
    },
    {
      "entropy": 6.193852949142456,
      "epoch": 1.459844446386437,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047904753926526777,
      "loss": 5.8891,
      "mean_token_accuracy": 0.14575171768665313,
      "num_tokens": 33829722.0,
      "step": 16705
    },
    {
      "entropy": 6.109727239608764,
      "epoch": 1.4602813947391418,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004790344062280683,
      "loss": 5.9019,
      "mean_token_accuracy": 0.15260518491268157,
      "num_tokens": 33839350.0,
      "step": 16710
    },
    {
      "entropy": 6.134222173690796,
      "epoch": 1.4607183430918464,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047902126927738196,
      "loss": 5.8518,
      "mean_token_accuracy": 0.15978336334228516,
      "num_tokens": 33849233.0,
      "step": 16715
    },
    {
      "entropy": 6.14475736618042,
      "epoch": 1.4611552914445514,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047900812841346086,
      "loss": 5.9247,
      "mean_token_accuracy": 0.14815541505813598,
      "num_tokens": 33859996.0,
      "step": 16720
    },
    {
      "entropy": 6.133819532394409,
      "epoch": 1.4615922397972558,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004789949836365569,
      "loss": 5.8895,
      "mean_token_accuracy": 0.15771810114383697,
      "num_tokens": 33870780.0,
      "step": 16725
    },
    {
      "entropy": 6.147455263137817,
      "epoch": 1.4620291881499607,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004789818349469222,
      "loss": 5.8763,
      "mean_token_accuracy": 0.15781297236680986,
      "num_tokens": 33880182.0,
      "step": 16730
    },
    {
      "entropy": 6.196677732467651,
      "epoch": 1.4624661365026654,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004789686823448091,
      "loss": 5.893,
      "mean_token_accuracy": 0.15342501997947694,
      "num_tokens": 33889464.0,
      "step": 16735
    },
    {
      "entropy": 6.157237815856933,
      "epoch": 1.4629030848553701,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004789555258304698,
      "loss": 5.8395,
      "mean_token_accuracy": 0.15037917494773864,
      "num_tokens": 33899530.0,
      "step": 16740
    },
    {
      "entropy": 6.101448488235474,
      "epoch": 1.4633400332080748,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004789423654041566,
      "loss": 5.9005,
      "mean_token_accuracy": 0.15652785450220108,
      "num_tokens": 33910094.0,
      "step": 16745
    },
    {
      "entropy": 6.1704583168029785,
      "epoch": 1.4637769815607795,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00047892920106612204,
      "loss": 5.8295,
      "mean_token_accuracy": 0.15678684264421464,
      "num_tokens": 33921181.0,
      "step": 16750
    },
    {
      "entropy": 6.175072479248047,
      "epoch": 1.4642139299134842,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004789160328166184,
      "loss": 5.9464,
      "mean_token_accuracy": 0.15273372381925582,
      "num_tokens": 33931567.0,
      "step": 16755
    },
    {
      "entropy": 6.192115306854248,
      "epoch": 1.464650878266189,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047890286065589856,
      "loss": 5.888,
      "mean_token_accuracy": 0.15326565504074097,
      "num_tokens": 33941513.0,
      "step": 16760
    },
    {
      "entropy": 6.104971075057984,
      "epoch": 1.4650878266188936,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047888968458421487,
      "loss": 5.8547,
      "mean_token_accuracy": 0.1533613994717598,
      "num_tokens": 33953400.0,
      "step": 16765
    },
    {
      "entropy": 6.130507564544677,
      "epoch": 1.4655247749715983,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004788765046018204,
      "loss": 5.8997,
      "mean_token_accuracy": 0.14915316104888915,
      "num_tokens": 33964394.0,
      "step": 16770
    },
    {
      "entropy": 6.193925952911377,
      "epoch": 1.4659617233243032,
      "grad_norm": 3.140625,
      "learning_rate": 0.00047886332070896747,
      "loss": 6.0017,
      "mean_token_accuracy": 0.141588294506073,
      "num_tokens": 33975266.0,
      "step": 16775
    },
    {
      "entropy": 6.223922300338745,
      "epoch": 1.4663986716770077,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004788501329059094,
      "loss": 5.941,
      "mean_token_accuracy": 0.14766502901911735,
      "num_tokens": 33986159.0,
      "step": 16780
    },
    {
      "entropy": 6.226293897628784,
      "epoch": 1.4668356200297126,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047883694119289896,
      "loss": 5.8964,
      "mean_token_accuracy": 0.15321390926837922,
      "num_tokens": 33997792.0,
      "step": 16785
    },
    {
      "entropy": 6.183793449401856,
      "epoch": 1.4672725683824173,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047882374557018916,
      "loss": 5.8602,
      "mean_token_accuracy": 0.155574768781662,
      "num_tokens": 34007945.0,
      "step": 16790
    },
    {
      "entropy": 6.199473142623901,
      "epoch": 1.467709516735122,
      "grad_norm": 3.96875,
      "learning_rate": 0.00047881054603803307,
      "loss": 5.9363,
      "mean_token_accuracy": 0.15616601407527925,
      "num_tokens": 34018399.0,
      "step": 16795
    },
    {
      "entropy": 6.166972923278808,
      "epoch": 1.4681464650878266,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004787973425966839,
      "loss": 5.9662,
      "mean_token_accuracy": 0.15201805159449577,
      "num_tokens": 34028267.0,
      "step": 16800
    },
    {
      "entropy": 6.185429906845092,
      "epoch": 1.4685834134405313,
      "grad_norm": 2.59375,
      "learning_rate": 0.000478784135246395,
      "loss": 5.9943,
      "mean_token_accuracy": 0.141352641582489,
      "num_tokens": 34037904.0,
      "step": 16805
    },
    {
      "entropy": 6.184894752502442,
      "epoch": 1.469020361793236,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004787709239874196,
      "loss": 5.8308,
      "mean_token_accuracy": 0.16382694393396377,
      "num_tokens": 34049203.0,
      "step": 16810
    },
    {
      "entropy": 6.192157649993897,
      "epoch": 1.4694573101459407,
      "grad_norm": 1.890625,
      "learning_rate": 0.000478757708820011,
      "loss": 5.9218,
      "mean_token_accuracy": 0.1531964898109436,
      "num_tokens": 34059670.0,
      "step": 16815
    },
    {
      "entropy": 6.1121491432189945,
      "epoch": 1.4698942584986454,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004787444897444229,
      "loss": 5.8363,
      "mean_token_accuracy": 0.15862599313259124,
      "num_tokens": 34069730.0,
      "step": 16820
    },
    {
      "entropy": 6.19559326171875,
      "epoch": 1.47033120685135,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004787312667609086,
      "loss": 5.865,
      "mean_token_accuracy": 0.153972090780735,
      "num_tokens": 34079315.0,
      "step": 16825
    },
    {
      "entropy": 6.145363664627075,
      "epoch": 1.470768155204055,
      "grad_norm": 2.453125,
      "learning_rate": 0.000478718039869722,
      "loss": 5.8624,
      "mean_token_accuracy": 0.15099019557237625,
      "num_tokens": 34089431.0,
      "step": 16830
    },
    {
      "entropy": 6.1299206733703615,
      "epoch": 1.4712051035567595,
      "grad_norm": 2.25,
      "learning_rate": 0.00047870480907111657,
      "loss": 5.8507,
      "mean_token_accuracy": 0.1506814807653427,
      "num_tokens": 34100120.0,
      "step": 16835
    },
    {
      "entropy": 6.155368232727051,
      "epoch": 1.4716420519094644,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004786915743653461,
      "loss": 5.8705,
      "mean_token_accuracy": 0.15664769858121871,
      "num_tokens": 34110271.0,
      "step": 16840
    },
    {
      "entropy": 6.202772951126098,
      "epoch": 1.472079000262169,
      "grad_norm": 2.71875,
      "learning_rate": 0.00047867833575266456,
      "loss": 5.934,
      "mean_token_accuracy": 0.14555186480283738,
      "num_tokens": 34120240.0,
      "step": 16845
    },
    {
      "entropy": 6.192800045013428,
      "epoch": 1.4725159486148738,
      "grad_norm": 4.625,
      "learning_rate": 0.0004786650932333259,
      "loss": 5.88,
      "mean_token_accuracy": 0.15328696072101594,
      "num_tokens": 34130813.0,
      "step": 16850
    },
    {
      "entropy": 6.159699964523315,
      "epoch": 1.4729528969675785,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004786518468075839,
      "loss": 5.8971,
      "mean_token_accuracy": 0.15006352439522744,
      "num_tokens": 34141169.0,
      "step": 16855
    },
    {
      "entropy": 6.1492616653442385,
      "epoch": 1.4733898453202832,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004786385964756928,
      "loss": 5.8825,
      "mean_token_accuracy": 0.15355351567268372,
      "num_tokens": 34151869.0,
      "step": 16860
    },
    {
      "entropy": 6.19665675163269,
      "epoch": 1.4738267936729879,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004786253422379068,
      "loss": 5.9562,
      "mean_token_accuracy": 0.14665970504283904,
      "num_tokens": 34162608.0,
      "step": 16865
    },
    {
      "entropy": 6.158228588104248,
      "epoch": 1.4742637420256925,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00047861208409448,
      "loss": 5.8708,
      "mean_token_accuracy": 0.15374520421028137,
      "num_tokens": 34172648.0,
      "step": 16870
    },
    {
      "entropy": 6.156950426101685,
      "epoch": 1.4747006903783972,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004785988220456666,
      "loss": 5.8943,
      "mean_token_accuracy": 0.15451784282922745,
      "num_tokens": 34183603.0,
      "step": 16875
    },
    {
      "entropy": 6.212537717819214,
      "epoch": 1.475137638731102,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004785855560917213,
      "loss": 5.901,
      "mean_token_accuracy": 0.14746190309524537,
      "num_tokens": 34193830.0,
      "step": 16880
    },
    {
      "entropy": 6.075336360931397,
      "epoch": 1.4755745870838066,
      "grad_norm": 4.34375,
      "learning_rate": 0.00047857228623289825,
      "loss": 5.7356,
      "mean_token_accuracy": 0.17408724129199982,
      "num_tokens": 34203074.0,
      "step": 16885
    },
    {
      "entropy": 6.114908838272095,
      "epoch": 1.4760115354365113,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00047855901246945205,
      "loss": 5.9593,
      "mean_token_accuracy": 0.14402876049280167,
      "num_tokens": 34212958.0,
      "step": 16890
    },
    {
      "entropy": 6.12075777053833,
      "epoch": 1.4764484837892162,
      "grad_norm": 4.25,
      "learning_rate": 0.0004785457348016374,
      "loss": 5.8837,
      "mean_token_accuracy": 0.14680899307131767,
      "num_tokens": 34223077.0,
      "step": 16895
    },
    {
      "entropy": 6.17345552444458,
      "epoch": 1.4768854321419207,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047853245322970885,
      "loss": 5.8748,
      "mean_token_accuracy": 0.15817682594060897,
      "num_tokens": 34233386.0,
      "step": 16900
    },
    {
      "entropy": 6.197332143783569,
      "epoch": 1.4773223804946256,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004785191677539212,
      "loss": 5.8864,
      "mean_token_accuracy": 0.15587843507528304,
      "num_tokens": 34244488.0,
      "step": 16905
    },
    {
      "entropy": 6.134504079818726,
      "epoch": 1.4777593288473303,
      "grad_norm": 1.9375,
      "learning_rate": 0.00047850587837452936,
      "loss": 5.8642,
      "mean_token_accuracy": 0.15110252797603607,
      "num_tokens": 34254047.0,
      "step": 16910
    },
    {
      "entropy": 6.137491989135742,
      "epoch": 1.478196277200035,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004784925850917881,
      "loss": 5.8483,
      "mean_token_accuracy": 0.14915971457958221,
      "num_tokens": 34264297.0,
      "step": 16915
    },
    {
      "entropy": 6.243809223175049,
      "epoch": 1.4786332255527397,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004784792879059524,
      "loss": 5.931,
      "mean_token_accuracy": 0.14868390411138535,
      "num_tokens": 34274945.0,
      "step": 16920
    },
    {
      "entropy": 6.139284992218018,
      "epoch": 1.4790701739054444,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047846598681727733,
      "loss": 5.8087,
      "mean_token_accuracy": 0.15301438719034194,
      "num_tokens": 34284914.0,
      "step": 16925
    },
    {
      "entropy": 6.06685848236084,
      "epoch": 1.479507122258149,
      "grad_norm": 1.9140625,
      "learning_rate": 0.000478452681826018,
      "loss": 5.8709,
      "mean_token_accuracy": 0.15490325689315795,
      "num_tokens": 34294277.0,
      "step": 16930
    },
    {
      "entropy": 6.2253889560699465,
      "epoch": 1.4799440706108538,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047843937293242967,
      "loss": 5.9426,
      "mean_token_accuracy": 0.15257695764303209,
      "num_tokens": 34304865.0,
      "step": 16935
    },
    {
      "entropy": 6.0872008323669435,
      "epoch": 1.4803810189635584,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047842606013676753,
      "loss": 5.7504,
      "mean_token_accuracy": 0.165998575091362,
      "num_tokens": 34315404.0,
      "step": 16940
    },
    {
      "entropy": 6.23872013092041,
      "epoch": 1.4808179673162631,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047841274343928697,
      "loss": 5.8733,
      "mean_token_accuracy": 0.1495609849691391,
      "num_tokens": 34325448.0,
      "step": 16945
    },
    {
      "entropy": 6.1050526142120365,
      "epoch": 1.481254915668968,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004783994228402434,
      "loss": 5.865,
      "mean_token_accuracy": 0.16074959486722945,
      "num_tokens": 34335266.0,
      "step": 16950
    },
    {
      "entropy": 6.163719987869262,
      "epoch": 1.4816918640216725,
      "grad_norm": 3.125,
      "learning_rate": 0.0004783860983398924,
      "loss": 5.8591,
      "mean_token_accuracy": 0.15349616706371308,
      "num_tokens": 34345847.0,
      "step": 16955
    },
    {
      "entropy": 6.157124805450439,
      "epoch": 1.4821288123743774,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004783727699384894,
      "loss": 5.8096,
      "mean_token_accuracy": 0.15444130301475525,
      "num_tokens": 34355718.0,
      "step": 16960
    },
    {
      "entropy": 6.073325824737549,
      "epoch": 1.4825657607270821,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004783594376362901,
      "loss": 5.8795,
      "mean_token_accuracy": 0.15474819391965866,
      "num_tokens": 34366100.0,
      "step": 16965
    },
    {
      "entropy": 6.178067350387574,
      "epoch": 1.4830027090797868,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004783461014335502,
      "loss": 5.9614,
      "mean_token_accuracy": 0.15003640651702882,
      "num_tokens": 34376977.0,
      "step": 16970
    },
    {
      "entropy": 6.2077248096466064,
      "epoch": 1.4834396574324915,
      "grad_norm": 2.375,
      "learning_rate": 0.0004783327613305256,
      "loss": 5.9,
      "mean_token_accuracy": 0.15152034908533096,
      "num_tokens": 34387782.0,
      "step": 16975
    },
    {
      "entropy": 6.092386054992676,
      "epoch": 1.4838766057851962,
      "grad_norm": 1.7890625,
      "learning_rate": 0.000478319417327472,
      "loss": 5.8218,
      "mean_token_accuracy": 0.1603291943669319,
      "num_tokens": 34397321.0,
      "step": 16980
    },
    {
      "entropy": 6.114581108093262,
      "epoch": 1.484313554137901,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004783060694246455,
      "loss": 5.899,
      "mean_token_accuracy": 0.1481311537325382,
      "num_tokens": 34407594.0,
      "step": 16985
    },
    {
      "entropy": 6.279845666885376,
      "epoch": 1.4847505024906056,
      "grad_norm": 2.375,
      "learning_rate": 0.0004782927176223021,
      "loss": 5.894,
      "mean_token_accuracy": 0.15344629883766175,
      "num_tokens": 34417337.0,
      "step": 16990
    },
    {
      "entropy": 6.093864631652832,
      "epoch": 1.4851874508433103,
      "grad_norm": 1.625,
      "learning_rate": 0.0004782793619206977,
      "loss": 5.7818,
      "mean_token_accuracy": 0.1617438644170761,
      "num_tokens": 34427643.0,
      "step": 16995
    },
    {
      "entropy": 6.112289524078369,
      "epoch": 1.485624399196015,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004782660023200887,
      "loss": 5.8903,
      "mean_token_accuracy": 0.15189099535346032,
      "num_tokens": 34437224.0,
      "step": 17000
    },
    {
      "entropy": 6.230650997161865,
      "epoch": 1.4860613475487199,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004782526388207313,
      "loss": 5.8937,
      "mean_token_accuracy": 0.14694590047001838,
      "num_tokens": 34447376.0,
      "step": 17005
    },
    {
      "entropy": 6.130454730987549,
      "epoch": 1.4864982959014243,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004782392714228818,
      "loss": 5.9059,
      "mean_token_accuracy": 0.1544419229030609,
      "num_tokens": 34457649.0,
      "step": 17010
    },
    {
      "entropy": 6.139226198196411,
      "epoch": 1.4869352442541293,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047822590012679654,
      "loss": 5.9019,
      "mean_token_accuracy": 0.157633076608181,
      "num_tokens": 34467437.0,
      "step": 17015
    },
    {
      "entropy": 6.196960973739624,
      "epoch": 1.487372192606834,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047821252493273204,
      "loss": 5.8212,
      "mean_token_accuracy": 0.1524882048368454,
      "num_tokens": 34477615.0,
      "step": 17020
    },
    {
      "entropy": 6.137586975097657,
      "epoch": 1.4878091409595386,
      "grad_norm": 2.625,
      "learning_rate": 0.0004781991458409448,
      "loss": 5.8225,
      "mean_token_accuracy": 0.1533292680978775,
      "num_tokens": 34487619.0,
      "step": 17025
    },
    {
      "entropy": 6.155377435684204,
      "epoch": 1.4882460893122433,
      "grad_norm": 1.96875,
      "learning_rate": 0.00047818576285169154,
      "loss": 5.8847,
      "mean_token_accuracy": 0.1548856280744076,
      "num_tokens": 34498837.0,
      "step": 17030
    },
    {
      "entropy": 6.130313444137573,
      "epoch": 1.488683037664948,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047817237596522885,
      "loss": 5.8574,
      "mean_token_accuracy": 0.15142044201493263,
      "num_tokens": 34508972.0,
      "step": 17035
    },
    {
      "entropy": 6.218903732299805,
      "epoch": 1.4891199860176527,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004781589851818135,
      "loss": 5.8451,
      "mean_token_accuracy": 0.1547143667936325,
      "num_tokens": 34519289.0,
      "step": 17040
    },
    {
      "entropy": 6.10342435836792,
      "epoch": 1.4895569343703574,
      "grad_norm": 3.25,
      "learning_rate": 0.00047814559050170236,
      "loss": 5.7664,
      "mean_token_accuracy": 0.16532005369663239,
      "num_tokens": 34528384.0,
      "step": 17045
    },
    {
      "entropy": 6.074532175064087,
      "epoch": 1.489993882723062,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00047813219192515235,
      "loss": 5.8363,
      "mean_token_accuracy": 0.14611624032258988,
      "num_tokens": 34537442.0,
      "step": 17050
    },
    {
      "entropy": 6.185675048828125,
      "epoch": 1.4904308310757668,
      "grad_norm": 2.0,
      "learning_rate": 0.0004781187894524205,
      "loss": 5.902,
      "mean_token_accuracy": 0.15343325957655907,
      "num_tokens": 34547960.0,
      "step": 17055
    },
    {
      "entropy": 6.139957427978516,
      "epoch": 1.4908677794284715,
      "grad_norm": 1.859375,
      "learning_rate": 0.00047810538308376376,
      "loss": 5.8101,
      "mean_token_accuracy": 0.14967811107635498,
      "num_tokens": 34556730.0,
      "step": 17060
    },
    {
      "entropy": 6.061622381210327,
      "epoch": 1.4913047277811762,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004780919728194394,
      "loss": 5.8754,
      "mean_token_accuracy": 0.1490348219871521,
      "num_tokens": 34566751.0,
      "step": 17065
    },
    {
      "entropy": 6.25036654472351,
      "epoch": 1.491741676133881,
      "grad_norm": 1.984375,
      "learning_rate": 0.00047807855865970454,
      "loss": 5.8678,
      "mean_token_accuracy": 0.15316146612167358,
      "num_tokens": 34576417.0,
      "step": 17070
    },
    {
      "entropy": 6.226842546463013,
      "epoch": 1.4921786244865856,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004780651406048166,
      "loss": 5.9638,
      "mean_token_accuracy": 0.1471695989370346,
      "num_tokens": 34586976.0,
      "step": 17075
    },
    {
      "entropy": 6.133858394622803,
      "epoch": 1.4926155728392905,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004780517186550328,
      "loss": 5.8593,
      "mean_token_accuracy": 0.15699538290500642,
      "num_tokens": 34597374.0,
      "step": 17080
    },
    {
      "entropy": 6.116224718093872,
      "epoch": 1.4930525211919952,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004780382928106107,
      "loss": 5.8185,
      "mean_token_accuracy": 0.1594260722398758,
      "num_tokens": 34607586.0,
      "step": 17085
    },
    {
      "entropy": 6.1564229965209964,
      "epoch": 1.4934894695446999,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047802486307180767,
      "loss": 5.9007,
      "mean_token_accuracy": 0.15488080382347108,
      "num_tokens": 34617152.0,
      "step": 17090
    },
    {
      "entropy": 6.124329280853272,
      "epoch": 1.4939264178974045,
      "grad_norm": 2.0,
      "learning_rate": 0.00047801142943888135,
      "loss": 5.8304,
      "mean_token_accuracy": 0.16123347729444504,
      "num_tokens": 34627421.0,
      "step": 17095
    },
    {
      "entropy": 6.161361265182495,
      "epoch": 1.4943633662501092,
      "grad_norm": 2.734375,
      "learning_rate": 0.00047799799191208955,
      "loss": 5.8443,
      "mean_token_accuracy": 0.15437712520360947,
      "num_tokens": 34637687.0,
      "step": 17100
    },
    {
      "entropy": 6.192191648483276,
      "epoch": 1.494800314602814,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004779845504916898,
      "loss": 5.901,
      "mean_token_accuracy": 0.1494130253791809,
      "num_tokens": 34647884.0,
      "step": 17105
    },
    {
      "entropy": 6.16855640411377,
      "epoch": 1.4952372629555186,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004779711051779401,
      "loss": 5.8809,
      "mean_token_accuracy": 0.1491269662976265,
      "num_tokens": 34658658.0,
      "step": 17110
    },
    {
      "entropy": 6.1588623046875,
      "epoch": 1.4956742113082233,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004779576559710982,
      "loss": 5.8907,
      "mean_token_accuracy": 0.15391636043787002,
      "num_tokens": 34668728.0,
      "step": 17115
    },
    {
      "entropy": 6.185825300216675,
      "epoch": 1.496111159660928,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047794420287142203,
      "loss": 5.8828,
      "mean_token_accuracy": 0.15038785636425017,
      "num_tokens": 34678576.0,
      "step": 17120
    },
    {
      "entropy": 6.2679033279418945,
      "epoch": 1.496548108013633,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047793074587916984,
      "loss": 5.9461,
      "mean_token_accuracy": 0.14634464979171752,
      "num_tokens": 34688920.0,
      "step": 17125
    },
    {
      "entropy": 6.149936866760254,
      "epoch": 1.4969850563663374,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047791728499459955,
      "loss": 5.8771,
      "mean_token_accuracy": 0.15319266617298127,
      "num_tokens": 34699348.0,
      "step": 17130
    },
    {
      "entropy": 6.107307958602905,
      "epoch": 1.4974220047190423,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004779038202179694,
      "loss": 5.7727,
      "mean_token_accuracy": 0.1620291978120804,
      "num_tokens": 34708991.0,
      "step": 17135
    },
    {
      "entropy": 6.164907312393188,
      "epoch": 1.497858953071747,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047789035154953776,
      "loss": 5.881,
      "mean_token_accuracy": 0.14633265882730484,
      "num_tokens": 34720869.0,
      "step": 17140
    },
    {
      "entropy": 6.117620849609375,
      "epoch": 1.4982959014244517,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004778768789895627,
      "loss": 5.8205,
      "mean_token_accuracy": 0.15676439553499222,
      "num_tokens": 34731025.0,
      "step": 17145
    },
    {
      "entropy": 6.097006177902221,
      "epoch": 1.4987328497771564,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004778634025383029,
      "loss": 5.8917,
      "mean_token_accuracy": 0.15143688768148422,
      "num_tokens": 34741846.0,
      "step": 17150
    },
    {
      "entropy": 6.106971263885498,
      "epoch": 1.499169798129861,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047784992219601673,
      "loss": 5.7706,
      "mean_token_accuracy": 0.17086363434791565,
      "num_tokens": 34753294.0,
      "step": 17155
    },
    {
      "entropy": 6.222295331954956,
      "epoch": 1.4996067464825658,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004778364379629627,
      "loss": 5.8604,
      "mean_token_accuracy": 0.15040425434708596,
      "num_tokens": 34763997.0,
      "step": 17160
    },
    {
      "entropy": 6.085229969024658,
      "epoch": 1.5000436948352704,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047782294983939955,
      "loss": 5.7707,
      "mean_token_accuracy": 0.1629262536764145,
      "num_tokens": 34772925.0,
      "step": 17165
    },
    {
      "entropy": 6.124052572250366,
      "epoch": 1.5004806431879751,
      "grad_norm": 2.125,
      "learning_rate": 0.0004778094578255859,
      "loss": 5.8837,
      "mean_token_accuracy": 0.15400953888893126,
      "num_tokens": 34782608.0,
      "step": 17170
    },
    {
      "entropy": 6.089442443847656,
      "epoch": 1.5009175915406798,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004777959619217807,
      "loss": 5.8225,
      "mean_token_accuracy": 0.16333565413951873,
      "num_tokens": 34792533.0,
      "step": 17175
    },
    {
      "entropy": 6.188339281082153,
      "epoch": 1.5013545398933847,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047778246212824263,
      "loss": 5.9562,
      "mean_token_accuracy": 0.14302828162908554,
      "num_tokens": 34803069.0,
      "step": 17180
    },
    {
      "entropy": 6.115234422683716,
      "epoch": 1.5017914882460892,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004777689584452306,
      "loss": 5.8841,
      "mean_token_accuracy": 0.15647624284029008,
      "num_tokens": 34811925.0,
      "step": 17185
    },
    {
      "entropy": 6.0832971096038815,
      "epoch": 1.5022284365987941,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004777554508730038,
      "loss": 5.8168,
      "mean_token_accuracy": 0.15610338002443314,
      "num_tokens": 34821511.0,
      "step": 17190
    },
    {
      "entropy": 6.138559246063233,
      "epoch": 1.5026653849514986,
      "grad_norm": 2.25,
      "learning_rate": 0.0004777419394118212,
      "loss": 5.9162,
      "mean_token_accuracy": 0.14611384645104408,
      "num_tokens": 34831683.0,
      "step": 17195
    },
    {
      "entropy": 6.227632617950439,
      "epoch": 1.5031023333042035,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004777284240619419,
      "loss": 5.8725,
      "mean_token_accuracy": 0.15374660789966582,
      "num_tokens": 34841578.0,
      "step": 17200
    },
    {
      "entropy": 6.162811422348023,
      "epoch": 1.5035392816569082,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004777149048236252,
      "loss": 5.8593,
      "mean_token_accuracy": 0.15482842922210693,
      "num_tokens": 34851959.0,
      "step": 17205
    },
    {
      "entropy": 6.082300519943237,
      "epoch": 1.503976230009613,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047770138169713047,
      "loss": 5.9133,
      "mean_token_accuracy": 0.14467027336359023,
      "num_tokens": 34862557.0,
      "step": 17210
    },
    {
      "entropy": 6.185537672042846,
      "epoch": 1.5044131783623176,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047768785468271704,
      "loss": 5.9212,
      "mean_token_accuracy": 0.15043128728866578,
      "num_tokens": 34872394.0,
      "step": 17215
    },
    {
      "entropy": 6.207429647445679,
      "epoch": 1.5048501267150223,
      "grad_norm": 1.828125,
      "learning_rate": 0.00047767432378064426,
      "loss": 5.8779,
      "mean_token_accuracy": 0.1513463541865349,
      "num_tokens": 34881938.0,
      "step": 17220
    },
    {
      "entropy": 6.158588123321533,
      "epoch": 1.505287075067727,
      "grad_norm": 2.125,
      "learning_rate": 0.00047766078899117174,
      "loss": 5.8561,
      "mean_token_accuracy": 0.1548664763569832,
      "num_tokens": 34892769.0,
      "step": 17225
    },
    {
      "entropy": 6.175883531570435,
      "epoch": 1.5057240234204317,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047764725031455927,
      "loss": 5.9726,
      "mean_token_accuracy": 0.14331179484724998,
      "num_tokens": 34903643.0,
      "step": 17230
    },
    {
      "entropy": 6.148700475692749,
      "epoch": 1.5061609717731366,
      "grad_norm": 2.5,
      "learning_rate": 0.0004776337077510662,
      "loss": 5.8744,
      "mean_token_accuracy": 0.14294183254241943,
      "num_tokens": 34913558.0,
      "step": 17235
    },
    {
      "entropy": 6.175105619430542,
      "epoch": 1.506597920125841,
      "grad_norm": 2.6875,
      "learning_rate": 0.00047762016130095246,
      "loss": 5.8489,
      "mean_token_accuracy": 0.15276338756084443,
      "num_tokens": 34923653.0,
      "step": 17240
    },
    {
      "entropy": 6.118389415740967,
      "epoch": 1.507034868478546,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047760661096447784,
      "loss": 5.8822,
      "mean_token_accuracy": 0.1490948513150215,
      "num_tokens": 34933381.0,
      "step": 17245
    },
    {
      "entropy": 6.124467945098877,
      "epoch": 1.5074718168312504,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004775930567419023,
      "loss": 5.7892,
      "mean_token_accuracy": 0.16079355478286744,
      "num_tokens": 34942486.0,
      "step": 17250
    },
    {
      "entropy": 6.147614049911499,
      "epoch": 1.5079087651839553,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004775794986334857,
      "loss": 5.925,
      "mean_token_accuracy": 0.14684897810220718,
      "num_tokens": 34952291.0,
      "step": 17255
    },
    {
      "entropy": 6.199529218673706,
      "epoch": 1.5083457135366598,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004775659366394883,
      "loss": 5.9296,
      "mean_token_accuracy": 0.1480625316500664,
      "num_tokens": 34963842.0,
      "step": 17260
    },
    {
      "entropy": 6.112629318237305,
      "epoch": 1.5087826618893647,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047755237076016997,
      "loss": 5.8021,
      "mean_token_accuracy": 0.15716639012098313,
      "num_tokens": 34973057.0,
      "step": 17265
    },
    {
      "entropy": 6.179137659072876,
      "epoch": 1.5092196102420694,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004775388009957911,
      "loss": 5.9797,
      "mean_token_accuracy": 0.14639114066958428,
      "num_tokens": 34983257.0,
      "step": 17270
    },
    {
      "entropy": 6.179181623458862,
      "epoch": 1.509656558594774,
      "grad_norm": 2.375,
      "learning_rate": 0.00047752522734661193,
      "loss": 5.8392,
      "mean_token_accuracy": 0.15043409168720245,
      "num_tokens": 34993066.0,
      "step": 17275
    },
    {
      "entropy": 6.176004648208618,
      "epoch": 1.5100935069474788,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047751164981289264,
      "loss": 5.8312,
      "mean_token_accuracy": 0.15084033384919165,
      "num_tokens": 35002830.0,
      "step": 17280
    },
    {
      "entropy": 6.094789361953735,
      "epoch": 1.5105304553001835,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004774980683948939,
      "loss": 5.842,
      "mean_token_accuracy": 0.15349375754594802,
      "num_tokens": 35011837.0,
      "step": 17285
    },
    {
      "entropy": 6.1739795207977295,
      "epoch": 1.5109674036528882,
      "grad_norm": 2.625,
      "learning_rate": 0.00047748448309287606,
      "loss": 5.8379,
      "mean_token_accuracy": 0.1536051645874977,
      "num_tokens": 35021589.0,
      "step": 17290
    },
    {
      "entropy": 6.149427890777588,
      "epoch": 1.5114043520055929,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047747089390709973,
      "loss": 5.9056,
      "mean_token_accuracy": 0.15581002384424208,
      "num_tokens": 35032207.0,
      "step": 17295
    },
    {
      "entropy": 6.164715242385864,
      "epoch": 1.5118413003582978,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004774573008378256,
      "loss": 5.9344,
      "mean_token_accuracy": 0.14515655487775803,
      "num_tokens": 35042661.0,
      "step": 17300
    },
    {
      "entropy": 6.180689668655395,
      "epoch": 1.5122782487110022,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004774437038853143,
      "loss": 5.8917,
      "mean_token_accuracy": 0.15099122375249863,
      "num_tokens": 35053050.0,
      "step": 17305
    },
    {
      "entropy": 6.247140979766845,
      "epoch": 1.5127151970637072,
      "grad_norm": 2.0,
      "learning_rate": 0.00047743010304982674,
      "loss": 5.9282,
      "mean_token_accuracy": 0.14915898814797401,
      "num_tokens": 35064381.0,
      "step": 17310
    },
    {
      "entropy": 6.180149269104004,
      "epoch": 1.5131521454164116,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004774164983316237,
      "loss": 5.8037,
      "mean_token_accuracy": 0.15262559577822685,
      "num_tokens": 35074787.0,
      "step": 17315
    },
    {
      "entropy": 6.035295391082764,
      "epoch": 1.5135890937691165,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004774028897309662,
      "loss": 5.7895,
      "mean_token_accuracy": 0.1611209601163864,
      "num_tokens": 35085801.0,
      "step": 17320
    },
    {
      "entropy": 6.099165201187134,
      "epoch": 1.5140260421218212,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047738927724811514,
      "loss": 5.8925,
      "mean_token_accuracy": 0.15450572967529297,
      "num_tokens": 35096800.0,
      "step": 17325
    },
    {
      "entropy": 6.124790954589844,
      "epoch": 1.514462990474526,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004773756608833318,
      "loss": 5.7183,
      "mean_token_accuracy": 0.17593078017234803,
      "num_tokens": 35106735.0,
      "step": 17330
    },
    {
      "entropy": 6.081604909896851,
      "epoch": 1.5148999388272306,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047736204063687717,
      "loss": 5.8269,
      "mean_token_accuracy": 0.15834816694259643,
      "num_tokens": 35116221.0,
      "step": 17335
    },
    {
      "entropy": 6.085692739486694,
      "epoch": 1.5153368871799353,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047734841650901255,
      "loss": 5.7771,
      "mean_token_accuracy": 0.1613197699189186,
      "num_tokens": 35124896.0,
      "step": 17340
    },
    {
      "entropy": 6.1304854393005375,
      "epoch": 1.51577383553264,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004773347884999993,
      "loss": 5.8639,
      "mean_token_accuracy": 0.15470646619796752,
      "num_tokens": 35134576.0,
      "step": 17345
    },
    {
      "entropy": 6.128907537460327,
      "epoch": 1.5162107838853447,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004773211566100988,
      "loss": 5.7862,
      "mean_token_accuracy": 0.16112884134054184,
      "num_tokens": 35144210.0,
      "step": 17350
    },
    {
      "entropy": 6.079049205780029,
      "epoch": 1.5166477322380496,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004773075208395725,
      "loss": 5.768,
      "mean_token_accuracy": 0.15846778452396393,
      "num_tokens": 35153669.0,
      "step": 17355
    },
    {
      "entropy": 6.149544954299927,
      "epoch": 1.517084680590754,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004772938811886818,
      "loss": 5.8897,
      "mean_token_accuracy": 0.1511499747633934,
      "num_tokens": 35163469.0,
      "step": 17360
    },
    {
      "entropy": 6.1565367698669435,
      "epoch": 1.517521628943459,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004772802376576887,
      "loss": 5.8992,
      "mean_token_accuracy": 0.15798701643943786,
      "num_tokens": 35173479.0,
      "step": 17365
    },
    {
      "entropy": 6.060641765594482,
      "epoch": 1.5179585772961635,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004772665902468545,
      "loss": 5.8632,
      "mean_token_accuracy": 0.15399756282567978,
      "num_tokens": 35184570.0,
      "step": 17370
    },
    {
      "entropy": 6.1313300132751465,
      "epoch": 1.5183955256488684,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004772529389564412,
      "loss": 5.8796,
      "mean_token_accuracy": 0.15239900052547456,
      "num_tokens": 35193714.0,
      "step": 17375
    },
    {
      "entropy": 6.14327483177185,
      "epoch": 1.518832474001573,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004772392837867104,
      "loss": 5.8906,
      "mean_token_accuracy": 0.1508115939795971,
      "num_tokens": 35203806.0,
      "step": 17380
    },
    {
      "entropy": 6.198706197738647,
      "epoch": 1.5192694223542778,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047722562473792433,
      "loss": 5.8987,
      "mean_token_accuracy": 0.15045736208558083,
      "num_tokens": 35213677.0,
      "step": 17385
    },
    {
      "entropy": 6.171533823013306,
      "epoch": 1.5197063707069824,
      "grad_norm": 1.84375,
      "learning_rate": 0.00047721196181034465,
      "loss": 5.7648,
      "mean_token_accuracy": 0.1512599065899849,
      "num_tokens": 35223663.0,
      "step": 17390
    },
    {
      "entropy": 6.112890720367432,
      "epoch": 1.5201433190596871,
      "grad_norm": 1.984375,
      "learning_rate": 0.00047719829500423374,
      "loss": 5.8337,
      "mean_token_accuracy": 0.15048087984323502,
      "num_tokens": 35234551.0,
      "step": 17395
    },
    {
      "entropy": 6.1172607898712155,
      "epoch": 1.5205802674123918,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047718462431985354,
      "loss": 5.8417,
      "mean_token_accuracy": 0.1563853070139885,
      "num_tokens": 35244350.0,
      "step": 17400
    },
    {
      "entropy": 6.143025302886963,
      "epoch": 1.5210172157650965,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004771709497574662,
      "loss": 5.7976,
      "mean_token_accuracy": 0.15802988708019255,
      "num_tokens": 35253970.0,
      "step": 17405
    },
    {
      "entropy": 6.09180474281311,
      "epoch": 1.5214541641178014,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047715727131733424,
      "loss": 5.8146,
      "mean_token_accuracy": 0.15859377682209014,
      "num_tokens": 35263316.0,
      "step": 17410
    },
    {
      "entropy": 6.159878158569336,
      "epoch": 1.521891112470506,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047714358899971984,
      "loss": 5.7915,
      "mean_token_accuracy": 0.16033942997455597,
      "num_tokens": 35273407.0,
      "step": 17415
    },
    {
      "entropy": 6.13733434677124,
      "epoch": 1.5223280608232108,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004771299028048855,
      "loss": 5.903,
      "mean_token_accuracy": 0.14890057668089868,
      "num_tokens": 35284225.0,
      "step": 17420
    },
    {
      "entropy": 6.1634420394897464,
      "epoch": 1.5227650091759153,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047711621273309365,
      "loss": 5.773,
      "mean_token_accuracy": 0.15862138643860818,
      "num_tokens": 35294747.0,
      "step": 17425
    },
    {
      "entropy": 6.1453797817230225,
      "epoch": 1.5232019575286202,
      "grad_norm": 3.984375,
      "learning_rate": 0.000477102518784607,
      "loss": 5.874,
      "mean_token_accuracy": 0.15801094472408295,
      "num_tokens": 35304856.0,
      "step": 17430
    },
    {
      "entropy": 6.1217848777771,
      "epoch": 1.5236389058813247,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047708882095968817,
      "loss": 5.903,
      "mean_token_accuracy": 0.16101155877113343,
      "num_tokens": 35315550.0,
      "step": 17435
    },
    {
      "entropy": 6.210644817352295,
      "epoch": 1.5240758542340296,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004770751192585997,
      "loss": 5.9505,
      "mean_token_accuracy": 0.14861477687954902,
      "num_tokens": 35325970.0,
      "step": 17440
    },
    {
      "entropy": 6.156119012832642,
      "epoch": 1.5245128025867343,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004770614136816047,
      "loss": 5.8299,
      "mean_token_accuracy": 0.1522540733218193,
      "num_tokens": 35337277.0,
      "step": 17445
    },
    {
      "entropy": 6.154503440856933,
      "epoch": 1.524949750939439,
      "grad_norm": 1.96875,
      "learning_rate": 0.00047704770422896586,
      "loss": 5.8872,
      "mean_token_accuracy": 0.14827334582805635,
      "num_tokens": 35347686.0,
      "step": 17450
    },
    {
      "entropy": 6.173165798187256,
      "epoch": 1.5253866992921437,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047703399090094623,
      "loss": 5.8353,
      "mean_token_accuracy": 0.15608186423778533,
      "num_tokens": 35358727.0,
      "step": 17455
    },
    {
      "entropy": 6.060405349731445,
      "epoch": 1.5258236476448483,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004770202736978087,
      "loss": 5.8818,
      "mean_token_accuracy": 0.15327811241149902,
      "num_tokens": 35369741.0,
      "step": 17460
    },
    {
      "entropy": 6.14546570777893,
      "epoch": 1.526260595997553,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004770065526198166,
      "loss": 5.9368,
      "mean_token_accuracy": 0.15440157651901246,
      "num_tokens": 35380213.0,
      "step": 17465
    },
    {
      "entropy": 6.137426805496216,
      "epoch": 1.5266975443502577,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004769928276672328,
      "loss": 5.8682,
      "mean_token_accuracy": 0.15861286073923112,
      "num_tokens": 35389937.0,
      "step": 17470
    },
    {
      "entropy": 6.143026542663574,
      "epoch": 1.5271344927029626,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004769790988403208,
      "loss": 5.8814,
      "mean_token_accuracy": 0.15100124180316926,
      "num_tokens": 35400534.0,
      "step": 17475
    },
    {
      "entropy": 6.158385133743286,
      "epoch": 1.5275714410556671,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004769653661393438,
      "loss": 5.8668,
      "mean_token_accuracy": 0.1581027925014496,
      "num_tokens": 35410607.0,
      "step": 17480
    },
    {
      "entropy": 6.174805927276611,
      "epoch": 1.528008389408372,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004769516295645654,
      "loss": 5.8717,
      "mean_token_accuracy": 0.15170571506023406,
      "num_tokens": 35420242.0,
      "step": 17485
    },
    {
      "entropy": 6.089843988418579,
      "epoch": 1.5284453377610765,
      "grad_norm": 1.75,
      "learning_rate": 0.00047693788911624876,
      "loss": 5.8229,
      "mean_token_accuracy": 0.1534021645784378,
      "num_tokens": 35429469.0,
      "step": 17490
    },
    {
      "entropy": 6.045228576660156,
      "epoch": 1.5288822861137814,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004769241447946577,
      "loss": 5.8742,
      "mean_token_accuracy": 0.15693089067935945,
      "num_tokens": 35439656.0,
      "step": 17495
    },
    {
      "entropy": 6.152874183654785,
      "epoch": 1.529319234466486,
      "grad_norm": 1.75,
      "learning_rate": 0.0004769103966000557,
      "loss": 5.7434,
      "mean_token_accuracy": 0.15863042771816255,
      "num_tokens": 35448600.0,
      "step": 17500
    },
    {
      "entropy": 6.086853170394898,
      "epoch": 1.5297561828191908,
      "grad_norm": 6.375,
      "learning_rate": 0.0004768966445327065,
      "loss": 5.8423,
      "mean_token_accuracy": 0.15816927403211595,
      "num_tokens": 35458124.0,
      "step": 17505
    },
    {
      "entropy": 6.153998327255249,
      "epoch": 1.5301931311718955,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004768828885928738,
      "loss": 5.8,
      "mean_token_accuracy": 0.15543074607849122,
      "num_tokens": 35467616.0,
      "step": 17510
    },
    {
      "entropy": 6.070661115646362,
      "epoch": 1.5306300795246002,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004768691287808216,
      "loss": 5.8839,
      "mean_token_accuracy": 0.15411746948957444,
      "num_tokens": 35478356.0,
      "step": 17515
    },
    {
      "entropy": 6.144384384155273,
      "epoch": 1.5310670278773049,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004768553650968137,
      "loss": 5.8968,
      "mean_token_accuracy": 0.15931062623858452,
      "num_tokens": 35490213.0,
      "step": 17520
    },
    {
      "entropy": 6.144395160675049,
      "epoch": 1.5315039762300096,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004768415975411142,
      "loss": 5.789,
      "mean_token_accuracy": 0.1629176154732704,
      "num_tokens": 35499982.0,
      "step": 17525
    },
    {
      "entropy": 6.12995457649231,
      "epoch": 1.5319409245827145,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047682782611398693,
      "loss": 5.8842,
      "mean_token_accuracy": 0.15302219837903977,
      "num_tokens": 35510572.0,
      "step": 17530
    },
    {
      "entropy": 6.234933853149414,
      "epoch": 1.532377872935419,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047681405081569637,
      "loss": 5.9495,
      "mean_token_accuracy": 0.14548078253865243,
      "num_tokens": 35521560.0,
      "step": 17535
    },
    {
      "entropy": 6.217502164840698,
      "epoch": 1.5328148212881239,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00047680027164650644,
      "loss": 5.9093,
      "mean_token_accuracy": 0.14836311414837838,
      "num_tokens": 35530879.0,
      "step": 17540
    },
    {
      "entropy": 6.141724014282227,
      "epoch": 1.5332517696408283,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047678648860668163,
      "loss": 5.909,
      "mean_token_accuracy": 0.15604118928313254,
      "num_tokens": 35541210.0,
      "step": 17545
    },
    {
      "entropy": 6.14900803565979,
      "epoch": 1.5336887179935332,
      "grad_norm": 2.0,
      "learning_rate": 0.00047677270169648616,
      "loss": 5.8811,
      "mean_token_accuracy": 0.1519891694188118,
      "num_tokens": 35550149.0,
      "step": 17550
    },
    {
      "entropy": 6.1639406204223635,
      "epoch": 1.534125666346238,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047675891091618464,
      "loss": 5.8188,
      "mean_token_accuracy": 0.15059865713119508,
      "num_tokens": 35561191.0,
      "step": 17555
    },
    {
      "entropy": 6.120050621032715,
      "epoch": 1.5345626146989426,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047674511626604144,
      "loss": 5.8153,
      "mean_token_accuracy": 0.16110145300626755,
      "num_tokens": 35571636.0,
      "step": 17560
    },
    {
      "entropy": 6.10401554107666,
      "epoch": 1.5349995630516473,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004767313177463212,
      "loss": 5.8247,
      "mean_token_accuracy": 0.151840540766716,
      "num_tokens": 35583191.0,
      "step": 17565
    },
    {
      "entropy": 6.083314085006714,
      "epoch": 1.535436511404352,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004767175153572885,
      "loss": 5.8139,
      "mean_token_accuracy": 0.15515257939696311,
      "num_tokens": 35592875.0,
      "step": 17570
    },
    {
      "entropy": 6.14441180229187,
      "epoch": 1.5358734597570567,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004767037090992083,
      "loss": 5.782,
      "mean_token_accuracy": 0.158903169631958,
      "num_tokens": 35602158.0,
      "step": 17575
    },
    {
      "entropy": 6.100735759735107,
      "epoch": 1.5363104081097614,
      "grad_norm": 2.25,
      "learning_rate": 0.00047668989897234516,
      "loss": 5.8161,
      "mean_token_accuracy": 0.16153561770915986,
      "num_tokens": 35612077.0,
      "step": 17580
    },
    {
      "entropy": 6.130440902709961,
      "epoch": 1.5367473564624663,
      "grad_norm": 1.75,
      "learning_rate": 0.00047667608497696415,
      "loss": 5.7458,
      "mean_token_accuracy": 0.16635396778583528,
      "num_tokens": 35621856.0,
      "step": 17585
    },
    {
      "entropy": 6.076784276962281,
      "epoch": 1.5371843048151708,
      "grad_norm": 3.28125,
      "learning_rate": 0.00047666226711333014,
      "loss": 5.8813,
      "mean_token_accuracy": 0.15905992388725282,
      "num_tokens": 35633160.0,
      "step": 17590
    },
    {
      "entropy": 6.200541257858276,
      "epoch": 1.5376212531678757,
      "grad_norm": 3.25,
      "learning_rate": 0.0004766484453817081,
      "loss": 5.8659,
      "mean_token_accuracy": 0.15177130997180938,
      "num_tokens": 35642477.0,
      "step": 17595
    },
    {
      "entropy": 6.158634042739868,
      "epoch": 1.5380582015205801,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004766346197823633,
      "loss": 5.8016,
      "mean_token_accuracy": 0.15790839940309526,
      "num_tokens": 35651938.0,
      "step": 17600
    },
    {
      "entropy": 6.116852426528931,
      "epoch": 1.538495149873285,
      "grad_norm": 3.1875,
      "learning_rate": 0.00047662079031556085,
      "loss": 5.8691,
      "mean_token_accuracy": 0.15003041326999664,
      "num_tokens": 35662881.0,
      "step": 17605
    },
    {
      "entropy": 6.122784233093261,
      "epoch": 1.5389320982259895,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004766069569815659,
      "loss": 5.8265,
      "mean_token_accuracy": 0.15284564197063447,
      "num_tokens": 35673436.0,
      "step": 17610
    },
    {
      "entropy": 6.138750267028809,
      "epoch": 1.5393690465786944,
      "grad_norm": 3.96875,
      "learning_rate": 0.000476593119780644,
      "loss": 5.8255,
      "mean_token_accuracy": 0.1549277886748314,
      "num_tokens": 35683411.0,
      "step": 17615
    },
    {
      "entropy": 6.106433486938476,
      "epoch": 1.5398059949313991,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047657927871306037,
      "loss": 5.8095,
      "mean_token_accuracy": 0.15640153139829635,
      "num_tokens": 35693806.0,
      "step": 17620
    },
    {
      "entropy": 6.098014307022095,
      "epoch": 1.5402429432841038,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004765654337790805,
      "loss": 5.8621,
      "mean_token_accuracy": 0.1564367577433586,
      "num_tokens": 35705015.0,
      "step": 17625
    },
    {
      "entropy": 6.186513805389405,
      "epoch": 1.5406798916368085,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004765515849789701,
      "loss": 5.8896,
      "mean_token_accuracy": 0.1558088928461075,
      "num_tokens": 35715668.0,
      "step": 17630
    },
    {
      "entropy": 6.206081104278565,
      "epoch": 1.5411168399895132,
      "grad_norm": 2.984375,
      "learning_rate": 0.00047653773231299455,
      "loss": 5.8467,
      "mean_token_accuracy": 0.156816578656435,
      "num_tokens": 35726381.0,
      "step": 17635
    },
    {
      "entropy": 6.152064275741577,
      "epoch": 1.541553788342218,
      "grad_norm": 2.71875,
      "learning_rate": 0.00047652387578141973,
      "loss": 5.869,
      "mean_token_accuracy": 0.1551175206899643,
      "num_tokens": 35736223.0,
      "step": 17640
    },
    {
      "entropy": 6.125501394271851,
      "epoch": 1.5419907366949226,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047651001538451146,
      "loss": 5.8569,
      "mean_token_accuracy": 0.15457146763801574,
      "num_tokens": 35746959.0,
      "step": 17645
    },
    {
      "entropy": 6.16798734664917,
      "epoch": 1.5424276850476275,
      "grad_norm": 3.0625,
      "learning_rate": 0.00047649615112253547,
      "loss": 5.8459,
      "mean_token_accuracy": 0.1611318990588188,
      "num_tokens": 35756975.0,
      "step": 17650
    },
    {
      "entropy": 6.124963617324829,
      "epoch": 1.542864633400332,
      "grad_norm": 5.09375,
      "learning_rate": 0.00047648228299575766,
      "loss": 5.8022,
      "mean_token_accuracy": 0.15940418988466262,
      "num_tokens": 35765890.0,
      "step": 17655
    },
    {
      "entropy": 6.081785678863525,
      "epoch": 1.5433015817530369,
      "grad_norm": 3.375,
      "learning_rate": 0.00047646841100444415,
      "loss": 5.7465,
      "mean_token_accuracy": 0.1595257118344307,
      "num_tokens": 35775630.0,
      "step": 17660
    },
    {
      "entropy": 6.075564384460449,
      "epoch": 1.5437385301057414,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047645453514886095,
      "loss": 5.8937,
      "mean_token_accuracy": 0.15650368481874466,
      "num_tokens": 35786053.0,
      "step": 17665
    },
    {
      "entropy": 6.1906568050384525,
      "epoch": 1.5441754784584463,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004764406554292742,
      "loss": 5.8739,
      "mean_token_accuracy": 0.15723182410001754,
      "num_tokens": 35795736.0,
      "step": 17670
    },
    {
      "entropy": 6.1809087753295895,
      "epoch": 1.544612426811151,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004764267718459501,
      "loss": 5.7909,
      "mean_token_accuracy": 0.15416981875896454,
      "num_tokens": 35805887.0,
      "step": 17675
    },
    {
      "entropy": 6.0569689750671385,
      "epoch": 1.5450493751638557,
      "grad_norm": 2.859375,
      "learning_rate": 0.000476412884399155,
      "loss": 5.8687,
      "mean_token_accuracy": 0.15537567883729936,
      "num_tokens": 35815625.0,
      "step": 17680
    },
    {
      "entropy": 6.149646234512329,
      "epoch": 1.5454863235165603,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004763989930891552,
      "loss": 5.8226,
      "mean_token_accuracy": 0.15840071737766265,
      "num_tokens": 35825617.0,
      "step": 17685
    },
    {
      "entropy": 6.105234050750733,
      "epoch": 1.545923271869265,
      "grad_norm": 2.375,
      "learning_rate": 0.00047638509791621714,
      "loss": 5.8399,
      "mean_token_accuracy": 0.15119030326604843,
      "num_tokens": 35835538.0,
      "step": 17690
    },
    {
      "entropy": 6.166472578048706,
      "epoch": 1.5463602202219697,
      "grad_norm": 3.109375,
      "learning_rate": 0.00047637119888060747,
      "loss": 5.7694,
      "mean_token_accuracy": 0.16132533103227614,
      "num_tokens": 35846302.0,
      "step": 17695
    },
    {
      "entropy": 6.2242824077606205,
      "epoch": 1.5467971685746744,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047635729598259266,
      "loss": 5.9383,
      "mean_token_accuracy": 0.14949096739292145,
      "num_tokens": 35857384.0,
      "step": 17700
    },
    {
      "entropy": 6.209799814224243,
      "epoch": 1.5472341169273793,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004763433892224394,
      "loss": 5.8718,
      "mean_token_accuracy": 0.14422939270734786,
      "num_tokens": 35867496.0,
      "step": 17705
    },
    {
      "entropy": 6.1603435516357425,
      "epoch": 1.5476710652800838,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004763294786004144,
      "loss": 5.8807,
      "mean_token_accuracy": 0.15611384958028793,
      "num_tokens": 35878051.0,
      "step": 17710
    },
    {
      "entropy": 6.131881952285767,
      "epoch": 1.5481080136327887,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004763155641167846,
      "loss": 5.8175,
      "mean_token_accuracy": 0.1604203164577484,
      "num_tokens": 35887926.0,
      "step": 17715
    },
    {
      "entropy": 6.1118607997894285,
      "epoch": 1.5485449619854932,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00047630164577181666,
      "loss": 5.8287,
      "mean_token_accuracy": 0.15149388164281846,
      "num_tokens": 35897982.0,
      "step": 17720
    },
    {
      "entropy": 6.113210725784302,
      "epoch": 1.548981910338198,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004762877235657777,
      "loss": 5.8803,
      "mean_token_accuracy": 0.15546282082796098,
      "num_tokens": 35908805.0,
      "step": 17725
    },
    {
      "entropy": 6.178672790527344,
      "epoch": 1.5494188586909028,
      "grad_norm": 2.25,
      "learning_rate": 0.0004762737974989348,
      "loss": 5.7969,
      "mean_token_accuracy": 0.15982719659805297,
      "num_tokens": 35918739.0,
      "step": 17730
    },
    {
      "entropy": 6.099252510070801,
      "epoch": 1.5498558070436075,
      "grad_norm": 2.546875,
      "learning_rate": 0.000476259867571555,
      "loss": 5.8749,
      "mean_token_accuracy": 0.15605874061584474,
      "num_tokens": 35928592.0,
      "step": 17735
    },
    {
      "entropy": 6.064162015914917,
      "epoch": 1.5502927553963122,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004762459337839055,
      "loss": 5.8197,
      "mean_token_accuracy": 0.15851555913686752,
      "num_tokens": 35938287.0,
      "step": 17740
    },
    {
      "entropy": 6.224677419662475,
      "epoch": 1.5507297037490169,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047623199613625343,
      "loss": 5.8184,
      "mean_token_accuracy": 0.1545197382569313,
      "num_tokens": 35947890.0,
      "step": 17745
    },
    {
      "entropy": 6.047910928726196,
      "epoch": 1.5511666521017216,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004762180546288663,
      "loss": 5.792,
      "mean_token_accuracy": 0.16004120260477067,
      "num_tokens": 35956513.0,
      "step": 17750
    },
    {
      "entropy": 6.07114953994751,
      "epoch": 1.5516036004544262,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004762041092620114,
      "loss": 5.8501,
      "mean_token_accuracy": 0.159542578458786,
      "num_tokens": 35966837.0,
      "step": 17755
    },
    {
      "entropy": 6.207669973373413,
      "epoch": 1.5520405488071312,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047619016003595625,
      "loss": 5.9687,
      "mean_token_accuracy": 0.14954447597265244,
      "num_tokens": 35977076.0,
      "step": 17760
    },
    {
      "entropy": 6.214081525802612,
      "epoch": 1.5524774971598356,
      "grad_norm": 2.953125,
      "learning_rate": 0.00047617620695096854,
      "loss": 5.799,
      "mean_token_accuracy": 0.15590979307889938,
      "num_tokens": 35986862.0,
      "step": 17765
    },
    {
      "entropy": 6.175775575637817,
      "epoch": 1.5529144455125405,
      "grad_norm": 2.65625,
      "learning_rate": 0.00047616225000731566,
      "loss": 5.8807,
      "mean_token_accuracy": 0.14411073699593543,
      "num_tokens": 35997488.0,
      "step": 17770
    },
    {
      "entropy": 6.140735149383545,
      "epoch": 1.553351393865245,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004761482892052655,
      "loss": 5.8541,
      "mean_token_accuracy": 0.14927130043506623,
      "num_tokens": 36008045.0,
      "step": 17775
    },
    {
      "entropy": 6.156619215011597,
      "epoch": 1.55378834221795,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047613432454508565,
      "loss": 5.8232,
      "mean_token_accuracy": 0.15075255781412125,
      "num_tokens": 36017864.0,
      "step": 17780
    },
    {
      "entropy": 6.14031400680542,
      "epoch": 1.5542252905706544,
      "grad_norm": 2.296875,
      "learning_rate": 0.00047612035602704407,
      "loss": 5.8403,
      "mean_token_accuracy": 0.15727190524339676,
      "num_tokens": 36027797.0,
      "step": 17785
    },
    {
      "entropy": 6.1969451904296875,
      "epoch": 1.5546622389233593,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047610638365140867,
      "loss": 5.8663,
      "mean_token_accuracy": 0.1498262345790863,
      "num_tokens": 36037627.0,
      "step": 17790
    },
    {
      "entropy": 6.133720731735229,
      "epoch": 1.555099187276064,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00047609240741844757,
      "loss": 5.8196,
      "mean_token_accuracy": 0.14968007132411004,
      "num_tokens": 36046324.0,
      "step": 17795
    },
    {
      "entropy": 6.094100475311279,
      "epoch": 1.5555361356287687,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004760784273284286,
      "loss": 5.7899,
      "mean_token_accuracy": 0.16053585708141327,
      "num_tokens": 36056212.0,
      "step": 17800
    },
    {
      "entropy": 6.148568534851075,
      "epoch": 1.5559730839814734,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004760644433816201,
      "loss": 5.9402,
      "mean_token_accuracy": 0.1476304531097412,
      "num_tokens": 36066745.0,
      "step": 17805
    },
    {
      "entropy": 6.216650438308716,
      "epoch": 1.556410032334178,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004760504555782902,
      "loss": 5.846,
      "mean_token_accuracy": 0.1519563153386116,
      "num_tokens": 36077415.0,
      "step": 17810
    },
    {
      "entropy": 6.103620147705078,
      "epoch": 1.5568469806868828,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047603646391870715,
      "loss": 5.8082,
      "mean_token_accuracy": 0.16240674257278442,
      "num_tokens": 36088381.0,
      "step": 17815
    },
    {
      "entropy": 6.0874169826507565,
      "epoch": 1.5572839290395875,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047602246840313945,
      "loss": 5.7962,
      "mean_token_accuracy": 0.1603674679994583,
      "num_tokens": 36098333.0,
      "step": 17820
    },
    {
      "entropy": 6.064192008972168,
      "epoch": 1.5577208773922924,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00047600846903185544,
      "loss": 5.8682,
      "mean_token_accuracy": 0.153440423309803,
      "num_tokens": 36107843.0,
      "step": 17825
    },
    {
      "entropy": 6.192847871780396,
      "epoch": 1.5581578257449968,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004759944658051237,
      "loss": 5.843,
      "mean_token_accuracy": 0.1583794519305229,
      "num_tokens": 36116948.0,
      "step": 17830
    },
    {
      "entropy": 6.153967332839966,
      "epoch": 1.5585947740977018,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004759804587232127,
      "loss": 5.8667,
      "mean_token_accuracy": 0.1532774269580841,
      "num_tokens": 36126808.0,
      "step": 17835
    },
    {
      "entropy": 6.127297449111938,
      "epoch": 1.5590317224504062,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047596644778639133,
      "loss": 5.9271,
      "mean_token_accuracy": 0.14905515313148499,
      "num_tokens": 36136630.0,
      "step": 17840
    },
    {
      "entropy": 6.2333721160888675,
      "epoch": 1.5594686708031111,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004759524329949281,
      "loss": 5.8645,
      "mean_token_accuracy": 0.14976508766412736,
      "num_tokens": 36146422.0,
      "step": 17845
    },
    {
      "entropy": 6.185214185714722,
      "epoch": 1.5599056191558158,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004759384143490919,
      "loss": 5.8501,
      "mean_token_accuracy": 0.14938322603702545,
      "num_tokens": 36156169.0,
      "step": 17850
    },
    {
      "entropy": 6.116262340545655,
      "epoch": 1.5603425675085205,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004759243918491517,
      "loss": 5.8568,
      "mean_token_accuracy": 0.1594856485724449,
      "num_tokens": 36165173.0,
      "step": 17855
    },
    {
      "entropy": 6.173662376403809,
      "epoch": 1.5607795158612252,
      "grad_norm": 1.828125,
      "learning_rate": 0.00047591036549537633,
      "loss": 5.8907,
      "mean_token_accuracy": 0.15109619051218032,
      "num_tokens": 36175852.0,
      "step": 17860
    },
    {
      "entropy": 6.13984637260437,
      "epoch": 1.56121646421393,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004758963352880348,
      "loss": 5.8446,
      "mean_token_accuracy": 0.15417944490909577,
      "num_tokens": 36185288.0,
      "step": 17865
    },
    {
      "entropy": 6.183798027038574,
      "epoch": 1.5616534125666346,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004758823012273964,
      "loss": 5.8696,
      "mean_token_accuracy": 0.15622807145118714,
      "num_tokens": 36196374.0,
      "step": 17870
    },
    {
      "entropy": 6.1149355411529545,
      "epoch": 1.5620903609193393,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047586826331373025,
      "loss": 5.9002,
      "mean_token_accuracy": 0.15695322304964066,
      "num_tokens": 36206714.0,
      "step": 17875
    },
    {
      "entropy": 6.19454517364502,
      "epoch": 1.5625273092720442,
      "grad_norm": 2.0,
      "learning_rate": 0.0004758542215473054,
      "loss": 5.9436,
      "mean_token_accuracy": 0.14614250063896178,
      "num_tokens": 36217891.0,
      "step": 17880
    },
    {
      "entropy": 6.176143741607666,
      "epoch": 1.5629642576247487,
      "grad_norm": 2.375,
      "learning_rate": 0.0004758401759283915,
      "loss": 5.8092,
      "mean_token_accuracy": 0.15781117528676986,
      "num_tokens": 36226776.0,
      "step": 17885
    },
    {
      "entropy": 6.14338641166687,
      "epoch": 1.5634012059774536,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047582612645725764,
      "loss": 5.8942,
      "mean_token_accuracy": 0.15400125086307526,
      "num_tokens": 36237362.0,
      "step": 17890
    },
    {
      "entropy": 6.109305906295776,
      "epoch": 1.563838154330158,
      "grad_norm": 2.0,
      "learning_rate": 0.00047581207313417346,
      "loss": 5.7884,
      "mean_token_accuracy": 0.15746268332004548,
      "num_tokens": 36246264.0,
      "step": 17895
    },
    {
      "entropy": 6.191521739959716,
      "epoch": 1.564275102682863,
      "grad_norm": 1.75,
      "learning_rate": 0.00047579801595940853,
      "loss": 5.8686,
      "mean_token_accuracy": 0.1532348468899727,
      "num_tokens": 36255676.0,
      "step": 17900
    },
    {
      "entropy": 6.137023687362671,
      "epoch": 1.5647120510355674,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047578395493323245,
      "loss": 5.729,
      "mean_token_accuracy": 0.16360267698764802,
      "num_tokens": 36264346.0,
      "step": 17905
    },
    {
      "entropy": 6.053152847290039,
      "epoch": 1.5651489993882723,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047576989005591495,
      "loss": 5.8377,
      "mean_token_accuracy": 0.14955184161663054,
      "num_tokens": 36273569.0,
      "step": 17910
    },
    {
      "entropy": 6.127921628952026,
      "epoch": 1.565585947740977,
      "grad_norm": 2.125,
      "learning_rate": 0.0004757558213277257,
      "loss": 5.7648,
      "mean_token_accuracy": 0.16324147805571557,
      "num_tokens": 36283708.0,
      "step": 17915
    },
    {
      "entropy": 6.1554240703582765,
      "epoch": 1.5660228960936817,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047574174874893456,
      "loss": 5.798,
      "mean_token_accuracy": 0.15344853103160858,
      "num_tokens": 36293184.0,
      "step": 17920
    },
    {
      "entropy": 6.090034246444702,
      "epoch": 1.5664598444463864,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004757276723198115,
      "loss": 5.8923,
      "mean_token_accuracy": 0.15130351930856706,
      "num_tokens": 36303071.0,
      "step": 17925
    },
    {
      "entropy": 6.147397661209107,
      "epoch": 1.566896792799091,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004757135920406266,
      "loss": 5.8016,
      "mean_token_accuracy": 0.15910038352012634,
      "num_tokens": 36313245.0,
      "step": 17930
    },
    {
      "entropy": 6.078237390518188,
      "epoch": 1.567333741151796,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004756995079116498,
      "loss": 5.8105,
      "mean_token_accuracy": 0.15247970968484878,
      "num_tokens": 36322711.0,
      "step": 17935
    },
    {
      "entropy": 6.0533184051513675,
      "epoch": 1.5677706895045005,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047568541993315124,
      "loss": 5.8381,
      "mean_token_accuracy": 0.15331623703241348,
      "num_tokens": 36333667.0,
      "step": 17940
    },
    {
      "entropy": 6.210267353057861,
      "epoch": 1.5682076378572054,
      "grad_norm": 1.75,
      "learning_rate": 0.00047567132810540113,
      "loss": 5.9076,
      "mean_token_accuracy": 0.14778622090816498,
      "num_tokens": 36343660.0,
      "step": 17945
    },
    {
      "entropy": 6.142909240722656,
      "epoch": 1.5686445862099099,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047565723242866983,
      "loss": 5.9106,
      "mean_token_accuracy": 0.15254026874899865,
      "num_tokens": 36353581.0,
      "step": 17950
    },
    {
      "entropy": 6.138402795791626,
      "epoch": 1.5690815345626148,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047564313290322774,
      "loss": 5.8026,
      "mean_token_accuracy": 0.15257705599069596,
      "num_tokens": 36363818.0,
      "step": 17955
    },
    {
      "entropy": 6.135894870758056,
      "epoch": 1.5695184829153193,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004756290295293451,
      "loss": 5.8455,
      "mean_token_accuracy": 0.15520919114351273,
      "num_tokens": 36373590.0,
      "step": 17960
    },
    {
      "entropy": 6.1613812923431395,
      "epoch": 1.5699554312680242,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004756149223072926,
      "loss": 5.8904,
      "mean_token_accuracy": 0.15242792218923568,
      "num_tokens": 36383556.0,
      "step": 17965
    },
    {
      "entropy": 6.152489709854126,
      "epoch": 1.5703923796207289,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004756008112373408,
      "loss": 5.8332,
      "mean_token_accuracy": 0.15248623639345169,
      "num_tokens": 36393194.0,
      "step": 17970
    },
    {
      "entropy": 6.120516300201416,
      "epoch": 1.5708293279734336,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004755866963197603,
      "loss": 5.8233,
      "mean_token_accuracy": 0.15515899062156677,
      "num_tokens": 36402962.0,
      "step": 17975
    },
    {
      "entropy": 6.1442992210388185,
      "epoch": 1.5712662763261382,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004755725775548219,
      "loss": 5.819,
      "mean_token_accuracy": 0.15354258567094803,
      "num_tokens": 36411932.0,
      "step": 17980
    },
    {
      "entropy": 6.1702721118927,
      "epoch": 1.571703224678843,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00047555845494279633,
      "loss": 5.8074,
      "mean_token_accuracy": 0.16097897440195083,
      "num_tokens": 36421363.0,
      "step": 17985
    },
    {
      "entropy": 6.132246685028076,
      "epoch": 1.5721401730315476,
      "grad_norm": 2.0,
      "learning_rate": 0.0004755443284839545,
      "loss": 5.8221,
      "mean_token_accuracy": 0.1544548377394676,
      "num_tokens": 36431422.0,
      "step": 17990
    },
    {
      "entropy": 6.075223112106324,
      "epoch": 1.5725771213842523,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047553019817856736,
      "loss": 5.813,
      "mean_token_accuracy": 0.15758510082960128,
      "num_tokens": 36441004.0,
      "step": 17995
    },
    {
      "entropy": 6.225890064239502,
      "epoch": 1.5730140697369572,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00047551606402690595,
      "loss": 5.9213,
      "mean_token_accuracy": 0.1466245472431183,
      "num_tokens": 36451474.0,
      "step": 18000
    },
    {
      "epoch": 1.5730140697369572,
      "eval_entropy": 5.992097050988548,
      "eval_loss": 5.897541522979736,
      "eval_mean_token_accuracy": 0.16076978383316498,
      "eval_num_tokens": 36451474.0,
      "eval_runtime": 28.2174,
      "eval_samples_per_second": 1311.744,
      "eval_steps_per_second": 163.977,
      "step": 18000
    },
    {
      "entropy": 6.158188962936402,
      "epoch": 1.5734510180896617,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004755019260292414,
      "loss": 5.8016,
      "mean_token_accuracy": 0.155093415081501,
      "num_tokens": 36461341.0,
      "step": 18005
    },
    {
      "entropy": 6.103657579421997,
      "epoch": 1.5738879664423666,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004754877841858448,
      "loss": 5.928,
      "mean_token_accuracy": 0.15227026119828224,
      "num_tokens": 36471679.0,
      "step": 18010
    },
    {
      "entropy": 6.188374614715576,
      "epoch": 1.574324914795071,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004754736384969875,
      "loss": 5.8603,
      "mean_token_accuracy": 0.1491597056388855,
      "num_tokens": 36481457.0,
      "step": 18015
    },
    {
      "entropy": 6.115581274032593,
      "epoch": 1.574761863147776,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004754594889629407,
      "loss": 5.7981,
      "mean_token_accuracy": 0.15590354502201081,
      "num_tokens": 36490596.0,
      "step": 18020
    },
    {
      "entropy": 6.189972972869873,
      "epoch": 1.5751988115004807,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004754453355839759,
      "loss": 5.907,
      "mean_token_accuracy": 0.1497448593378067,
      "num_tokens": 36501027.0,
      "step": 18025
    },
    {
      "entropy": 6.140265607833863,
      "epoch": 1.5756357598531854,
      "grad_norm": 1.71875,
      "learning_rate": 0.00047543117836036446,
      "loss": 5.8032,
      "mean_token_accuracy": 0.15489476919174194,
      "num_tokens": 36512089.0,
      "step": 18030
    },
    {
      "entropy": 6.076355457305908,
      "epoch": 1.57607270820589,
      "grad_norm": 1.703125,
      "learning_rate": 0.00047541701729237815,
      "loss": 5.822,
      "mean_token_accuracy": 0.15851605236530303,
      "num_tokens": 36522521.0,
      "step": 18035
    },
    {
      "entropy": 6.234769821166992,
      "epoch": 1.5765096565585948,
      "grad_norm": 2.125,
      "learning_rate": 0.0004754028523802883,
      "loss": 5.9214,
      "mean_token_accuracy": 0.14585841074585915,
      "num_tokens": 36533111.0,
      "step": 18040
    },
    {
      "entropy": 6.116734504699707,
      "epoch": 1.5769466049112995,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004753886836243668,
      "loss": 5.8059,
      "mean_token_accuracy": 0.1609452873468399,
      "num_tokens": 36542477.0,
      "step": 18045
    },
    {
      "entropy": 6.050258493423462,
      "epoch": 1.5773835532640041,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047537451102488526,
      "loss": 5.8604,
      "mean_token_accuracy": 0.15373194292187692,
      "num_tokens": 36552137.0,
      "step": 18050
    },
    {
      "entropy": 6.17083477973938,
      "epoch": 1.577820501616709,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004753603345821157,
      "loss": 5.874,
      "mean_token_accuracy": 0.15040013045072556,
      "num_tokens": 36562440.0,
      "step": 18055
    },
    {
      "entropy": 6.1897135257720945,
      "epoch": 1.5782574499694135,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047534615429632986,
      "loss": 5.814,
      "mean_token_accuracy": 0.15815496444702148,
      "num_tokens": 36571863.0,
      "step": 18060
    },
    {
      "entropy": 6.081523466110229,
      "epoch": 1.5786943983221184,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004753319701677999,
      "loss": 5.8188,
      "mean_token_accuracy": 0.1590537667274475,
      "num_tokens": 36581409.0,
      "step": 18065
    },
    {
      "entropy": 6.135543918609619,
      "epoch": 1.579131346674823,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047531778219679773,
      "loss": 5.8387,
      "mean_token_accuracy": 0.1500946596264839,
      "num_tokens": 36592779.0,
      "step": 18070
    },
    {
      "entropy": 6.099080991744995,
      "epoch": 1.5795682950275278,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047530359038359547,
      "loss": 5.8261,
      "mean_token_accuracy": 0.16549460887908934,
      "num_tokens": 36601173.0,
      "step": 18075
    },
    {
      "entropy": 6.092098569869995,
      "epoch": 1.5800052433802323,
      "grad_norm": 2.0,
      "learning_rate": 0.0004752893947284654,
      "loss": 5.8677,
      "mean_token_accuracy": 0.15613431632518768,
      "num_tokens": 36610147.0,
      "step": 18080
    },
    {
      "entropy": 6.12001142501831,
      "epoch": 1.5804421917329372,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004752751952316799,
      "loss": 5.8327,
      "mean_token_accuracy": 0.15582821518182755,
      "num_tokens": 36620093.0,
      "step": 18085
    },
    {
      "entropy": 6.163449096679687,
      "epoch": 1.580879140085642,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047526099189351115,
      "loss": 5.869,
      "mean_token_accuracy": 0.15226275324821473,
      "num_tokens": 36631833.0,
      "step": 18090
    },
    {
      "entropy": 6.121299171447754,
      "epoch": 1.5813160884383466,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004752467847142316,
      "loss": 5.799,
      "mean_token_accuracy": 0.15259799361228943,
      "num_tokens": 36641123.0,
      "step": 18095
    },
    {
      "entropy": 6.122568988800049,
      "epoch": 1.5817530367910513,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047523257369411383,
      "loss": 5.8833,
      "mean_token_accuracy": 0.15033481940627097,
      "num_tokens": 36652310.0,
      "step": 18100
    },
    {
      "entropy": 6.199804639816284,
      "epoch": 1.582189985143756,
      "grad_norm": 2.375,
      "learning_rate": 0.0004752183588334303,
      "loss": 5.9005,
      "mean_token_accuracy": 0.152802512049675,
      "num_tokens": 36662783.0,
      "step": 18105
    },
    {
      "entropy": 6.155228853225708,
      "epoch": 1.5826269334964607,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047520414013245383,
      "loss": 5.822,
      "mean_token_accuracy": 0.1583334967494011,
      "num_tokens": 36672933.0,
      "step": 18110
    },
    {
      "entropy": 6.1429215431213375,
      "epoch": 1.5830638818491654,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004751899175914571,
      "loss": 5.8927,
      "mean_token_accuracy": 0.15098801255226135,
      "num_tokens": 36682570.0,
      "step": 18115
    },
    {
      "entropy": 6.102208280563355,
      "epoch": 1.5835008302018703,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004751756912107127,
      "loss": 5.8016,
      "mean_token_accuracy": 0.16016732454299926,
      "num_tokens": 36693008.0,
      "step": 18120
    },
    {
      "entropy": 6.047227907180786,
      "epoch": 1.5839377785545747,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047516146099049376,
      "loss": 5.7122,
      "mean_token_accuracy": 0.16307833045721054,
      "num_tokens": 36702611.0,
      "step": 18125
    },
    {
      "entropy": 6.099211072921753,
      "epoch": 1.5843747269072797,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047514722693107304,
      "loss": 5.8812,
      "mean_token_accuracy": 0.15601153075695037,
      "num_tokens": 36712470.0,
      "step": 18130
    },
    {
      "entropy": 6.157591247558594,
      "epoch": 1.5848116752599841,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004751329890327237,
      "loss": 5.8888,
      "mean_token_accuracy": 0.15223745182156562,
      "num_tokens": 36722578.0,
      "step": 18135
    },
    {
      "entropy": 6.162668514251709,
      "epoch": 1.585248623612689,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004751187472957187,
      "loss": 5.8746,
      "mean_token_accuracy": 0.15409443974494935,
      "num_tokens": 36732334.0,
      "step": 18140
    },
    {
      "entropy": 6.123752069473267,
      "epoch": 1.5856855719653937,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004751045017203313,
      "loss": 5.8885,
      "mean_token_accuracy": 0.15324715673923492,
      "num_tokens": 36742328.0,
      "step": 18145
    },
    {
      "entropy": 6.141130352020264,
      "epoch": 1.5861225203180984,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004750902523068347,
      "loss": 5.8926,
      "mean_token_accuracy": 0.15117437094449998,
      "num_tokens": 36754363.0,
      "step": 18150
    },
    {
      "entropy": 6.151986789703369,
      "epoch": 1.586559468670803,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004750759990555022,
      "loss": 5.8848,
      "mean_token_accuracy": 0.14788235649466513,
      "num_tokens": 36765588.0,
      "step": 18155
    },
    {
      "entropy": 6.08484959602356,
      "epoch": 1.5869964170235078,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004750617419666072,
      "loss": 5.7995,
      "mean_token_accuracy": 0.15464927703142167,
      "num_tokens": 36774851.0,
      "step": 18160
    },
    {
      "entropy": 6.132309865951538,
      "epoch": 1.5874333653762125,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047504748104042324,
      "loss": 5.8627,
      "mean_token_accuracy": 0.15192834734916688,
      "num_tokens": 36785836.0,
      "step": 18165
    },
    {
      "entropy": 6.061438465118409,
      "epoch": 1.5878703137289172,
      "grad_norm": 2.71875,
      "learning_rate": 0.00047503321627722375,
      "loss": 5.797,
      "mean_token_accuracy": 0.16071885600686073,
      "num_tokens": 36795367.0,
      "step": 18170
    },
    {
      "entropy": 6.141823673248291,
      "epoch": 1.588307262081622,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047501894767728225,
      "loss": 5.7842,
      "mean_token_accuracy": 0.1577644795179367,
      "num_tokens": 36804294.0,
      "step": 18175
    },
    {
      "entropy": 6.135769701004028,
      "epoch": 1.5887442104343266,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00047500467524087266,
      "loss": 5.8456,
      "mean_token_accuracy": 0.1532744884490967,
      "num_tokens": 36813711.0,
      "step": 18180
    },
    {
      "entropy": 6.156258630752563,
      "epoch": 1.5891811587870315,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004749903989682685,
      "loss": 5.8345,
      "mean_token_accuracy": 0.1577165961265564,
      "num_tokens": 36824498.0,
      "step": 18185
    },
    {
      "entropy": 6.1636247634887695,
      "epoch": 1.589618107139736,
      "grad_norm": 1.671875,
      "learning_rate": 0.00047497611885974366,
      "loss": 5.7983,
      "mean_token_accuracy": 0.16096200197935104,
      "num_tokens": 36833329.0,
      "step": 18190
    },
    {
      "entropy": 6.105648612976074,
      "epoch": 1.5900550554924409,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047496183491557215,
      "loss": 5.854,
      "mean_token_accuracy": 0.1531764507293701,
      "num_tokens": 36842783.0,
      "step": 18195
    },
    {
      "entropy": 6.066834402084351,
      "epoch": 1.5904920038451456,
      "grad_norm": 1.875,
      "learning_rate": 0.00047494754713602787,
      "loss": 5.8337,
      "mean_token_accuracy": 0.15330330207943915,
      "num_tokens": 36852785.0,
      "step": 18200
    },
    {
      "entropy": 6.11262698173523,
      "epoch": 1.5909289521978502,
      "grad_norm": 2.0,
      "learning_rate": 0.00047493325552138484,
      "loss": 5.8639,
      "mean_token_accuracy": 0.1529523327946663,
      "num_tokens": 36862337.0,
      "step": 18205
    },
    {
      "entropy": 6.165375089645385,
      "epoch": 1.591365900550555,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004749189600719172,
      "loss": 5.8814,
      "mean_token_accuracy": 0.15142595618963242,
      "num_tokens": 36872026.0,
      "step": 18210
    },
    {
      "entropy": 6.114049291610717,
      "epoch": 1.5918028489032596,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004749046607878992,
      "loss": 5.8663,
      "mean_token_accuracy": 0.15473776459693908,
      "num_tokens": 36882520.0,
      "step": 18215
    },
    {
      "entropy": 6.075957012176514,
      "epoch": 1.5922397972559643,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047489035766960504,
      "loss": 5.7429,
      "mean_token_accuracy": 0.16983606517314911,
      "num_tokens": 36892630.0,
      "step": 18220
    },
    {
      "entropy": 6.19117112159729,
      "epoch": 1.592676745608669,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047487605071730904,
      "loss": 5.8687,
      "mean_token_accuracy": 0.15230597406625748,
      "num_tokens": 36902692.0,
      "step": 18225
    },
    {
      "entropy": 6.2098315238952635,
      "epoch": 1.593113693961374,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047486173993128566,
      "loss": 5.974,
      "mean_token_accuracy": 0.14443593248724937,
      "num_tokens": 36913581.0,
      "step": 18230
    },
    {
      "entropy": 6.11932864189148,
      "epoch": 1.5935506423140784,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004748474253118094,
      "loss": 5.8162,
      "mean_token_accuracy": 0.15359503030776978,
      "num_tokens": 36924031.0,
      "step": 18235
    },
    {
      "entropy": 6.147268056869507,
      "epoch": 1.5939875906667833,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004748331068591548,
      "loss": 5.8445,
      "mean_token_accuracy": 0.1566885754466057,
      "num_tokens": 36934622.0,
      "step": 18240
    },
    {
      "entropy": 6.1849205493927,
      "epoch": 1.5944245390194878,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004748187845735965,
      "loss": 5.8724,
      "mean_token_accuracy": 0.158174766600132,
      "num_tokens": 36944374.0,
      "step": 18245
    },
    {
      "entropy": 6.136397647857666,
      "epoch": 1.5948614873721927,
      "grad_norm": 2.0,
      "learning_rate": 0.00047480445845540927,
      "loss": 5.8145,
      "mean_token_accuracy": 0.15433474332094194,
      "num_tokens": 36954371.0,
      "step": 18250
    },
    {
      "entropy": 6.053894710540772,
      "epoch": 1.5952984357248972,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00047479012850486777,
      "loss": 5.8279,
      "mean_token_accuracy": 0.1602454349398613,
      "num_tokens": 36964217.0,
      "step": 18255
    },
    {
      "entropy": 6.2213554859161375,
      "epoch": 1.595735384077602,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047477579472224687,
      "loss": 5.9019,
      "mean_token_accuracy": 0.150271500647068,
      "num_tokens": 36974580.0,
      "step": 18260
    },
    {
      "entropy": 6.0925116539001465,
      "epoch": 1.5961723324303068,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004747614571078216,
      "loss": 5.6725,
      "mean_token_accuracy": 0.17123343348503112,
      "num_tokens": 36983434.0,
      "step": 18265
    },
    {
      "entropy": 6.057025814056397,
      "epoch": 1.5966092807830115,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004747471156618669,
      "loss": 5.9073,
      "mean_token_accuracy": 0.1516721248626709,
      "num_tokens": 36993813.0,
      "step": 18270
    },
    {
      "entropy": 6.1624181270599365,
      "epoch": 1.5970462291357161,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004747327703846579,
      "loss": 5.8415,
      "mean_token_accuracy": 0.15800213813781738,
      "num_tokens": 37003855.0,
      "step": 18275
    },
    {
      "entropy": 6.102229309082031,
      "epoch": 1.5974831774884208,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047471842127646966,
      "loss": 5.8066,
      "mean_token_accuracy": 0.16011677086353301,
      "num_tokens": 37013660.0,
      "step": 18280
    },
    {
      "entropy": 6.042488145828247,
      "epoch": 1.5979201258411255,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047470406833757746,
      "loss": 5.7613,
      "mean_token_accuracy": 0.1555236518383026,
      "num_tokens": 37023888.0,
      "step": 18285
    },
    {
      "entropy": 6.171698427200317,
      "epoch": 1.5983570741938302,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047468971156825656,
      "loss": 5.9277,
      "mean_token_accuracy": 0.14409274607896805,
      "num_tokens": 37034260.0,
      "step": 18290
    },
    {
      "entropy": 6.1402528285980225,
      "epoch": 1.5987940225465351,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004746753509687824,
      "loss": 5.8863,
      "mean_token_accuracy": 0.15345453470945358,
      "num_tokens": 37045300.0,
      "step": 18295
    },
    {
      "entropy": 6.163219738006592,
      "epoch": 1.5992309708992396,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004746609865394303,
      "loss": 5.7962,
      "mean_token_accuracy": 0.16205774247646332,
      "num_tokens": 37054898.0,
      "step": 18300
    },
    {
      "entropy": 6.053664112091065,
      "epoch": 1.5996679192519445,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004746466182804759,
      "loss": 5.7697,
      "mean_token_accuracy": 0.156445737183094,
      "num_tokens": 37064519.0,
      "step": 18305
    },
    {
      "entropy": 6.072087907791138,
      "epoch": 1.600104867604649,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004746322461921948,
      "loss": 5.8531,
      "mean_token_accuracy": 0.15501998960971833,
      "num_tokens": 37074967.0,
      "step": 18310
    },
    {
      "entropy": 6.115574693679809,
      "epoch": 1.600541815957354,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004746178702748626,
      "loss": 5.8371,
      "mean_token_accuracy": 0.1587224781513214,
      "num_tokens": 37084736.0,
      "step": 18315
    },
    {
      "entropy": 6.170813703536988,
      "epoch": 1.6009787643100586,
      "grad_norm": 2.140625,
      "learning_rate": 0.000474603490528755,
      "loss": 5.8044,
      "mean_token_accuracy": 0.16121101081371308,
      "num_tokens": 37094327.0,
      "step": 18320
    },
    {
      "entropy": 6.140086221694946,
      "epoch": 1.6014157126627633,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004745891069541479,
      "loss": 5.819,
      "mean_token_accuracy": 0.15752982199192048,
      "num_tokens": 37104099.0,
      "step": 18325
    },
    {
      "entropy": 6.114892101287841,
      "epoch": 1.601852661015468,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047457471955131714,
      "loss": 5.8892,
      "mean_token_accuracy": 0.14979428946971893,
      "num_tokens": 37115337.0,
      "step": 18330
    },
    {
      "entropy": 6.170300102233886,
      "epoch": 1.6022896093681727,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004745603283205387,
      "loss": 5.8199,
      "mean_token_accuracy": 0.15776073485612868,
      "num_tokens": 37125522.0,
      "step": 18335
    },
    {
      "entropy": 6.154092025756836,
      "epoch": 1.6027265577208774,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004745459332620885,
      "loss": 5.9082,
      "mean_token_accuracy": 0.15259279161691666,
      "num_tokens": 37135847.0,
      "step": 18340
    },
    {
      "entropy": 6.01115345954895,
      "epoch": 1.603163506073582,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004745315343762428,
      "loss": 5.7863,
      "mean_token_accuracy": 0.16372239738702773,
      "num_tokens": 37145714.0,
      "step": 18345
    },
    {
      "entropy": 6.101547622680664,
      "epoch": 1.603600454426287,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047451713166327776,
      "loss": 5.8202,
      "mean_token_accuracy": 0.15872190445661544,
      "num_tokens": 37154921.0,
      "step": 18350
    },
    {
      "entropy": 6.17178316116333,
      "epoch": 1.6040374027789914,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047450272512346953,
      "loss": 5.8216,
      "mean_token_accuracy": 0.15789951384067535,
      "num_tokens": 37165303.0,
      "step": 18355
    },
    {
      "entropy": 6.102764511108399,
      "epoch": 1.6044743511316963,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047448831475709454,
      "loss": 5.7883,
      "mean_token_accuracy": 0.15744060277938843,
      "num_tokens": 37175014.0,
      "step": 18360
    },
    {
      "entropy": 6.079820156097412,
      "epoch": 1.6049112994844008,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004744739005644291,
      "loss": 5.8088,
      "mean_token_accuracy": 0.16320455074310303,
      "num_tokens": 37185830.0,
      "step": 18365
    },
    {
      "entropy": 6.082566165924073,
      "epoch": 1.6053482478371057,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004744594825457498,
      "loss": 5.736,
      "mean_token_accuracy": 0.16083633452653884,
      "num_tokens": 37196264.0,
      "step": 18370
    },
    {
      "entropy": 6.16067361831665,
      "epoch": 1.6057851961898104,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047444506070133305,
      "loss": 5.8702,
      "mean_token_accuracy": 0.15172916948795317,
      "num_tokens": 37206338.0,
      "step": 18375
    },
    {
      "entropy": 6.133966636657715,
      "epoch": 1.606222144542515,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004744306350314555,
      "loss": 5.834,
      "mean_token_accuracy": 0.15540405511856079,
      "num_tokens": 37216231.0,
      "step": 18380
    },
    {
      "entropy": 6.255458927154541,
      "epoch": 1.6066590928952198,
      "grad_norm": 1.625,
      "learning_rate": 0.0004744162055363939,
      "loss": 5.9554,
      "mean_token_accuracy": 0.14620960950851442,
      "num_tokens": 37227209.0,
      "step": 18385
    },
    {
      "entropy": 6.1480175971984865,
      "epoch": 1.6070960412479245,
      "grad_norm": 2.765625,
      "learning_rate": 0.000474401772216425,
      "loss": 5.9332,
      "mean_token_accuracy": 0.1554884910583496,
      "num_tokens": 37238056.0,
      "step": 18390
    },
    {
      "entropy": 6.062219905853271,
      "epoch": 1.6075329896006292,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004743873350718257,
      "loss": 5.8097,
      "mean_token_accuracy": 0.15636790841817855,
      "num_tokens": 37247864.0,
      "step": 18395
    },
    {
      "entropy": 6.048687553405761,
      "epoch": 1.6079699379533339,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004743728941028727,
      "loss": 5.7532,
      "mean_token_accuracy": 0.16187262088060378,
      "num_tokens": 37256418.0,
      "step": 18400
    },
    {
      "entropy": 6.119430303573608,
      "epoch": 1.6084068863060388,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004743584493098431,
      "loss": 5.8693,
      "mean_token_accuracy": 0.15558625459671022,
      "num_tokens": 37266540.0,
      "step": 18405
    },
    {
      "entropy": 6.188225746154785,
      "epoch": 1.6088438346587433,
      "grad_norm": 2.578125,
      "learning_rate": 0.00047434400069301414,
      "loss": 5.8421,
      "mean_token_accuracy": 0.1538287803530693,
      "num_tokens": 37277011.0,
      "step": 18410
    },
    {
      "entropy": 6.105555009841919,
      "epoch": 1.6092807830114482,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004743295482526627,
      "loss": 5.8359,
      "mean_token_accuracy": 0.157236485183239,
      "num_tokens": 37287203.0,
      "step": 18415
    },
    {
      "entropy": 6.083509492874145,
      "epoch": 1.6097177313641526,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004743150919890661,
      "loss": 5.7794,
      "mean_token_accuracy": 0.1586715191602707,
      "num_tokens": 37297293.0,
      "step": 18420
    },
    {
      "entropy": 6.166322374343872,
      "epoch": 1.6101546797168576,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00047430063190250155,
      "loss": 5.8815,
      "mean_token_accuracy": 0.1494002342224121,
      "num_tokens": 37307882.0,
      "step": 18425
    },
    {
      "entropy": 6.082572603225708,
      "epoch": 1.610591628069562,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004742861679932464,
      "loss": 5.7658,
      "mean_token_accuracy": 0.16378538608551024,
      "num_tokens": 37317466.0,
      "step": 18430
    },
    {
      "entropy": 6.054345893859863,
      "epoch": 1.611028576422267,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047427170026157823,
      "loss": 5.8851,
      "mean_token_accuracy": 0.15313297808170317,
      "num_tokens": 37327544.0,
      "step": 18435
    },
    {
      "entropy": 6.156347465515137,
      "epoch": 1.6114655247749716,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004742572287077744,
      "loss": 5.874,
      "mean_token_accuracy": 0.15231260806322097,
      "num_tokens": 37338082.0,
      "step": 18440
    },
    {
      "entropy": 6.187852287292481,
      "epoch": 1.6119024731276763,
      "grad_norm": 1.703125,
      "learning_rate": 0.00047424275333211256,
      "loss": 5.8248,
      "mean_token_accuracy": 0.15846317410469055,
      "num_tokens": 37348127.0,
      "step": 18445
    },
    {
      "entropy": 6.189578294754028,
      "epoch": 1.612339421480381,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004742282741348702,
      "loss": 5.9039,
      "mean_token_accuracy": 0.1473035603761673,
      "num_tokens": 37359547.0,
      "step": 18450
    },
    {
      "entropy": 6.147730827331543,
      "epoch": 1.6127763698330857,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004742137911163253,
      "loss": 5.9371,
      "mean_token_accuracy": 0.15470682680606843,
      "num_tokens": 37369547.0,
      "step": 18455
    },
    {
      "entropy": 6.110079622268676,
      "epoch": 1.6132133181857904,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004741993042767554,
      "loss": 5.9168,
      "mean_token_accuracy": 0.157265368103981,
      "num_tokens": 37380055.0,
      "step": 18460
    },
    {
      "entropy": 6.183764362335205,
      "epoch": 1.613650266538495,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00047418481361643844,
      "loss": 5.9211,
      "mean_token_accuracy": 0.15204641371965408,
      "num_tokens": 37390179.0,
      "step": 18465
    },
    {
      "entropy": 6.236109256744385,
      "epoch": 1.6140872148912,
      "grad_norm": 2.5,
      "learning_rate": 0.00047417031913565247,
      "loss": 5.879,
      "mean_token_accuracy": 0.15373657792806625,
      "num_tokens": 37400663.0,
      "step": 18470
    },
    {
      "entropy": 6.129943609237671,
      "epoch": 1.6145241632439045,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004741558208346753,
      "loss": 5.8659,
      "mean_token_accuracy": 0.1560629978775978,
      "num_tokens": 37410988.0,
      "step": 18475
    },
    {
      "entropy": 6.199072647094726,
      "epoch": 1.6149611115966094,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047414131871378526,
      "loss": 5.9526,
      "mean_token_accuracy": 0.15012071579694747,
      "num_tokens": 37422092.0,
      "step": 18480
    },
    {
      "entropy": 6.131946229934693,
      "epoch": 1.6153980599493138,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004741268127732603,
      "loss": 5.835,
      "mean_token_accuracy": 0.15567414164543153,
      "num_tokens": 37431637.0,
      "step": 18485
    },
    {
      "entropy": 6.142140293121338,
      "epoch": 1.6158350083020188,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047411230301337877,
      "loss": 5.9197,
      "mean_token_accuracy": 0.14138541519641876,
      "num_tokens": 37442678.0,
      "step": 18490
    },
    {
      "entropy": 6.176050090789795,
      "epoch": 1.6162719566547235,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004740977894344189,
      "loss": 5.8537,
      "mean_token_accuracy": 0.15359829142689704,
      "num_tokens": 37452324.0,
      "step": 18495
    },
    {
      "entropy": 6.142911100387574,
      "epoch": 1.6167089050074281,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004740832720366591,
      "loss": 5.8174,
      "mean_token_accuracy": 0.1494448497891426,
      "num_tokens": 37462940.0,
      "step": 18500
    },
    {
      "entropy": 6.132896661758423,
      "epoch": 1.6171458533601328,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004740687508203779,
      "loss": 5.9082,
      "mean_token_accuracy": 0.15300915539264678,
      "num_tokens": 37474121.0,
      "step": 18505
    },
    {
      "entropy": 6.110058116912842,
      "epoch": 1.6175828017128375,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004740542257858536,
      "loss": 5.868,
      "mean_token_accuracy": 0.15394898653030395,
      "num_tokens": 37483439.0,
      "step": 18510
    },
    {
      "entropy": 6.196117782592774,
      "epoch": 1.6180197500655422,
      "grad_norm": 3.734375,
      "learning_rate": 0.00047403969693336507,
      "loss": 5.8701,
      "mean_token_accuracy": 0.15491663068532943,
      "num_tokens": 37493804.0,
      "step": 18515
    },
    {
      "entropy": 6.145570993423462,
      "epoch": 1.618456698418247,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004740251642631909,
      "loss": 5.7606,
      "mean_token_accuracy": 0.15633058845996856,
      "num_tokens": 37504158.0,
      "step": 18520
    },
    {
      "entropy": 6.094492197036743,
      "epoch": 1.6188936467709518,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004740106277756097,
      "loss": 5.8435,
      "mean_token_accuracy": 0.15085844844579696,
      "num_tokens": 37515069.0,
      "step": 18525
    },
    {
      "entropy": 6.0258582592010494,
      "epoch": 1.6193305951236563,
      "grad_norm": 1.828125,
      "learning_rate": 0.00047399608747090033,
      "loss": 5.7402,
      "mean_token_accuracy": 0.15899247005581857,
      "num_tokens": 37525486.0,
      "step": 18530
    },
    {
      "entropy": 6.142672967910767,
      "epoch": 1.6197675434763612,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047398154334934187,
      "loss": 5.8674,
      "mean_token_accuracy": 0.15258204638957978,
      "num_tokens": 37535616.0,
      "step": 18535
    },
    {
      "entropy": 6.094711589813232,
      "epoch": 1.6202044918290657,
      "grad_norm": 1.765625,
      "learning_rate": 0.00047396699541121296,
      "loss": 5.8246,
      "mean_token_accuracy": 0.1578874409198761,
      "num_tokens": 37546084.0,
      "step": 18540
    },
    {
      "entropy": 6.058980131149292,
      "epoch": 1.6206414401817706,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004739524436567929,
      "loss": 5.8388,
      "mean_token_accuracy": 0.1601293243467808,
      "num_tokens": 37555865.0,
      "step": 18545
    },
    {
      "entropy": 6.090431213378906,
      "epoch": 1.6210783885344753,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004739378880863608,
      "loss": 5.8292,
      "mean_token_accuracy": 0.15517839342355727,
      "num_tokens": 37565985.0,
      "step": 18550
    },
    {
      "entropy": 6.1478002071380615,
      "epoch": 1.62151533688718,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004739233287001957,
      "loss": 5.9188,
      "mean_token_accuracy": 0.14992354512214662,
      "num_tokens": 37575887.0,
      "step": 18555
    },
    {
      "entropy": 6.139444732666016,
      "epoch": 1.6219522852398847,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047390876549857695,
      "loss": 5.7669,
      "mean_token_accuracy": 0.1608128160238266,
      "num_tokens": 37585346.0,
      "step": 18560
    },
    {
      "entropy": 6.144708108901978,
      "epoch": 1.6223892335925894,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047389419848178384,
      "loss": 5.843,
      "mean_token_accuracy": 0.1571737825870514,
      "num_tokens": 37595674.0,
      "step": 18565
    },
    {
      "entropy": 6.109766626358033,
      "epoch": 1.622826181945294,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047387962765009576,
      "loss": 5.837,
      "mean_token_accuracy": 0.15036114826798438,
      "num_tokens": 37605537.0,
      "step": 18570
    },
    {
      "entropy": 6.148508358001709,
      "epoch": 1.6232631302979987,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047386505300379224,
      "loss": 5.8813,
      "mean_token_accuracy": 0.15324413031339645,
      "num_tokens": 37615801.0,
      "step": 18575
    },
    {
      "entropy": 6.13037633895874,
      "epoch": 1.6237000786507036,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004738504745431528,
      "loss": 5.7696,
      "mean_token_accuracy": 0.1616264298558235,
      "num_tokens": 37625582.0,
      "step": 18580
    },
    {
      "entropy": 6.104911661148071,
      "epoch": 1.6241370270034081,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047383589226845707,
      "loss": 5.8896,
      "mean_token_accuracy": 0.15660351365804673,
      "num_tokens": 37636075.0,
      "step": 18585
    },
    {
      "entropy": 6.046101427078247,
      "epoch": 1.624573975356113,
      "grad_norm": 2.125,
      "learning_rate": 0.00047382130617998477,
      "loss": 5.7254,
      "mean_token_accuracy": 0.1683705359697342,
      "num_tokens": 37645692.0,
      "step": 18590
    },
    {
      "entropy": 6.128520774841308,
      "epoch": 1.6250109237088175,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004738067162780156,
      "loss": 5.8307,
      "mean_token_accuracy": 0.1556292802095413,
      "num_tokens": 37655035.0,
      "step": 18595
    },
    {
      "entropy": 6.148236703872681,
      "epoch": 1.6254478720615224,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004737921225628294,
      "loss": 5.8172,
      "mean_token_accuracy": 0.15634016543626786,
      "num_tokens": 37665829.0,
      "step": 18600
    },
    {
      "entropy": 6.067884206771851,
      "epoch": 1.6258848204142269,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047377752503470613,
      "loss": 5.8063,
      "mean_token_accuracy": 0.16721004694700242,
      "num_tokens": 37676318.0,
      "step": 18605
    },
    {
      "entropy": 6.114659547805786,
      "epoch": 1.6263217687669318,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047376292369392584,
      "loss": 5.8606,
      "mean_token_accuracy": 0.15416946560144423,
      "num_tokens": 37686977.0,
      "step": 18610
    },
    {
      "entropy": 6.085484218597412,
      "epoch": 1.6267587171196365,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047374831854076843,
      "loss": 5.8032,
      "mean_token_accuracy": 0.15572132617235185,
      "num_tokens": 37697004.0,
      "step": 18615
    },
    {
      "entropy": 6.105016708374023,
      "epoch": 1.6271956654723412,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047373370957551423,
      "loss": 5.8355,
      "mean_token_accuracy": 0.15412657111883163,
      "num_tokens": 37707613.0,
      "step": 18620
    },
    {
      "entropy": 6.154040145874023,
      "epoch": 1.6276326138250459,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004737190967984433,
      "loss": 5.815,
      "mean_token_accuracy": 0.15441539734601975,
      "num_tokens": 37717167.0,
      "step": 18625
    },
    {
      "entropy": 6.135318040847778,
      "epoch": 1.6280695621777506,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004737044802098359,
      "loss": 5.915,
      "mean_token_accuracy": 0.14939552247524263,
      "num_tokens": 37727496.0,
      "step": 18630
    },
    {
      "entropy": 6.093013715744019,
      "epoch": 1.6285065105304553,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004736898598099725,
      "loss": 5.8211,
      "mean_token_accuracy": 0.1553942784667015,
      "num_tokens": 37736844.0,
      "step": 18635
    },
    {
      "entropy": 6.165453815460205,
      "epoch": 1.62894345888316,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047367523559913347,
      "loss": 5.8521,
      "mean_token_accuracy": 0.15251405388116837,
      "num_tokens": 37746027.0,
      "step": 18640
    },
    {
      "entropy": 6.174861478805542,
      "epoch": 1.6293804072358649,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047366060757759933,
      "loss": 5.8098,
      "mean_token_accuracy": 0.15763675719499587,
      "num_tokens": 37756499.0,
      "step": 18645
    },
    {
      "entropy": 6.101605224609375,
      "epoch": 1.6298173555885693,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00047364597574565064,
      "loss": 5.8026,
      "mean_token_accuracy": 0.15793321132659913,
      "num_tokens": 37766533.0,
      "step": 18650
    },
    {
      "entropy": 6.1170854568481445,
      "epoch": 1.6302543039412742,
      "grad_norm": 2.390625,
      "learning_rate": 0.000473631340103568,
      "loss": 5.8193,
      "mean_token_accuracy": 0.1622927963733673,
      "num_tokens": 37775937.0,
      "step": 18655
    },
    {
      "entropy": 6.122182750701905,
      "epoch": 1.6306912522939787,
      "grad_norm": 3.03125,
      "learning_rate": 0.00047361670065163225,
      "loss": 5.8239,
      "mean_token_accuracy": 0.1569531351327896,
      "num_tokens": 37786587.0,
      "step": 18660
    },
    {
      "entropy": 6.128579950332641,
      "epoch": 1.6311282006466836,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047360205739012406,
      "loss": 5.8439,
      "mean_token_accuracy": 0.15462930351495743,
      "num_tokens": 37797200.0,
      "step": 18665
    },
    {
      "entropy": 6.066535949707031,
      "epoch": 1.6315651489993883,
      "grad_norm": 3.125,
      "learning_rate": 0.0004735874103193244,
      "loss": 5.8188,
      "mean_token_accuracy": 0.16094542890787125,
      "num_tokens": 37806652.0,
      "step": 18670
    },
    {
      "entropy": 6.137418413162232,
      "epoch": 1.632002097352093,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004735727594395141,
      "loss": 5.8656,
      "mean_token_accuracy": 0.1501804545521736,
      "num_tokens": 37817057.0,
      "step": 18675
    },
    {
      "entropy": 6.117572450637818,
      "epoch": 1.6324390457047977,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00047355810475097423,
      "loss": 5.8236,
      "mean_token_accuracy": 0.15169511288404464,
      "num_tokens": 37826216.0,
      "step": 18680
    },
    {
      "entropy": 6.1833806991577145,
      "epoch": 1.6328759940575024,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00047354344625398596,
      "loss": 5.8884,
      "mean_token_accuracy": 0.1534342348575592,
      "num_tokens": 37835834.0,
      "step": 18685
    },
    {
      "entropy": 6.10155930519104,
      "epoch": 1.633312942410207,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004735287839488302,
      "loss": 5.8167,
      "mean_token_accuracy": 0.16098785549402236,
      "num_tokens": 37846252.0,
      "step": 18690
    },
    {
      "entropy": 6.101823854446411,
      "epoch": 1.6337498907629118,
      "grad_norm": 1.953125,
      "learning_rate": 0.00047351411783578846,
      "loss": 5.8107,
      "mean_token_accuracy": 0.15409183204174043,
      "num_tokens": 37856020.0,
      "step": 18695
    },
    {
      "entropy": 6.188598203659057,
      "epoch": 1.6341868391156167,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004734994479151418,
      "loss": 5.8483,
      "mean_token_accuracy": 0.15436530858278275,
      "num_tokens": 37866132.0,
      "step": 18700
    },
    {
      "entropy": 6.106948757171631,
      "epoch": 1.6346237874683212,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004734847741871718,
      "loss": 5.8997,
      "mean_token_accuracy": 0.15431200116872787,
      "num_tokens": 37876779.0,
      "step": 18705
    },
    {
      "entropy": 6.1267928123474125,
      "epoch": 1.635060735821026,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00047347009665215983,
      "loss": 5.8291,
      "mean_token_accuracy": 0.15666446685791016,
      "num_tokens": 37886981.0,
      "step": 18710
    },
    {
      "entropy": 6.103256511688232,
      "epoch": 1.6354976841737305,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004734554153103874,
      "loss": 5.7996,
      "mean_token_accuracy": 0.1629627212882042,
      "num_tokens": 37896974.0,
      "step": 18715
    },
    {
      "entropy": 6.1627795696258545,
      "epoch": 1.6359346325264355,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047344073016213615,
      "loss": 5.9436,
      "mean_token_accuracy": 0.14571316093206405,
      "num_tokens": 37909020.0,
      "step": 18720
    },
    {
      "entropy": 6.198339128494263,
      "epoch": 1.63637158087914,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047342604120768764,
      "loss": 5.796,
      "mean_token_accuracy": 0.1588145151734352,
      "num_tokens": 37919417.0,
      "step": 18725
    },
    {
      "entropy": 6.1158068656921385,
      "epoch": 1.6368085292318448,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004734113484473237,
      "loss": 5.9177,
      "mean_token_accuracy": 0.1477796733379364,
      "num_tokens": 37929030.0,
      "step": 18730
    },
    {
      "entropy": 6.090813541412354,
      "epoch": 1.6372454775845495,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004733966518813262,
      "loss": 5.8007,
      "mean_token_accuracy": 0.15683868974447251,
      "num_tokens": 37939157.0,
      "step": 18735
    },
    {
      "entropy": 6.121892595291138,
      "epoch": 1.6376824259372542,
      "grad_norm": 1.828125,
      "learning_rate": 0.00047338195150997686,
      "loss": 5.805,
      "mean_token_accuracy": 0.15147726386785507,
      "num_tokens": 37950544.0,
      "step": 18740
    },
    {
      "entropy": 6.08565993309021,
      "epoch": 1.638119374289959,
      "grad_norm": 2.78125,
      "learning_rate": 0.00047336724733355786,
      "loss": 5.7644,
      "mean_token_accuracy": 0.16694549173116685,
      "num_tokens": 37960059.0,
      "step": 18745
    },
    {
      "entropy": 6.093255233764649,
      "epoch": 1.6385563226426636,
      "grad_norm": 2.625,
      "learning_rate": 0.00047335253935235094,
      "loss": 5.7961,
      "mean_token_accuracy": 0.15961635261774063,
      "num_tokens": 37970669.0,
      "step": 18750
    },
    {
      "entropy": 6.077090787887573,
      "epoch": 1.6389932709953685,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047333782756663853,
      "loss": 5.8211,
      "mean_token_accuracy": 0.1623784527182579,
      "num_tokens": 37980088.0,
      "step": 18755
    },
    {
      "entropy": 6.130883264541626,
      "epoch": 1.639430219348073,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047332311197670264,
      "loss": 5.8651,
      "mean_token_accuracy": 0.15645087212324144,
      "num_tokens": 37989433.0,
      "step": 18760
    },
    {
      "entropy": 6.168463802337646,
      "epoch": 1.639867167700778,
      "grad_norm": 1.96875,
      "learning_rate": 0.00047330839258282544,
      "loss": 5.8538,
      "mean_token_accuracy": 0.15782661885023117,
      "num_tokens": 37999735.0,
      "step": 18765
    },
    {
      "entropy": 6.086714363098144,
      "epoch": 1.6403041160534824,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004732936693852895,
      "loss": 5.7306,
      "mean_token_accuracy": 0.1667976751923561,
      "num_tokens": 38008515.0,
      "step": 18770
    },
    {
      "entropy": 6.1335205078125,
      "epoch": 1.6407410644061873,
      "grad_norm": 1.921875,
      "learning_rate": 0.00047327894238437696,
      "loss": 5.8851,
      "mean_token_accuracy": 0.15473941415548326,
      "num_tokens": 38018778.0,
      "step": 18775
    },
    {
      "entropy": 6.169872426986695,
      "epoch": 1.6411780127588917,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047326421158037045,
      "loss": 5.8763,
      "mean_token_accuracy": 0.15240313559770585,
      "num_tokens": 38029062.0,
      "step": 18780
    },
    {
      "entropy": 6.091077470779419,
      "epoch": 1.6416149611115967,
      "grad_norm": 6.03125,
      "learning_rate": 0.00047324947697355253,
      "loss": 5.7618,
      "mean_token_accuracy": 0.16436618864536284,
      "num_tokens": 38039945.0,
      "step": 18785
    },
    {
      "entropy": 6.13494725227356,
      "epoch": 1.6420519094643014,
      "grad_norm": 1.75,
      "learning_rate": 0.00047323473856420573,
      "loss": 5.852,
      "mean_token_accuracy": 0.153858582675457,
      "num_tokens": 38050213.0,
      "step": 18790
    },
    {
      "entropy": 6.117273807525635,
      "epoch": 1.642488857817006,
      "grad_norm": 1.921875,
      "learning_rate": 0.00047321999635261276,
      "loss": 5.7821,
      "mean_token_accuracy": 0.15843318849802018,
      "num_tokens": 38060147.0,
      "step": 18795
    },
    {
      "entropy": 6.100334501266479,
      "epoch": 1.6429258061697107,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004732052503390563,
      "loss": 5.8576,
      "mean_token_accuracy": 0.15376311391592026,
      "num_tokens": 38070854.0,
      "step": 18800
    },
    {
      "entropy": 6.118924236297607,
      "epoch": 1.6433627545224154,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004731905005238194,
      "loss": 5.9109,
      "mean_token_accuracy": 0.14683781787753106,
      "num_tokens": 38080998.0,
      "step": 18805
    },
    {
      "entropy": 6.145289516448974,
      "epoch": 1.6437997028751201,
      "grad_norm": 2.25,
      "learning_rate": 0.00047317574690718486,
      "loss": 5.8302,
      "mean_token_accuracy": 0.15473107993602753,
      "num_tokens": 38091421.0,
      "step": 18810
    },
    {
      "entropy": 6.117181921005249,
      "epoch": 1.6442366512278248,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047316098948943557,
      "loss": 5.802,
      "mean_token_accuracy": 0.16150670647621154,
      "num_tokens": 38101106.0,
      "step": 18815
    },
    {
      "entropy": 6.124206638336181,
      "epoch": 1.6446735995805297,
      "grad_norm": 1.984375,
      "learning_rate": 0.00047314622827085466,
      "loss": 5.8331,
      "mean_token_accuracy": 0.15121359154582023,
      "num_tokens": 38111346.0,
      "step": 18820
    },
    {
      "entropy": 6.194262552261352,
      "epoch": 1.6451105479332342,
      "grad_norm": 1.65625,
      "learning_rate": 0.00047313146325172534,
      "loss": 5.922,
      "mean_token_accuracy": 0.147097672522068,
      "num_tokens": 38121798.0,
      "step": 18825
    },
    {
      "entropy": 6.124340009689331,
      "epoch": 1.645547496285939,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004731166944323307,
      "loss": 5.8881,
      "mean_token_accuracy": 0.14912372529506684,
      "num_tokens": 38134145.0,
      "step": 18830
    },
    {
      "entropy": 6.161386203765869,
      "epoch": 1.6459844446386436,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047310192181295406,
      "loss": 5.8036,
      "mean_token_accuracy": 0.15616045147180557,
      "num_tokens": 38144213.0,
      "step": 18835
    },
    {
      "entropy": 6.122765636444091,
      "epoch": 1.6464213929913485,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004730871453938788,
      "loss": 5.814,
      "mean_token_accuracy": 0.16242776662111283,
      "num_tokens": 38154245.0,
      "step": 18840
    },
    {
      "entropy": 6.099372720718383,
      "epoch": 1.6468583413440532,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004730723651753882,
      "loss": 5.8227,
      "mean_token_accuracy": 0.1565498560667038,
      "num_tokens": 38163969.0,
      "step": 18845
    },
    {
      "entropy": 6.059172439575195,
      "epoch": 1.6472952896967579,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004730575811577659,
      "loss": 5.7527,
      "mean_token_accuracy": 0.15794841796159745,
      "num_tokens": 38173586.0,
      "step": 18850
    },
    {
      "entropy": 6.0792951583862305,
      "epoch": 1.6477322380494626,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004730427933412954,
      "loss": 5.777,
      "mean_token_accuracy": 0.1548040106892586,
      "num_tokens": 38182702.0,
      "step": 18855
    },
    {
      "entropy": 6.07518162727356,
      "epoch": 1.6481691864021673,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004730280017262604,
      "loss": 5.806,
      "mean_token_accuracy": 0.16442347913980485,
      "num_tokens": 38191937.0,
      "step": 18860
    },
    {
      "entropy": 6.09440131187439,
      "epoch": 1.648606134754872,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004730132063129445,
      "loss": 5.7839,
      "mean_token_accuracy": 0.159683395922184,
      "num_tokens": 38201234.0,
      "step": 18865
    },
    {
      "entropy": 6.1160468578338625,
      "epoch": 1.6490430831075766,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00047299840710163157,
      "loss": 5.8284,
      "mean_token_accuracy": 0.1586320295929909,
      "num_tokens": 38211790.0,
      "step": 18870
    },
    {
      "entropy": 6.147962951660157,
      "epoch": 1.6494800314602815,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047298360409260543,
      "loss": 5.8082,
      "mean_token_accuracy": 0.15903422236442566,
      "num_tokens": 38221946.0,
      "step": 18875
    },
    {
      "entropy": 6.0754194259643555,
      "epoch": 1.649916979812986,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047296879728615,
      "loss": 5.8554,
      "mean_token_accuracy": 0.15347450822591782,
      "num_tokens": 38232366.0,
      "step": 18880
    },
    {
      "entropy": 6.133282041549682,
      "epoch": 1.650353928165691,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047295398668254933,
      "loss": 5.8837,
      "mean_token_accuracy": 0.1509290099143982,
      "num_tokens": 38242504.0,
      "step": 18885
    },
    {
      "entropy": 6.182716655731201,
      "epoch": 1.6507908765183954,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004729391722820875,
      "loss": 5.8571,
      "mean_token_accuracy": 0.15664587765932084,
      "num_tokens": 38251916.0,
      "step": 18890
    },
    {
      "entropy": 6.072823715209961,
      "epoch": 1.6512278248711003,
      "grad_norm": 2.5,
      "learning_rate": 0.00047292435408504855,
      "loss": 5.7837,
      "mean_token_accuracy": 0.15972300320863725,
      "num_tokens": 38262438.0,
      "step": 18895
    },
    {
      "entropy": 6.039656114578247,
      "epoch": 1.6516647732238048,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004729095320917169,
      "loss": 5.7734,
      "mean_token_accuracy": 0.17215760201215743,
      "num_tokens": 38273095.0,
      "step": 18900
    },
    {
      "entropy": 6.167916774749756,
      "epoch": 1.6521017215765097,
      "grad_norm": 2.5,
      "learning_rate": 0.0004728947063023766,
      "loss": 5.796,
      "mean_token_accuracy": 0.1562114581465721,
      "num_tokens": 38282827.0,
      "step": 18905
    },
    {
      "entropy": 6.219447088241577,
      "epoch": 1.6525386699292144,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004728798767173122,
      "loss": 5.806,
      "mean_token_accuracy": 0.15486627519130708,
      "num_tokens": 38292827.0,
      "step": 18910
    },
    {
      "entropy": 6.125799322128296,
      "epoch": 1.652975618281919,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047286504333680804,
      "loss": 5.8488,
      "mean_token_accuracy": 0.1487625904381275,
      "num_tokens": 38302658.0,
      "step": 18915
    },
    {
      "entropy": 6.119754648208618,
      "epoch": 1.6534125666346238,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004728502061611488,
      "loss": 5.8125,
      "mean_token_accuracy": 0.15366054475307464,
      "num_tokens": 38312154.0,
      "step": 18920
    },
    {
      "entropy": 6.12138671875,
      "epoch": 1.6538495149873285,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047283536519061875,
      "loss": 5.7915,
      "mean_token_accuracy": 0.1609782636165619,
      "num_tokens": 38321317.0,
      "step": 18925
    },
    {
      "entropy": 6.093803071975708,
      "epoch": 1.6542864633400332,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047282052042550276,
      "loss": 5.8078,
      "mean_token_accuracy": 0.15576285868883133,
      "num_tokens": 38331632.0,
      "step": 18930
    },
    {
      "entropy": 6.1320899486541744,
      "epoch": 1.6547234116927378,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047280567186608563,
      "loss": 5.8679,
      "mean_token_accuracy": 0.1505403511226177,
      "num_tokens": 38342072.0,
      "step": 18935
    },
    {
      "entropy": 6.211898517608643,
      "epoch": 1.6551603600454428,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000472790819512652,
      "loss": 5.811,
      "mean_token_accuracy": 0.15256388932466508,
      "num_tokens": 38351105.0,
      "step": 18940
    },
    {
      "entropy": 6.193249702453613,
      "epoch": 1.6555973083981472,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004727759633654868,
      "loss": 5.8789,
      "mean_token_accuracy": 0.15076037496328354,
      "num_tokens": 38361521.0,
      "step": 18945
    },
    {
      "entropy": 6.070543527603149,
      "epoch": 1.6560342567508521,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004727611034248751,
      "loss": 5.7791,
      "mean_token_accuracy": 0.15336894989013672,
      "num_tokens": 38370720.0,
      "step": 18950
    },
    {
      "entropy": 6.0799023628234865,
      "epoch": 1.6564712051035566,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004727462396911017,
      "loss": 5.8515,
      "mean_token_accuracy": 0.15835435688495636,
      "num_tokens": 38380434.0,
      "step": 18955
    },
    {
      "entropy": 6.022542905807495,
      "epoch": 1.6569081534562615,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047273137216445176,
      "loss": 5.7317,
      "mean_token_accuracy": 0.173357991874218,
      "num_tokens": 38390503.0,
      "step": 18960
    },
    {
      "entropy": 6.16728367805481,
      "epoch": 1.6573451018089662,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004727165008452106,
      "loss": 5.952,
      "mean_token_accuracy": 0.1479262411594391,
      "num_tokens": 38401757.0,
      "step": 18965
    },
    {
      "entropy": 6.112488460540772,
      "epoch": 1.657782050161671,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004727016257336633,
      "loss": 5.8418,
      "mean_token_accuracy": 0.15839479118585587,
      "num_tokens": 38412130.0,
      "step": 18970
    },
    {
      "entropy": 6.101836204528809,
      "epoch": 1.6582189985143756,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047268674683009526,
      "loss": 5.8219,
      "mean_token_accuracy": 0.15906714797019958,
      "num_tokens": 38422155.0,
      "step": 18975
    },
    {
      "entropy": 6.116298866271973,
      "epoch": 1.6586559468670803,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047267186413479185,
      "loss": 5.7519,
      "mean_token_accuracy": 0.16271159201860427,
      "num_tokens": 38431621.0,
      "step": 18980
    },
    {
      "entropy": 6.178564023971558,
      "epoch": 1.659092895219785,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047265697764803843,
      "loss": 5.9011,
      "mean_token_accuracy": 0.15169056057929992,
      "num_tokens": 38442354.0,
      "step": 18985
    },
    {
      "entropy": 6.085093355178833,
      "epoch": 1.6595298435724897,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00047264208737012064,
      "loss": 5.8335,
      "mean_token_accuracy": 0.15349179804325103,
      "num_tokens": 38452932.0,
      "step": 18990
    },
    {
      "entropy": 6.094286775588989,
      "epoch": 1.6599667919251946,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004726271933013241,
      "loss": 5.8281,
      "mean_token_accuracy": 0.1560909390449524,
      "num_tokens": 38464442.0,
      "step": 18995
    },
    {
      "entropy": 6.140982151031494,
      "epoch": 1.660403740277899,
      "grad_norm": 1.953125,
      "learning_rate": 0.00047261229544193434,
      "loss": 5.8461,
      "mean_token_accuracy": 0.15364985167980194,
      "num_tokens": 38475718.0,
      "step": 19000
    },
    {
      "entropy": 6.0793756484985355,
      "epoch": 1.660840688630604,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047259739379223726,
      "loss": 5.8179,
      "mean_token_accuracy": 0.15528008192777634,
      "num_tokens": 38485091.0,
      "step": 19005
    },
    {
      "entropy": 6.151020765304565,
      "epoch": 1.6612776369833084,
      "grad_norm": 2.421875,
      "learning_rate": 0.00047258248835251867,
      "loss": 5.8278,
      "mean_token_accuracy": 0.15394909381866456,
      "num_tokens": 38495748.0,
      "step": 19010
    },
    {
      "entropy": 6.190218257904053,
      "epoch": 1.6617145853360134,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047256757912306435,
      "loss": 5.8852,
      "mean_token_accuracy": 0.1539991468191147,
      "num_tokens": 38506152.0,
      "step": 19015
    },
    {
      "entropy": 6.156981611251831,
      "epoch": 1.662151533688718,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004725526661041604,
      "loss": 5.7962,
      "mean_token_accuracy": 0.15531576201319694,
      "num_tokens": 38515722.0,
      "step": 19020
    },
    {
      "entropy": 6.083573865890503,
      "epoch": 1.6625884820414227,
      "grad_norm": 2.125,
      "learning_rate": 0.0004725377492960927,
      "loss": 5.7999,
      "mean_token_accuracy": 0.16602276414632797,
      "num_tokens": 38526270.0,
      "step": 19025
    },
    {
      "entropy": 6.136768865585327,
      "epoch": 1.6630254303941274,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004725228286991475,
      "loss": 5.7949,
      "mean_token_accuracy": 0.15266606360673904,
      "num_tokens": 38537436.0,
      "step": 19030
    },
    {
      "entropy": 6.1395446300506595,
      "epoch": 1.6634623787468321,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004725079043136111,
      "loss": 5.8012,
      "mean_token_accuracy": 0.1607723727822304,
      "num_tokens": 38547298.0,
      "step": 19035
    },
    {
      "entropy": 6.068498325347901,
      "epoch": 1.6638993270995368,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004724929761397694,
      "loss": 5.8054,
      "mean_token_accuracy": 0.1583385407924652,
      "num_tokens": 38556465.0,
      "step": 19040
    },
    {
      "entropy": 6.072360706329346,
      "epoch": 1.6643362754522415,
      "grad_norm": 1.8203125,
      "learning_rate": 0.000472478044177909,
      "loss": 5.8425,
      "mean_token_accuracy": 0.1597535043954849,
      "num_tokens": 38568009.0,
      "step": 19045
    },
    {
      "entropy": 6.146525239944458,
      "epoch": 1.6647732238049464,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004724631084283163,
      "loss": 5.8573,
      "mean_token_accuracy": 0.15977909564971923,
      "num_tokens": 38578778.0,
      "step": 19050
    },
    {
      "entropy": 6.0575133800506595,
      "epoch": 1.6652101721576509,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047244816889127763,
      "loss": 5.7483,
      "mean_token_accuracy": 0.1671575352549553,
      "num_tokens": 38588991.0,
      "step": 19055
    },
    {
      "entropy": 6.155195903778076,
      "epoch": 1.6656471205103558,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004724332255670797,
      "loss": 5.7757,
      "mean_token_accuracy": 0.16592033207416534,
      "num_tokens": 38598800.0,
      "step": 19060
    },
    {
      "entropy": 6.070187377929687,
      "epoch": 1.6660840688630603,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00047241827845600896,
      "loss": 5.8069,
      "mean_token_accuracy": 0.15673100650310517,
      "num_tokens": 38608680.0,
      "step": 19065
    },
    {
      "entropy": 6.124496603012085,
      "epoch": 1.6665210172157652,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047240332755835217,
      "loss": 5.8361,
      "mean_token_accuracy": 0.15894126668572425,
      "num_tokens": 38618402.0,
      "step": 19070
    },
    {
      "entropy": 6.26353907585144,
      "epoch": 1.6669579655684696,
      "grad_norm": 5.9375,
      "learning_rate": 0.00047238837287439625,
      "loss": 5.8538,
      "mean_token_accuracy": 0.15350157767534256,
      "num_tokens": 38628370.0,
      "step": 19075
    },
    {
      "entropy": 6.1332025051116945,
      "epoch": 1.6673949139211746,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004723734144044278,
      "loss": 5.784,
      "mean_token_accuracy": 0.1663918063044548,
      "num_tokens": 38638138.0,
      "step": 19080
    },
    {
      "entropy": 6.117161989212036,
      "epoch": 1.6678318622738793,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047235845214873385,
      "loss": 5.8663,
      "mean_token_accuracy": 0.15420200526714326,
      "num_tokens": 38649412.0,
      "step": 19085
    },
    {
      "entropy": 6.145470809936524,
      "epoch": 1.668268810626584,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047234348610760145,
      "loss": 5.7974,
      "mean_token_accuracy": 0.1533084586262703,
      "num_tokens": 38660048.0,
      "step": 19090
    },
    {
      "entropy": 6.105442905426026,
      "epoch": 1.6687057589792886,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004723285162813175,
      "loss": 5.7969,
      "mean_token_accuracy": 0.16411686688661575,
      "num_tokens": 38671246.0,
      "step": 19095
    },
    {
      "entropy": 6.117104768753052,
      "epoch": 1.6691427073319933,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004723135426701693,
      "loss": 5.8273,
      "mean_token_accuracy": 0.1577977180480957,
      "num_tokens": 38680880.0,
      "step": 19100
    },
    {
      "entropy": 6.044180011749267,
      "epoch": 1.669579655684698,
      "grad_norm": 6.53125,
      "learning_rate": 0.0004722985652744438,
      "loss": 5.8667,
      "mean_token_accuracy": 0.15735935270786286,
      "num_tokens": 38691440.0,
      "step": 19105
    },
    {
      "entropy": 6.119514274597168,
      "epoch": 1.6700166040374027,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00047228358409442863,
      "loss": 5.8091,
      "mean_token_accuracy": 0.1662777468562126,
      "num_tokens": 38700846.0,
      "step": 19110
    },
    {
      "entropy": 6.238430547714233,
      "epoch": 1.6704535523901076,
      "grad_norm": 1.828125,
      "learning_rate": 0.00047226859913041084,
      "loss": 5.8645,
      "mean_token_accuracy": 0.16070397943258286,
      "num_tokens": 38712130.0,
      "step": 19115
    },
    {
      "entropy": 6.180017757415771,
      "epoch": 1.670890500742812,
      "grad_norm": 2.984375,
      "learning_rate": 0.00047225361038267795,
      "loss": 5.9067,
      "mean_token_accuracy": 0.15520768612623215,
      "num_tokens": 38722482.0,
      "step": 19120
    },
    {
      "entropy": 5.9815174579620365,
      "epoch": 1.671327449095517,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047223861785151744,
      "loss": 5.7566,
      "mean_token_accuracy": 0.15863682925701142,
      "num_tokens": 38731936.0,
      "step": 19125
    },
    {
      "entropy": 6.0875166893005375,
      "epoch": 1.6717643974482215,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004722236215372169,
      "loss": 5.7405,
      "mean_token_accuracy": 0.15893969535827637,
      "num_tokens": 38741444.0,
      "step": 19130
    },
    {
      "entropy": 6.21986083984375,
      "epoch": 1.6722013458009264,
      "grad_norm": 2.09375,
      "learning_rate": 0.000472208621440064,
      "loss": 5.8781,
      "mean_token_accuracy": 0.14849231764674187,
      "num_tokens": 38751454.0,
      "step": 19135
    },
    {
      "entropy": 6.1371866226196286,
      "epoch": 1.672638294153631,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004721936175603463,
      "loss": 5.8272,
      "mean_token_accuracy": 0.14809708148241044,
      "num_tokens": 38761373.0,
      "step": 19140
    },
    {
      "entropy": 6.120560216903686,
      "epoch": 1.6730752425063358,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047217860989835184,
      "loss": 5.8891,
      "mean_token_accuracy": 0.16308124959468842,
      "num_tokens": 38772047.0,
      "step": 19145
    },
    {
      "entropy": 6.145443487167358,
      "epoch": 1.6735121908590405,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004721635984543682,
      "loss": 5.7741,
      "mean_token_accuracy": 0.16155510991811753,
      "num_tokens": 38781393.0,
      "step": 19150
    },
    {
      "entropy": 6.145397710800171,
      "epoch": 1.6739491392117452,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004721485832286835,
      "loss": 5.8036,
      "mean_token_accuracy": 0.15633226037025452,
      "num_tokens": 38791104.0,
      "step": 19155
    },
    {
      "entropy": 6.039845895767212,
      "epoch": 1.6743860875644498,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004721335642215856,
      "loss": 5.8627,
      "mean_token_accuracy": 0.15498319491744042,
      "num_tokens": 38801855.0,
      "step": 19160
    },
    {
      "entropy": 6.044262790679932,
      "epoch": 1.6748230359171545,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004721185414333627,
      "loss": 5.7402,
      "mean_token_accuracy": 0.16360869854688645,
      "num_tokens": 38811983.0,
      "step": 19165
    },
    {
      "entropy": 6.076495170593262,
      "epoch": 1.6752599842698594,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004721035148643029,
      "loss": 5.7706,
      "mean_token_accuracy": 0.15775015950202942,
      "num_tokens": 38821762.0,
      "step": 19170
    },
    {
      "entropy": 6.044222021102906,
      "epoch": 1.675696932622564,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004720884845146944,
      "loss": 5.8175,
      "mean_token_accuracy": 0.16202907115221024,
      "num_tokens": 38831840.0,
      "step": 19175
    },
    {
      "entropy": 6.102727460861206,
      "epoch": 1.6761338809752688,
      "grad_norm": 2.796875,
      "learning_rate": 0.00047207345038482545,
      "loss": 5.7381,
      "mean_token_accuracy": 0.1704510048031807,
      "num_tokens": 38841506.0,
      "step": 19180
    },
    {
      "entropy": 6.145182085037232,
      "epoch": 1.6765708293279733,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004720584124749844,
      "loss": 5.8712,
      "mean_token_accuracy": 0.15555896311998368,
      "num_tokens": 38851793.0,
      "step": 19185
    },
    {
      "entropy": 6.011753129959106,
      "epoch": 1.6770077776806782,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047204337078545985,
      "loss": 5.7769,
      "mean_token_accuracy": 0.16332475394010543,
      "num_tokens": 38861958.0,
      "step": 19190
    },
    {
      "entropy": 6.086695909500122,
      "epoch": 1.677444726033383,
      "grad_norm": 2.375,
      "learning_rate": 0.0004720283253165402,
      "loss": 5.8019,
      "mean_token_accuracy": 0.15991731584072114,
      "num_tokens": 38872181.0,
      "step": 19195
    },
    {
      "entropy": 6.173571586608887,
      "epoch": 1.6778816743860876,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047201327606851394,
      "loss": 5.8367,
      "mean_token_accuracy": 0.15976209789514542,
      "num_tokens": 38882423.0,
      "step": 19200
    },
    {
      "entropy": 6.16213698387146,
      "epoch": 1.6783186227387923,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004719982230416699,
      "loss": 5.8349,
      "mean_token_accuracy": 0.15263132005929947,
      "num_tokens": 38892833.0,
      "step": 19205
    },
    {
      "entropy": 6.139116191864014,
      "epoch": 1.678755571091497,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004719831662362967,
      "loss": 5.8888,
      "mean_token_accuracy": 0.15749704241752624,
      "num_tokens": 38904037.0,
      "step": 19210
    },
    {
      "entropy": 6.0898291110992435,
      "epoch": 1.6791925194442017,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004719681056526831,
      "loss": 5.795,
      "mean_token_accuracy": 0.164544478058815,
      "num_tokens": 38914194.0,
      "step": 19215
    },
    {
      "entropy": 6.039488077163696,
      "epoch": 1.6796294677969064,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047195304129111804,
      "loss": 5.8509,
      "mean_token_accuracy": 0.15357363373041152,
      "num_tokens": 38925354.0,
      "step": 19220
    },
    {
      "entropy": 6.172190713882446,
      "epoch": 1.6800664161496113,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047193797315189047,
      "loss": 5.8548,
      "mean_token_accuracy": 0.15298427045345306,
      "num_tokens": 38936170.0,
      "step": 19225
    },
    {
      "entropy": 6.172828149795532,
      "epoch": 1.6805033645023157,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004719229012352894,
      "loss": 5.8449,
      "mean_token_accuracy": 0.15547781437635422,
      "num_tokens": 38945316.0,
      "step": 19230
    },
    {
      "entropy": 6.130415439605713,
      "epoch": 1.6809403128550207,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047190782554160386,
      "loss": 5.8396,
      "mean_token_accuracy": 0.15261305794119834,
      "num_tokens": 38955444.0,
      "step": 19235
    },
    {
      "entropy": 6.080860996246338,
      "epoch": 1.6813772612077251,
      "grad_norm": 2.25,
      "learning_rate": 0.0004718927460711231,
      "loss": 5.7725,
      "mean_token_accuracy": 0.16373928785324096,
      "num_tokens": 38965202.0,
      "step": 19240
    },
    {
      "entropy": 6.053822183609009,
      "epoch": 1.68181420956043,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047187766282413623,
      "loss": 5.8399,
      "mean_token_accuracy": 0.15780514031648635,
      "num_tokens": 38975822.0,
      "step": 19245
    },
    {
      "entropy": 6.110651063919067,
      "epoch": 1.6822511579131345,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004718625758009327,
      "loss": 5.8095,
      "mean_token_accuracy": 0.16059374213218688,
      "num_tokens": 38985890.0,
      "step": 19250
    },
    {
      "entropy": 6.125767564773559,
      "epoch": 1.6826881062658394,
      "grad_norm": 3.0,
      "learning_rate": 0.00047184748500180175,
      "loss": 5.7942,
      "mean_token_accuracy": 0.16184118539094924,
      "num_tokens": 38996374.0,
      "step": 19255
    },
    {
      "entropy": 6.08174500465393,
      "epoch": 1.6831250546185441,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004718323904270329,
      "loss": 5.8103,
      "mean_token_accuracy": 0.1581805855035782,
      "num_tokens": 39006371.0,
      "step": 19260
    },
    {
      "entropy": 6.079828882217408,
      "epoch": 1.6835620029712488,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004718172920769157,
      "loss": 5.7721,
      "mean_token_accuracy": 0.16084895133972169,
      "num_tokens": 39015514.0,
      "step": 19265
    },
    {
      "entropy": 6.096600675582886,
      "epoch": 1.6839989513239535,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004718021899517397,
      "loss": 5.79,
      "mean_token_accuracy": 0.16337790489196777,
      "num_tokens": 39025700.0,
      "step": 19270
    },
    {
      "entropy": 6.063145446777344,
      "epoch": 1.6844358996766582,
      "grad_norm": 2.953125,
      "learning_rate": 0.00047178708405179454,
      "loss": 5.8365,
      "mean_token_accuracy": 0.1638899862766266,
      "num_tokens": 39036157.0,
      "step": 19275
    },
    {
      "entropy": 6.153424978256226,
      "epoch": 1.6848728480293629,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047177197437737003,
      "loss": 5.9458,
      "mean_token_accuracy": 0.15151374638080597,
      "num_tokens": 39046044.0,
      "step": 19280
    },
    {
      "entropy": 6.163524580001831,
      "epoch": 1.6853097963820676,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004717568609287559,
      "loss": 5.8601,
      "mean_token_accuracy": 0.14842982366681098,
      "num_tokens": 39056034.0,
      "step": 19285
    },
    {
      "entropy": 6.140909671783447,
      "epoch": 1.6857467447347725,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004717417437062421,
      "loss": 5.8127,
      "mean_token_accuracy": 0.15855249464511872,
      "num_tokens": 39065687.0,
      "step": 19290
    },
    {
      "entropy": 6.110138702392578,
      "epoch": 1.686183693087477,
      "grad_norm": 2.125,
      "learning_rate": 0.00047172662271011855,
      "loss": 5.8198,
      "mean_token_accuracy": 0.15501875281333924,
      "num_tokens": 39077107.0,
      "step": 19295
    },
    {
      "entropy": 6.084283828735352,
      "epoch": 1.6866206414401819,
      "grad_norm": 1.9375,
      "learning_rate": 0.00047171149794067533,
      "loss": 5.8218,
      "mean_token_accuracy": 0.15895448625087738,
      "num_tokens": 39087220.0,
      "step": 19300
    },
    {
      "entropy": 6.1323915958404545,
      "epoch": 1.6870575897928863,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004716963693982025,
      "loss": 5.7649,
      "mean_token_accuracy": 0.16237185448408126,
      "num_tokens": 39097706.0,
      "step": 19305
    },
    {
      "entropy": 6.125293827056884,
      "epoch": 1.6874945381455912,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004716812370829902,
      "loss": 5.8325,
      "mean_token_accuracy": 0.1505938410758972,
      "num_tokens": 39108044.0,
      "step": 19310
    },
    {
      "entropy": 6.0772223472595215,
      "epoch": 1.687931486498296,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004716661009953287,
      "loss": 5.8118,
      "mean_token_accuracy": 0.1537209965288639,
      "num_tokens": 39116917.0,
      "step": 19315
    },
    {
      "entropy": 6.056534147262573,
      "epoch": 1.6883684348510006,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004716509611355084,
      "loss": 5.8115,
      "mean_token_accuracy": 0.16177838891744614,
      "num_tokens": 39126770.0,
      "step": 19320
    },
    {
      "entropy": 6.069213485717773,
      "epoch": 1.6888053832037053,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047163581750381956,
      "loss": 5.8373,
      "mean_token_accuracy": 0.15334124118089676,
      "num_tokens": 39137406.0,
      "step": 19325
    },
    {
      "entropy": 6.111293697357178,
      "epoch": 1.68924233155641,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047162067010055275,
      "loss": 5.7981,
      "mean_token_accuracy": 0.15638425350189208,
      "num_tokens": 39146950.0,
      "step": 19330
    },
    {
      "entropy": 6.128775310516358,
      "epoch": 1.6896792799091147,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004716055189259984,
      "loss": 5.7575,
      "mean_token_accuracy": 0.1567702740430832,
      "num_tokens": 39157382.0,
      "step": 19335
    },
    {
      "entropy": 6.082703018188477,
      "epoch": 1.6901162282618194,
      "grad_norm": 2.734375,
      "learning_rate": 0.00047159036398044727,
      "loss": 5.8044,
      "mean_token_accuracy": 0.15351771414279938,
      "num_tokens": 39166641.0,
      "step": 19340
    },
    {
      "entropy": 6.073806095123291,
      "epoch": 1.6905531766145243,
      "grad_norm": 2.0,
      "learning_rate": 0.0004715752052641899,
      "loss": 5.8544,
      "mean_token_accuracy": 0.15833919048309325,
      "num_tokens": 39176731.0,
      "step": 19345
    },
    {
      "entropy": 6.172596979141235,
      "epoch": 1.6909901249672288,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004715600427775171,
      "loss": 5.8432,
      "mean_token_accuracy": 0.1556394711136818,
      "num_tokens": 39187202.0,
      "step": 19350
    },
    {
      "entropy": 6.141682434082031,
      "epoch": 1.6914270733199337,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004715448765207196,
      "loss": 5.8361,
      "mean_token_accuracy": 0.15623832643032073,
      "num_tokens": 39197302.0,
      "step": 19355
    },
    {
      "entropy": 6.161933422088623,
      "epoch": 1.6918640216726382,
      "grad_norm": 2.25,
      "learning_rate": 0.0004715297064940885,
      "loss": 5.8449,
      "mean_token_accuracy": 0.15721532255411147,
      "num_tokens": 39208382.0,
      "step": 19360
    },
    {
      "entropy": 6.108320522308349,
      "epoch": 1.692300970025343,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004715145326979146,
      "loss": 5.7858,
      "mean_token_accuracy": 0.15833992809057235,
      "num_tokens": 39218529.0,
      "step": 19365
    },
    {
      "entropy": 6.094034099578858,
      "epoch": 1.6927379183780478,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047149935513248896,
      "loss": 5.8515,
      "mean_token_accuracy": 0.15539976805448533,
      "num_tokens": 39228976.0,
      "step": 19370
    },
    {
      "entropy": 6.103112697601318,
      "epoch": 1.6931748667307525,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004714841737981028,
      "loss": 5.738,
      "mean_token_accuracy": 0.1658128723502159,
      "num_tokens": 39238921.0,
      "step": 19375
    },
    {
      "entropy": 6.018022108078003,
      "epoch": 1.6936118150834572,
      "grad_norm": 2.125,
      "learning_rate": 0.00047146898869504717,
      "loss": 5.7631,
      "mean_token_accuracy": 0.1520024523139,
      "num_tokens": 39247978.0,
      "step": 19380
    },
    {
      "entropy": 6.070518350601196,
      "epoch": 1.6940487634361618,
      "grad_norm": 2.65625,
      "learning_rate": 0.00047145379982361336,
      "loss": 5.784,
      "mean_token_accuracy": 0.16389472037553787,
      "num_tokens": 39257487.0,
      "step": 19385
    },
    {
      "entropy": 6.11506609916687,
      "epoch": 1.6944857117888665,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00047143860718409285,
      "loss": 5.772,
      "mean_token_accuracy": 0.16077075898647308,
      "num_tokens": 39267819.0,
      "step": 19390
    },
    {
      "entropy": 6.141073894500733,
      "epoch": 1.6949226601415712,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047142341077677685,
      "loss": 5.8704,
      "mean_token_accuracy": 0.1514420062303543,
      "num_tokens": 39277673.0,
      "step": 19395
    },
    {
      "entropy": 6.106815814971924,
      "epoch": 1.6953596084942761,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004714082106019569,
      "loss": 5.7257,
      "mean_token_accuracy": 0.15973930805921555,
      "num_tokens": 39287340.0,
      "step": 19400
    },
    {
      "entropy": 6.112748098373413,
      "epoch": 1.6957965568469806,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004713930066599246,
      "loss": 5.8611,
      "mean_token_accuracy": 0.16122769713401794,
      "num_tokens": 39298277.0,
      "step": 19405
    },
    {
      "entropy": 6.123789072036743,
      "epoch": 1.6962335051996855,
      "grad_norm": 3.15625,
      "learning_rate": 0.00047137779895097145,
      "loss": 5.8217,
      "mean_token_accuracy": 0.15225579738616943,
      "num_tokens": 39307996.0,
      "step": 19410
    },
    {
      "entropy": 6.097979784011841,
      "epoch": 1.69667045355239,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047136258747538926,
      "loss": 5.803,
      "mean_token_accuracy": 0.15645243972539902,
      "num_tokens": 39317327.0,
      "step": 19415
    },
    {
      "entropy": 6.063478755950928,
      "epoch": 1.697107401905095,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004713473722334697,
      "loss": 5.7537,
      "mean_token_accuracy": 0.16891108006238936,
      "num_tokens": 39326728.0,
      "step": 19420
    },
    {
      "entropy": 6.1465741157531735,
      "epoch": 1.6975443502577994,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00047133215322550464,
      "loss": 5.9263,
      "mean_token_accuracy": 0.15052735805511475,
      "num_tokens": 39336606.0,
      "step": 19425
    },
    {
      "entropy": 6.161265277862549,
      "epoch": 1.6979812986105043,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004713169304517861,
      "loss": 5.8177,
      "mean_token_accuracy": 0.15517356991767883,
      "num_tokens": 39345622.0,
      "step": 19430
    },
    {
      "entropy": 6.162582921981811,
      "epoch": 1.698418246963209,
      "grad_norm": 1.859375,
      "learning_rate": 0.00047130170391260597,
      "loss": 5.8144,
      "mean_token_accuracy": 0.15732862651348115,
      "num_tokens": 39356868.0,
      "step": 19435
    },
    {
      "entropy": 6.071225881576538,
      "epoch": 1.6988551953159137,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004712864736082562,
      "loss": 5.7278,
      "mean_token_accuracy": 0.16240954250097275,
      "num_tokens": 39366866.0,
      "step": 19440
    },
    {
      "entropy": 6.0279035568237305,
      "epoch": 1.6992921436686184,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00047127123953902907,
      "loss": 5.8019,
      "mean_token_accuracy": 0.1561470329761505,
      "num_tokens": 39377303.0,
      "step": 19445
    },
    {
      "entropy": 6.094523334503174,
      "epoch": 1.699729092021323,
      "grad_norm": 2.375,
      "learning_rate": 0.0004712560017052167,
      "loss": 5.7729,
      "mean_token_accuracy": 0.16477225571870804,
      "num_tokens": 39387191.0,
      "step": 19450
    },
    {
      "entropy": 6.0450057029724125,
      "epoch": 1.7001660403740277,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004712407601071114,
      "loss": 5.8029,
      "mean_token_accuracy": 0.1633811041712761,
      "num_tokens": 39397416.0,
      "step": 19455
    },
    {
      "entropy": 6.122595977783203,
      "epoch": 1.7006029887267324,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047122551474500555,
      "loss": 5.8209,
      "mean_token_accuracy": 0.15754495412111283,
      "num_tokens": 39407855.0,
      "step": 19460
    },
    {
      "entropy": 6.141034698486328,
      "epoch": 1.7010399370794373,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047121026561919137,
      "loss": 5.8404,
      "mean_token_accuracy": 0.15303106158971785,
      "num_tokens": 39417735.0,
      "step": 19465
    },
    {
      "entropy": 6.131769800186158,
      "epoch": 1.7014768854321418,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004711950127299616,
      "loss": 5.8011,
      "mean_token_accuracy": 0.15741673707962037,
      "num_tokens": 39427641.0,
      "step": 19470
    },
    {
      "entropy": 6.0494544982910154,
      "epoch": 1.7019138337848467,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004711797560776087,
      "loss": 5.7202,
      "mean_token_accuracy": 0.1707633838057518,
      "num_tokens": 39437372.0,
      "step": 19475
    },
    {
      "entropy": 6.125210428237915,
      "epoch": 1.7023507821375512,
      "grad_norm": 2.90625,
      "learning_rate": 0.00047116449566242516,
      "loss": 5.8116,
      "mean_token_accuracy": 0.16120726615190506,
      "num_tokens": 39447800.0,
      "step": 19480
    },
    {
      "entropy": 6.1110752582550045,
      "epoch": 1.7027877304902561,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004711492314847038,
      "loss": 5.8512,
      "mean_token_accuracy": 0.1558827742934227,
      "num_tokens": 39457310.0,
      "step": 19485
    },
    {
      "entropy": 6.0604277610778805,
      "epoch": 1.7032246788429608,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047113396354473753,
      "loss": 5.8505,
      "mean_token_accuracy": 0.15217647403478624,
      "num_tokens": 39468579.0,
      "step": 19490
    },
    {
      "entropy": 6.14478440284729,
      "epoch": 1.7036616271956655,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000471118691842819,
      "loss": 5.7832,
      "mean_token_accuracy": 0.15656710267066956,
      "num_tokens": 39479027.0,
      "step": 19495
    },
    {
      "entropy": 6.078792572021484,
      "epoch": 1.7040985755483702,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047110341637924126,
      "loss": 5.7844,
      "mean_token_accuracy": 0.1674623131752014,
      "num_tokens": 39488651.0,
      "step": 19500
    },
    {
      "entropy": 6.016928243637085,
      "epoch": 1.7045355239010749,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004710881371542972,
      "loss": 5.8205,
      "mean_token_accuracy": 0.1599768027663231,
      "num_tokens": 39498633.0,
      "step": 19505
    },
    {
      "entropy": 6.077724409103394,
      "epoch": 1.7049724722537796,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047107285416827984,
      "loss": 5.7245,
      "mean_token_accuracy": 0.16976630091667175,
      "num_tokens": 39509351.0,
      "step": 19510
    },
    {
      "entropy": 6.059860515594482,
      "epoch": 1.7054094206064843,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004710575674214824,
      "loss": 5.7652,
      "mean_token_accuracy": 0.17138244360685348,
      "num_tokens": 39519821.0,
      "step": 19515
    },
    {
      "entropy": 6.08597354888916,
      "epoch": 1.7058463689591892,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004710422769141982,
      "loss": 5.8304,
      "mean_token_accuracy": 0.1571711540222168,
      "num_tokens": 39530040.0,
      "step": 19520
    },
    {
      "entropy": 6.1210408210754395,
      "epoch": 1.7062833173118936,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004710269826467203,
      "loss": 5.8161,
      "mean_token_accuracy": 0.15810015946626663,
      "num_tokens": 39540669.0,
      "step": 19525
    },
    {
      "entropy": 6.138817501068115,
      "epoch": 1.7067202656645986,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047101168461934225,
      "loss": 5.8071,
      "mean_token_accuracy": 0.15717188715934755,
      "num_tokens": 39550651.0,
      "step": 19530
    },
    {
      "entropy": 6.082176446914673,
      "epoch": 1.707157214017303,
      "grad_norm": 2.125,
      "learning_rate": 0.0004709963828323573,
      "loss": 5.8054,
      "mean_token_accuracy": 0.14699531793594361,
      "num_tokens": 39561748.0,
      "step": 19535
    },
    {
      "entropy": 6.058559131622315,
      "epoch": 1.707594162370008,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004709810772860591,
      "loss": 5.7454,
      "mean_token_accuracy": 0.1647449776530266,
      "num_tokens": 39571328.0,
      "step": 19540
    },
    {
      "entropy": 6.062957096099853,
      "epoch": 1.7080311107227124,
      "grad_norm": 1.65625,
      "learning_rate": 0.000470965767980741,
      "loss": 5.7642,
      "mean_token_accuracy": 0.16907085478305817,
      "num_tokens": 39581931.0,
      "step": 19545
    },
    {
      "entropy": 6.106709575653076,
      "epoch": 1.7084680590754173,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047095045491669696,
      "loss": 5.9287,
      "mean_token_accuracy": 0.15409351736307145,
      "num_tokens": 39592818.0,
      "step": 19550
    },
    {
      "entropy": 6.0509696960449215,
      "epoch": 1.708905007428122,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047093513809422043,
      "loss": 5.8039,
      "mean_token_accuracy": 0.159418548643589,
      "num_tokens": 39602867.0,
      "step": 19555
    },
    {
      "entropy": 6.10002760887146,
      "epoch": 1.7093419557808267,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004709198175136053,
      "loss": 5.8111,
      "mean_token_accuracy": 0.15785362124443053,
      "num_tokens": 39613185.0,
      "step": 19560
    },
    {
      "entropy": 6.111972522735596,
      "epoch": 1.7097789041335314,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004709044931751454,
      "loss": 5.7734,
      "mean_token_accuracy": 0.16488297283649445,
      "num_tokens": 39623262.0,
      "step": 19565
    },
    {
      "entropy": 6.152746105194092,
      "epoch": 1.710215852486236,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004708891650791347,
      "loss": 5.8064,
      "mean_token_accuracy": 0.1590514674782753,
      "num_tokens": 39633184.0,
      "step": 19570
    },
    {
      "entropy": 6.069907903671265,
      "epoch": 1.710652800838941,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004708738332258671,
      "loss": 5.8354,
      "mean_token_accuracy": 0.15927487015724182,
      "num_tokens": 39643778.0,
      "step": 19575
    },
    {
      "entropy": 6.024876642227173,
      "epoch": 1.7110897491916455,
      "grad_norm": 2.125,
      "learning_rate": 0.0004708584976156368,
      "loss": 5.7518,
      "mean_token_accuracy": 0.16237813532352446,
      "num_tokens": 39653519.0,
      "step": 19580
    },
    {
      "entropy": 6.065796899795532,
      "epoch": 1.7115266975443504,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047084315824873793,
      "loss": 5.8202,
      "mean_token_accuracy": 0.15744800567626954,
      "num_tokens": 39663824.0,
      "step": 19585
    },
    {
      "entropy": 6.117559957504272,
      "epoch": 1.7119636458970549,
      "grad_norm": 1.875,
      "learning_rate": 0.0004708278151254646,
      "loss": 5.8159,
      "mean_token_accuracy": 0.15706327557563782,
      "num_tokens": 39673631.0,
      "step": 19590
    },
    {
      "entropy": 6.20908145904541,
      "epoch": 1.7124005942497598,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004708124682461112,
      "loss": 5.9043,
      "mean_token_accuracy": 0.1504289224743843,
      "num_tokens": 39683118.0,
      "step": 19595
    },
    {
      "entropy": 6.143590927124023,
      "epoch": 1.7128375426024642,
      "grad_norm": 3.9375,
      "learning_rate": 0.00047079711761097197,
      "loss": 5.8102,
      "mean_token_accuracy": 0.15656877309083939,
      "num_tokens": 39693947.0,
      "step": 19600
    },
    {
      "entropy": 6.1515895366668705,
      "epoch": 1.7132744909551691,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004707817632203415,
      "loss": 5.822,
      "mean_token_accuracy": 0.15703746378421785,
      "num_tokens": 39703874.0,
      "step": 19605
    },
    {
      "entropy": 6.100787544250489,
      "epoch": 1.7137114393078738,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047076640507451425,
      "loss": 5.7793,
      "mean_token_accuracy": 0.16517805308103561,
      "num_tokens": 39714997.0,
      "step": 19610
    },
    {
      "entropy": 6.133305835723877,
      "epoch": 1.7141483876605785,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004707510431737847,
      "loss": 5.7837,
      "mean_token_accuracy": 0.1590082861483097,
      "num_tokens": 39726042.0,
      "step": 19615
    },
    {
      "entropy": 6.072143936157227,
      "epoch": 1.7145853360132832,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004707356775184476,
      "loss": 5.81,
      "mean_token_accuracy": 0.15487070977687836,
      "num_tokens": 39736733.0,
      "step": 19620
    },
    {
      "entropy": 6.172472286224365,
      "epoch": 1.715022284365988,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004707203081087977,
      "loss": 5.9035,
      "mean_token_accuracy": 0.14709966480731965,
      "num_tokens": 39746811.0,
      "step": 19625
    },
    {
      "entropy": 6.170523166656494,
      "epoch": 1.7154592327186926,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004707049349451297,
      "loss": 5.8335,
      "mean_token_accuracy": 0.15364097952842712,
      "num_tokens": 39757371.0,
      "step": 19630
    },
    {
      "entropy": 6.069514322280884,
      "epoch": 1.7158961810713973,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004706895580277384,
      "loss": 5.7897,
      "mean_token_accuracy": 0.1589108169078827,
      "num_tokens": 39767430.0,
      "step": 19635
    },
    {
      "entropy": 6.100810766220093,
      "epoch": 1.7163331294241022,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047067417735691884,
      "loss": 5.7911,
      "mean_token_accuracy": 0.16221286505460739,
      "num_tokens": 39777270.0,
      "step": 19640
    },
    {
      "entropy": 6.0711774826049805,
      "epoch": 1.7167700777768067,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004706587929329662,
      "loss": 5.8088,
      "mean_token_accuracy": 0.16016956865787507,
      "num_tokens": 39787647.0,
      "step": 19645
    },
    {
      "entropy": 6.07291316986084,
      "epoch": 1.7172070261295116,
      "grad_norm": 1.921875,
      "learning_rate": 0.00047064340475617527,
      "loss": 5.8504,
      "mean_token_accuracy": 0.15247760564088822,
      "num_tokens": 39797315.0,
      "step": 19650
    },
    {
      "entropy": 6.114043951034546,
      "epoch": 1.717643974482216,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004706280128268413,
      "loss": 5.7363,
      "mean_token_accuracy": 0.16428982317447663,
      "num_tokens": 39806625.0,
      "step": 19655
    },
    {
      "entropy": 6.085983753204346,
      "epoch": 1.718080922834921,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004706126171452596,
      "loss": 5.8655,
      "mean_token_accuracy": 0.15948590785264968,
      "num_tokens": 39818113.0,
      "step": 19660
    },
    {
      "entropy": 6.079538297653198,
      "epoch": 1.7185178711876257,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004705972177117254,
      "loss": 5.8459,
      "mean_token_accuracy": 0.15430212914943695,
      "num_tokens": 39827897.0,
      "step": 19665
    },
    {
      "entropy": 6.110194683074951,
      "epoch": 1.7189548195403304,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047058181452653403,
      "loss": 5.6778,
      "mean_token_accuracy": 0.17461654245853425,
      "num_tokens": 39837209.0,
      "step": 19670
    },
    {
      "entropy": 6.115376710891724,
      "epoch": 1.719391767893035,
      "grad_norm": 2.4375,
      "learning_rate": 0.000470566407589981,
      "loss": 5.8055,
      "mean_token_accuracy": 0.16116696149110793,
      "num_tokens": 39848023.0,
      "step": 19675
    },
    {
      "entropy": 6.088941621780395,
      "epoch": 1.7198287162457397,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004705509969023618,
      "loss": 5.859,
      "mean_token_accuracy": 0.15758197903633117,
      "num_tokens": 39857992.0,
      "step": 19680
    },
    {
      "entropy": 6.10357027053833,
      "epoch": 1.7202656645984444,
      "grad_norm": 2.34375,
      "learning_rate": 0.000470535582463972,
      "loss": 5.8544,
      "mean_token_accuracy": 0.15936634242534636,
      "num_tokens": 39867927.0,
      "step": 19685
    },
    {
      "entropy": 6.083196306228638,
      "epoch": 1.7207026129511491,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004705201642751073,
      "loss": 5.7606,
      "mean_token_accuracy": 0.16738648116588592,
      "num_tokens": 39877327.0,
      "step": 19690
    },
    {
      "entropy": 6.135484075546264,
      "epoch": 1.721139561303854,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047050474233606334,
      "loss": 5.7982,
      "mean_token_accuracy": 0.15428519546985625,
      "num_tokens": 39887369.0,
      "step": 19695
    },
    {
      "entropy": 6.169403553009033,
      "epoch": 1.7215765096565585,
      "grad_norm": 2.015625,
      "learning_rate": 0.000470489316647136,
      "loss": 5.8366,
      "mean_token_accuracy": 0.15813133269548416,
      "num_tokens": 39897530.0,
      "step": 19700
    },
    {
      "entropy": 6.11139063835144,
      "epoch": 1.7220134580092634,
      "grad_norm": 2.875,
      "learning_rate": 0.0004704738872086212,
      "loss": 5.7868,
      "mean_token_accuracy": 0.15899520814418794,
      "num_tokens": 39907234.0,
      "step": 19705
    },
    {
      "entropy": 6.1176755905151365,
      "epoch": 1.722450406361968,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004704584540208148,
      "loss": 5.8914,
      "mean_token_accuracy": 0.14965191781520842,
      "num_tokens": 39917940.0,
      "step": 19710
    },
    {
      "entropy": 6.061934232711792,
      "epoch": 1.7228873547146728,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004704430170840127,
      "loss": 5.7535,
      "mean_token_accuracy": 0.17103257924318313,
      "num_tokens": 39927569.0,
      "step": 19715
    },
    {
      "entropy": 6.062588214874268,
      "epoch": 1.7233243030673773,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004704275763985113,
      "loss": 5.8255,
      "mean_token_accuracy": 0.1563766971230507,
      "num_tokens": 39938302.0,
      "step": 19720
    },
    {
      "entropy": 6.180183267593383,
      "epoch": 1.7237612514200822,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004704121319646065,
      "loss": 5.7812,
      "mean_token_accuracy": 0.16535097658634185,
      "num_tokens": 39948133.0,
      "step": 19725
    },
    {
      "entropy": 6.109431028366089,
      "epoch": 1.7241981997727869,
      "grad_norm": 1.828125,
      "learning_rate": 0.00047039668378259455,
      "loss": 5.7704,
      "mean_token_accuracy": 0.16346433609724045,
      "num_tokens": 39958924.0,
      "step": 19730
    },
    {
      "entropy": 6.093697357177734,
      "epoch": 1.7246351481254916,
      "grad_norm": 2.296875,
      "learning_rate": 0.00047038123185277187,
      "loss": 5.7305,
      "mean_token_accuracy": 0.1630522519350052,
      "num_tokens": 39968578.0,
      "step": 19735
    },
    {
      "entropy": 6.050151920318603,
      "epoch": 1.7250720964781963,
      "grad_norm": 3.234375,
      "learning_rate": 0.00047036577617543484,
      "loss": 5.75,
      "mean_token_accuracy": 0.16871791779994966,
      "num_tokens": 39978112.0,
      "step": 19740
    },
    {
      "entropy": 6.04722638130188,
      "epoch": 1.725509044830901,
      "grad_norm": 3.53125,
      "learning_rate": 0.00047035031675087983,
      "loss": 5.843,
      "mean_token_accuracy": 0.15301677137613295,
      "num_tokens": 39988071.0,
      "step": 19745
    },
    {
      "entropy": 6.165513944625855,
      "epoch": 1.7259459931836056,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004703348535794034,
      "loss": 5.8487,
      "mean_token_accuracy": 0.15659181028604507,
      "num_tokens": 39998112.0,
      "step": 19750
    },
    {
      "entropy": 6.10497145652771,
      "epoch": 1.7263829415363103,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00047031938666130206,
      "loss": 5.8109,
      "mean_token_accuracy": 0.1588652789592743,
      "num_tokens": 40008869.0,
      "step": 19755
    },
    {
      "entropy": 6.133491516113281,
      "epoch": 1.7268198898890152,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004703039159968726,
      "loss": 5.8208,
      "mean_token_accuracy": 0.16081639528274536,
      "num_tokens": 40018711.0,
      "step": 19760
    },
    {
      "entropy": 6.127039194107056,
      "epoch": 1.7272568382417197,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047028844158641166,
      "loss": 5.8113,
      "mean_token_accuracy": 0.1599874258041382,
      "num_tokens": 40029209.0,
      "step": 19765
    },
    {
      "entropy": 6.124913072586059,
      "epoch": 1.7276937865944246,
      "grad_norm": 2.046875,
      "learning_rate": 0.000470272963430216,
      "loss": 5.7925,
      "mean_token_accuracy": 0.15902002155780792,
      "num_tokens": 40039071.0,
      "step": 19770
    },
    {
      "entropy": 6.07090311050415,
      "epoch": 1.728130734947129,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004702574815285827,
      "loss": 5.7495,
      "mean_token_accuracy": 0.16340671926736833,
      "num_tokens": 40049765.0,
      "step": 19775
    },
    {
      "entropy": 6.014097356796265,
      "epoch": 1.728567683299834,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004702419958818085,
      "loss": 5.7395,
      "mean_token_accuracy": 0.16334608048200608,
      "num_tokens": 40060020.0,
      "step": 19780
    },
    {
      "entropy": 6.131642818450928,
      "epoch": 1.7290046316525387,
      "grad_norm": 2.625,
      "learning_rate": 0.00047022650649019063,
      "loss": 5.7997,
      "mean_token_accuracy": 0.158540341258049,
      "num_tokens": 40070081.0,
      "step": 19785
    },
    {
      "entropy": 6.169640636444091,
      "epoch": 1.7294415800052434,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000470211013354026,
      "loss": 5.8532,
      "mean_token_accuracy": 0.15497546046972274,
      "num_tokens": 40080532.0,
      "step": 19790
    },
    {
      "entropy": 6.08638219833374,
      "epoch": 1.729878528357948,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004701955164736118,
      "loss": 5.8145,
      "mean_token_accuracy": 0.16103439629077912,
      "num_tokens": 40091288.0,
      "step": 19795
    },
    {
      "entropy": 6.105102586746216,
      "epoch": 1.7303154767106528,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004701800158492454,
      "loss": 5.7804,
      "mean_token_accuracy": 0.15797901302576065,
      "num_tokens": 40100461.0,
      "step": 19800
    },
    {
      "entropy": 6.130492925643921,
      "epoch": 1.7307524250633575,
      "grad_norm": 2.046875,
      "learning_rate": 0.000470164511481224,
      "loss": 5.7543,
      "mean_token_accuracy": 0.16075600385665895,
      "num_tokens": 40109627.0,
      "step": 19805
    },
    {
      "entropy": 6.056947946548462,
      "epoch": 1.7311893734160622,
      "grad_norm": 1.796875,
      "learning_rate": 0.000470149003369845,
      "loss": 5.7948,
      "mean_token_accuracy": 0.1607701726257801,
      "num_tokens": 40119183.0,
      "step": 19810
    },
    {
      "entropy": 6.099366903305054,
      "epoch": 1.731626321768767,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004701334915154058,
      "loss": 5.8212,
      "mean_token_accuracy": 0.1572314754128456,
      "num_tokens": 40129569.0,
      "step": 19815
    },
    {
      "entropy": 6.15400652885437,
      "epoch": 1.7320632701214715,
      "grad_norm": 2.25,
      "learning_rate": 0.00047011797591820404,
      "loss": 5.7387,
      "mean_token_accuracy": 0.161799655854702,
      "num_tokens": 40139618.0,
      "step": 19820
    },
    {
      "entropy": 6.015894174575806,
      "epoch": 1.7325002184741765,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047010245657853726,
      "loss": 5.6981,
      "mean_token_accuracy": 0.1670214295387268,
      "num_tokens": 40149349.0,
      "step": 19825
    },
    {
      "entropy": 6.086278915405273,
      "epoch": 1.732937166826881,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004700869334967031,
      "loss": 5.8248,
      "mean_token_accuracy": 0.15957651287317276,
      "num_tokens": 40159161.0,
      "step": 19830
    },
    {
      "entropy": 6.098497915267944,
      "epoch": 1.7333741151795858,
      "grad_norm": 3.09375,
      "learning_rate": 0.00047007140667299944,
      "loss": 5.7915,
      "mean_token_accuracy": 0.15914907604455947,
      "num_tokens": 40169006.0,
      "step": 19835
    },
    {
      "entropy": 6.0696619033813475,
      "epoch": 1.7338110635322905,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047005587610772384,
      "loss": 5.7878,
      "mean_token_accuracy": 0.15699501782655717,
      "num_tokens": 40179964.0,
      "step": 19840
    },
    {
      "entropy": 6.0839104652404785,
      "epoch": 1.7342480118849952,
      "grad_norm": 1.84375,
      "learning_rate": 0.00047004034180117446,
      "loss": 5.8392,
      "mean_token_accuracy": 0.15348096936941147,
      "num_tokens": 40190598.0,
      "step": 19845
    },
    {
      "entropy": 6.194391250610352,
      "epoch": 1.7346849602377,
      "grad_norm": 2.828125,
      "learning_rate": 0.000470024803753649,
      "loss": 5.8188,
      "mean_token_accuracy": 0.15980052649974824,
      "num_tokens": 40199537.0,
      "step": 19850
    },
    {
      "entropy": 6.089365434646607,
      "epoch": 1.7351219085904046,
      "grad_norm": 2.0,
      "learning_rate": 0.0004700092619654456,
      "loss": 5.7672,
      "mean_token_accuracy": 0.1661269783973694,
      "num_tokens": 40209167.0,
      "step": 19855
    },
    {
      "entropy": 6.115077066421509,
      "epoch": 1.7355588569431093,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004699937164368625,
      "loss": 5.7892,
      "mean_token_accuracy": 0.16164094358682632,
      "num_tokens": 40218738.0,
      "step": 19860
    },
    {
      "entropy": 6.13235993385315,
      "epoch": 1.735995805295814,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004699781671681977,
      "loss": 5.7523,
      "mean_token_accuracy": 0.1674604207277298,
      "num_tokens": 40227897.0,
      "step": 19865
    },
    {
      "entropy": 6.098686265945434,
      "epoch": 1.736432753648519,
      "grad_norm": 2.921875,
      "learning_rate": 0.00046996261415974946,
      "loss": 5.8531,
      "mean_token_accuracy": 0.15154337733983994,
      "num_tokens": 40238870.0,
      "step": 19870
    },
    {
      "entropy": 6.06974081993103,
      "epoch": 1.7368697020012234,
      "grad_norm": 2.265625,
      "learning_rate": 0.00046994705741181615,
      "loss": 5.7654,
      "mean_token_accuracy": 0.1588371604681015,
      "num_tokens": 40247832.0,
      "step": 19875
    },
    {
      "entropy": 6.133505582809448,
      "epoch": 1.7373066503539283,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004699314969246961,
      "loss": 5.7738,
      "mean_token_accuracy": 0.16144887804985047,
      "num_tokens": 40258696.0,
      "step": 19880
    },
    {
      "entropy": 6.135479545593261,
      "epoch": 1.7377435987066328,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046991593269868776,
      "loss": 5.7799,
      "mean_token_accuracy": 0.16415976583957673,
      "num_tokens": 40267767.0,
      "step": 19885
    },
    {
      "entropy": 6.124272394180298,
      "epoch": 1.7381805470593377,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004699003647340898,
      "loss": 5.8364,
      "mean_token_accuracy": 0.1598043531179428,
      "num_tokens": 40277813.0,
      "step": 19890
    },
    {
      "entropy": 6.1051390171051025,
      "epoch": 1.7386174954120421,
      "grad_norm": 3.40625,
      "learning_rate": 0.00046988479303120066,
      "loss": 5.8418,
      "mean_token_accuracy": 0.16099047660827637,
      "num_tokens": 40289031.0,
      "step": 19895
    },
    {
      "entropy": 6.124701452255249,
      "epoch": 1.739054443764747,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046986921759031906,
      "loss": 5.7612,
      "mean_token_accuracy": 0.1616186574101448,
      "num_tokens": 40299428.0,
      "step": 19900
    },
    {
      "entropy": 6.097035360336304,
      "epoch": 1.7394913921174517,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004698536384117438,
      "loss": 5.7642,
      "mean_token_accuracy": 0.15883323699235916,
      "num_tokens": 40308996.0,
      "step": 19905
    },
    {
      "entropy": 6.114970874786377,
      "epoch": 1.7399283404701564,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046983805549577365,
      "loss": 5.8343,
      "mean_token_accuracy": 0.16042929142713547,
      "num_tokens": 40319753.0,
      "step": 19910
    },
    {
      "entropy": 6.181920528411865,
      "epoch": 1.7403652888228611,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004698224688427074,
      "loss": 5.9631,
      "mean_token_accuracy": 0.14773821085691452,
      "num_tokens": 40330729.0,
      "step": 19915
    },
    {
      "entropy": 6.173805475234985,
      "epoch": 1.7408022371755658,
      "grad_norm": 8.0,
      "learning_rate": 0.0004698068784528442,
      "loss": 5.8712,
      "mean_token_accuracy": 0.15089573711156845,
      "num_tokens": 40340246.0,
      "step": 19920
    },
    {
      "entropy": 6.140525674819946,
      "epoch": 1.7412391855282705,
      "grad_norm": 1.8125,
      "learning_rate": 0.000469791284326483,
      "loss": 5.7339,
      "mean_token_accuracy": 0.15793695002794267,
      "num_tokens": 40350099.0,
      "step": 19925
    },
    {
      "entropy": 6.124837207794189,
      "epoch": 1.7416761338809752,
      "grad_norm": 2.09375,
      "learning_rate": 0.00046977568646392284,
      "loss": 5.8228,
      "mean_token_accuracy": 0.16175366342067718,
      "num_tokens": 40360825.0,
      "step": 19930
    },
    {
      "entropy": 6.138683843612671,
      "epoch": 1.74211308223368,
      "grad_norm": 2.671875,
      "learning_rate": 0.000469760084865463,
      "loss": 5.8945,
      "mean_token_accuracy": 0.14274640902876853,
      "num_tokens": 40371025.0,
      "step": 19935
    },
    {
      "entropy": 6.1342261791229244,
      "epoch": 1.7425500305863846,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004697444795314027,
      "loss": 5.745,
      "mean_token_accuracy": 0.16824313402175903,
      "num_tokens": 40380338.0,
      "step": 19940
    },
    {
      "entropy": 6.112724494934082,
      "epoch": 1.7429869789390895,
      "grad_norm": 2.953125,
      "learning_rate": 0.00046972887046204114,
      "loss": 5.8305,
      "mean_token_accuracy": 0.1612398386001587,
      "num_tokens": 40391814.0,
      "step": 19945
    },
    {
      "entropy": 6.1076788902282715,
      "epoch": 1.743423927291794,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046971325765767793,
      "loss": 5.801,
      "mean_token_accuracy": 0.15103936195373535,
      "num_tokens": 40402013.0,
      "step": 19950
    },
    {
      "entropy": 6.149267053604126,
      "epoch": 1.7438608756444989,
      "grad_norm": 2.90625,
      "learning_rate": 0.00046969764111861234,
      "loss": 5.7764,
      "mean_token_accuracy": 0.15401173681020736,
      "num_tokens": 40412205.0,
      "step": 19955
    },
    {
      "entropy": 6.018934965133667,
      "epoch": 1.7442978239972036,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046968202084514405,
      "loss": 5.7452,
      "mean_token_accuracy": 0.17005741745233535,
      "num_tokens": 40423212.0,
      "step": 19960
    },
    {
      "entropy": 6.087668323516846,
      "epoch": 1.7447347723499083,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004696663968375725,
      "loss": 5.7369,
      "mean_token_accuracy": 0.15847148001194,
      "num_tokens": 40433272.0,
      "step": 19965
    },
    {
      "entropy": 6.170419836044312,
      "epoch": 1.745171720702613,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004696507690961975,
      "loss": 5.889,
      "mean_token_accuracy": 0.15732436180114745,
      "num_tokens": 40443931.0,
      "step": 19970
    },
    {
      "entropy": 6.11259183883667,
      "epoch": 1.7456086690553176,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004696351376213188,
      "loss": 5.7941,
      "mean_token_accuracy": 0.16061891317367555,
      "num_tokens": 40453008.0,
      "step": 19975
    },
    {
      "entropy": 6.029903888702393,
      "epoch": 1.7460456174080223,
      "grad_norm": 2.078125,
      "learning_rate": 0.000469619502413236,
      "loss": 5.8557,
      "mean_token_accuracy": 0.15949919819831848,
      "num_tokens": 40463081.0,
      "step": 19980
    },
    {
      "entropy": 6.105974817276001,
      "epoch": 1.746482565760727,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004696038634722493,
      "loss": 5.6849,
      "mean_token_accuracy": 0.16437639892101288,
      "num_tokens": 40472390.0,
      "step": 19985
    },
    {
      "entropy": 6.113411855697632,
      "epoch": 1.746919514113432,
      "grad_norm": 4.125,
      "learning_rate": 0.00046958822079865855,
      "loss": 5.8554,
      "mean_token_accuracy": 0.1559371531009674,
      "num_tokens": 40483068.0,
      "step": 19990
    },
    {
      "entropy": 6.165650415420532,
      "epoch": 1.7473564624661364,
      "grad_norm": 8.625,
      "learning_rate": 0.0004695725743927637,
      "loss": 5.8203,
      "mean_token_accuracy": 0.16495466828346253,
      "num_tokens": 40493125.0,
      "step": 19995
    },
    {
      "entropy": 6.093226909637451,
      "epoch": 1.7477934108188413,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046955692425486496,
      "loss": 5.8068,
      "mean_token_accuracy": 0.1584664985537529,
      "num_tokens": 40502618.0,
      "step": 20000
    },
    {
      "entropy": 6.1111351490020756,
      "epoch": 1.7482303591715458,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004695412703852624,
      "loss": 5.8404,
      "mean_token_accuracy": 0.16017089933156967,
      "num_tokens": 40512878.0,
      "step": 20005
    },
    {
      "entropy": 6.048735857009888,
      "epoch": 1.7486673075242507,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004695256127842563,
      "loss": 5.7857,
      "mean_token_accuracy": 0.15966059416532516,
      "num_tokens": 40521983.0,
      "step": 20010
    },
    {
      "entropy": 6.018918180465699,
      "epoch": 1.7491042558769554,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046950995145214716,
      "loss": 5.7894,
      "mean_token_accuracy": 0.1607269287109375,
      "num_tokens": 40533126.0,
      "step": 20015
    },
    {
      "entropy": 6.080697250366211,
      "epoch": 1.74954120422966,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004694942863892352,
      "loss": 5.7748,
      "mean_token_accuracy": 0.16380762457847595,
      "num_tokens": 40543321.0,
      "step": 20020
    },
    {
      "entropy": 6.116272592544556,
      "epoch": 1.7499781525823648,
      "grad_norm": 5.4375,
      "learning_rate": 0.0004694786175958209,
      "loss": 5.8036,
      "mean_token_accuracy": 0.162904292345047,
      "num_tokens": 40554208.0,
      "step": 20025
    },
    {
      "entropy": 6.147489881515503,
      "epoch": 1.7504151009350695,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00046946294507220477,
      "loss": 5.7652,
      "mean_token_accuracy": 0.15288180112838745,
      "num_tokens": 40563365.0,
      "step": 20030
    },
    {
      "entropy": 5.9902637004852295,
      "epoch": 1.7508520492877742,
      "grad_norm": 4.375,
      "learning_rate": 0.0004694472688186874,
      "loss": 5.6902,
      "mean_token_accuracy": 0.16462101489305497,
      "num_tokens": 40573765.0,
      "step": 20035
    },
    {
      "entropy": 6.040267848968506,
      "epoch": 1.7512889976404789,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004694315888355697,
      "loss": 5.7955,
      "mean_token_accuracy": 0.16260012239217758,
      "num_tokens": 40583786.0,
      "step": 20040
    },
    {
      "entropy": 6.088585710525512,
      "epoch": 1.7517259459931838,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046941590512315215,
      "loss": 5.8312,
      "mean_token_accuracy": 0.15657771080732347,
      "num_tokens": 40593686.0,
      "step": 20045
    },
    {
      "entropy": 6.089376735687256,
      "epoch": 1.7521628943458882,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004694002176817356,
      "loss": 5.7958,
      "mean_token_accuracy": 0.15806715488433837,
      "num_tokens": 40604326.0,
      "step": 20050
    },
    {
      "entropy": 6.134919023513794,
      "epoch": 1.7525998426985931,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004693845265116211,
      "loss": 5.7826,
      "mean_token_accuracy": 0.15690571218729019,
      "num_tokens": 40613683.0,
      "step": 20055
    },
    {
      "entropy": 6.098470735549927,
      "epoch": 1.7530367910512976,
      "grad_norm": 2.5,
      "learning_rate": 0.00046936883161310944,
      "loss": 5.8124,
      "mean_token_accuracy": 0.16023394614458084,
      "num_tokens": 40623675.0,
      "step": 20060
    },
    {
      "entropy": 6.1270934581756595,
      "epoch": 1.7534737394040025,
      "grad_norm": 2.703125,
      "learning_rate": 0.00046935313298650185,
      "loss": 5.8249,
      "mean_token_accuracy": 0.15611762255430223,
      "num_tokens": 40634006.0,
      "step": 20065
    },
    {
      "entropy": 6.041053342819214,
      "epoch": 1.753910687756707,
      "grad_norm": 3.34375,
      "learning_rate": 0.00046933743063209927,
      "loss": 5.697,
      "mean_token_accuracy": 0.17064322680234909,
      "num_tokens": 40643175.0,
      "step": 20070
    },
    {
      "entropy": 6.15368161201477,
      "epoch": 1.754347636109412,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046932172455020296,
      "loss": 5.8431,
      "mean_token_accuracy": 0.14859526455402375,
      "num_tokens": 40653447.0,
      "step": 20075
    },
    {
      "entropy": 6.164422464370728,
      "epoch": 1.7547845844621166,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004693060147411141,
      "loss": 5.7652,
      "mean_token_accuracy": 0.1601276680827141,
      "num_tokens": 40663499.0,
      "step": 20080
    },
    {
      "entropy": 6.063589191436767,
      "epoch": 1.7552215328148213,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004692903012051341,
      "loss": 5.7773,
      "mean_token_accuracy": 0.16496762931346892,
      "num_tokens": 40672943.0,
      "step": 20085
    },
    {
      "entropy": 6.052312898635864,
      "epoch": 1.755658481167526,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004692745839425643,
      "loss": 5.8293,
      "mean_token_accuracy": 0.1603731095790863,
      "num_tokens": 40683659.0,
      "step": 20090
    },
    {
      "entropy": 6.176554393768311,
      "epoch": 1.7560954295202307,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004692588629537062,
      "loss": 5.8437,
      "mean_token_accuracy": 0.15700934082269669,
      "num_tokens": 40693829.0,
      "step": 20095
    },
    {
      "entropy": 6.1258251667022705,
      "epoch": 1.7565323778729354,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004692431382388613,
      "loss": 5.8517,
      "mean_token_accuracy": 0.15558656603097915,
      "num_tokens": 40704717.0,
      "step": 20100
    },
    {
      "entropy": 6.1134437084197994,
      "epoch": 1.75696932622564,
      "grad_norm": 4.65625,
      "learning_rate": 0.00046922740979833135,
      "loss": 5.8156,
      "mean_token_accuracy": 0.15428268760442734,
      "num_tokens": 40714970.0,
      "step": 20105
    },
    {
      "entropy": 6.109387731552124,
      "epoch": 1.757406274578345,
      "grad_norm": 3.21875,
      "learning_rate": 0.00046921167763241777,
      "loss": 5.7751,
      "mean_token_accuracy": 0.16221875548362732,
      "num_tokens": 40723664.0,
      "step": 20110
    },
    {
      "entropy": 6.10299654006958,
      "epoch": 1.7578432229310494,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046919594174142247,
      "loss": 5.7692,
      "mean_token_accuracy": 0.15864422172307968,
      "num_tokens": 40733735.0,
      "step": 20115
    },
    {
      "entropy": 6.117214059829712,
      "epoch": 1.7582801712837544,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046918020212564725,
      "loss": 5.8432,
      "mean_token_accuracy": 0.15609552711248398,
      "num_tokens": 40743699.0,
      "step": 20120
    },
    {
      "entropy": 6.076550769805908,
      "epoch": 1.7587171196364588,
      "grad_norm": 2.890625,
      "learning_rate": 0.000469164458785394,
      "loss": 5.8734,
      "mean_token_accuracy": 0.1531803399324417,
      "num_tokens": 40754553.0,
      "step": 20125
    },
    {
      "entropy": 6.1531970500946045,
      "epoch": 1.7591540679891637,
      "grad_norm": 2.59375,
      "learning_rate": 0.00046914871172096473,
      "loss": 5.832,
      "mean_token_accuracy": 0.15300681218504905,
      "num_tokens": 40764660.0,
      "step": 20130
    },
    {
      "entropy": 6.124754905700684,
      "epoch": 1.7595910163418684,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004691329609326614,
      "loss": 5.6663,
      "mean_token_accuracy": 0.17002324759960175,
      "num_tokens": 40774237.0,
      "step": 20135
    },
    {
      "entropy": 6.064556074142456,
      "epoch": 1.7600279646945731,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004691172064207862,
      "loss": 5.7629,
      "mean_token_accuracy": 0.1665944427251816,
      "num_tokens": 40783608.0,
      "step": 20140
    },
    {
      "entropy": 6.151745176315307,
      "epoch": 1.7604649130472778,
      "grad_norm": 2.5,
      "learning_rate": 0.0004691014481856412,
      "loss": 5.8235,
      "mean_token_accuracy": 0.15317616909742354,
      "num_tokens": 40793987.0,
      "step": 20145
    },
    {
      "entropy": 6.174080038070679,
      "epoch": 1.7609018613999825,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046908568622752874,
      "loss": 5.8482,
      "mean_token_accuracy": 0.1541568398475647,
      "num_tokens": 40805600.0,
      "step": 20150
    },
    {
      "entropy": 6.106844186782837,
      "epoch": 1.7613388097526872,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046906992054675115,
      "loss": 5.805,
      "mean_token_accuracy": 0.1637753963470459,
      "num_tokens": 40815252.0,
      "step": 20155
    },
    {
      "entropy": 6.04176287651062,
      "epoch": 1.7617757581053919,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004690541511436107,
      "loss": 5.7887,
      "mean_token_accuracy": 0.16401199549436568,
      "num_tokens": 40825945.0,
      "step": 20160
    },
    {
      "entropy": 6.106495428085327,
      "epoch": 1.7622127064580968,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004690383780184101,
      "loss": 5.8429,
      "mean_token_accuracy": 0.15902678966522216,
      "num_tokens": 40835811.0,
      "step": 20165
    },
    {
      "entropy": 6.141741943359375,
      "epoch": 1.7626496548108013,
      "grad_norm": 1.890625,
      "learning_rate": 0.00046902260117145163,
      "loss": 5.8062,
      "mean_token_accuracy": 0.15804900377988815,
      "num_tokens": 40845088.0,
      "step": 20170
    },
    {
      "entropy": 6.0984584331512455,
      "epoch": 1.7630866031635062,
      "grad_norm": 1.9375,
      "learning_rate": 0.000469006820603038,
      "loss": 5.8553,
      "mean_token_accuracy": 0.157698193192482,
      "num_tokens": 40854703.0,
      "step": 20175
    },
    {
      "entropy": 6.179224252700806,
      "epoch": 1.7635235515162107,
      "grad_norm": 2.34375,
      "learning_rate": 0.00046899103631347196,
      "loss": 5.8021,
      "mean_token_accuracy": 0.15572601705789565,
      "num_tokens": 40865056.0,
      "step": 20180
    },
    {
      "entropy": 6.078614711761475,
      "epoch": 1.7639604998689156,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004689752483030561,
      "loss": 5.7023,
      "mean_token_accuracy": 0.16415783613920212,
      "num_tokens": 40874355.0,
      "step": 20185
    },
    {
      "entropy": 6.075523948669433,
      "epoch": 1.7643974482216203,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004689594565720934,
      "loss": 5.7778,
      "mean_token_accuracy": 0.1595476284623146,
      "num_tokens": 40884329.0,
      "step": 20190
    },
    {
      "entropy": 6.078005838394165,
      "epoch": 1.764834396574325,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004689436611208867,
      "loss": 5.7986,
      "mean_token_accuracy": 0.1628919407725334,
      "num_tokens": 40894702.0,
      "step": 20195
    },
    {
      "entropy": 6.1492271900177,
      "epoch": 1.7652713449270296,
      "grad_norm": 3.1875,
      "learning_rate": 0.00046892786194973897,
      "loss": 5.7791,
      "mean_token_accuracy": 0.15836129039525987,
      "num_tokens": 40905467.0,
      "step": 20200
    },
    {
      "entropy": 6.12502088546753,
      "epoch": 1.7657082932797343,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004689120590589532,
      "loss": 5.83,
      "mean_token_accuracy": 0.15785814374685286,
      "num_tokens": 40915915.0,
      "step": 20205
    },
    {
      "entropy": 6.054952096939087,
      "epoch": 1.766145241632439,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004688962524488325,
      "loss": 5.7665,
      "mean_token_accuracy": 0.16187303513288498,
      "num_tokens": 40925941.0,
      "step": 20210
    },
    {
      "entropy": 6.108825159072876,
      "epoch": 1.7665821899851437,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004688804421196801,
      "loss": 5.8356,
      "mean_token_accuracy": 0.15991260409355162,
      "num_tokens": 40935831.0,
      "step": 20215
    },
    {
      "entropy": 6.034203100204468,
      "epoch": 1.7670191383378486,
      "grad_norm": 1.796875,
      "learning_rate": 0.00046886462807179926,
      "loss": 5.7838,
      "mean_token_accuracy": 0.16537935733795167,
      "num_tokens": 40945341.0,
      "step": 20220
    },
    {
      "entropy": 6.136110496520996,
      "epoch": 1.767456086690553,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004688488103054931,
      "loss": 5.8317,
      "mean_token_accuracy": 0.1633661061525345,
      "num_tokens": 40955759.0,
      "step": 20225
    },
    {
      "entropy": 6.170142745971679,
      "epoch": 1.767893035043258,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004688329888210654,
      "loss": 5.7527,
      "mean_token_accuracy": 0.1559879496693611,
      "num_tokens": 40966184.0,
      "step": 20230
    },
    {
      "entropy": 6.114583969116211,
      "epoch": 1.7683299833959625,
      "grad_norm": 1.75,
      "learning_rate": 0.0004688171636188194,
      "loss": 5.8194,
      "mean_token_accuracy": 0.16104481220245362,
      "num_tokens": 40975797.0,
      "step": 20235
    },
    {
      "entropy": 6.142330169677734,
      "epoch": 1.7687669317486674,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004688013346990585,
      "loss": 5.8128,
      "mean_token_accuracy": 0.1605291798710823,
      "num_tokens": 40985830.0,
      "step": 20240
    },
    {
      "entropy": 6.036077928543091,
      "epoch": 1.7692038801013719,
      "grad_norm": 1.84375,
      "learning_rate": 0.00046878550206208656,
      "loss": 5.7106,
      "mean_token_accuracy": 0.16255198419094086,
      "num_tokens": 40995692.0,
      "step": 20245
    },
    {
      "entropy": 6.117742252349854,
      "epoch": 1.7696408284540768,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004687696657082071,
      "loss": 5.8295,
      "mean_token_accuracy": 0.1577209010720253,
      "num_tokens": 41005506.0,
      "step": 20250
    },
    {
      "entropy": 6.119452285766601,
      "epoch": 1.7700777768067815,
      "grad_norm": 2.640625,
      "learning_rate": 0.00046875382563772396,
      "loss": 5.7293,
      "mean_token_accuracy": 0.16344009190797806,
      "num_tokens": 41014944.0,
      "step": 20255
    },
    {
      "entropy": 6.145107364654541,
      "epoch": 1.7705147251594862,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046873798185094093,
      "loss": 5.8847,
      "mean_token_accuracy": 0.15175872445106506,
      "num_tokens": 41025424.0,
      "step": 20260
    },
    {
      "entropy": 6.010427761077881,
      "epoch": 1.7709516735121908,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004687221343481618,
      "loss": 5.7275,
      "mean_token_accuracy": 0.1676472619175911,
      "num_tokens": 41035817.0,
      "step": 20265
    },
    {
      "entropy": 6.0415114402771,
      "epoch": 1.7713886218648955,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046870628312969064,
      "loss": 5.7668,
      "mean_token_accuracy": 0.16444457173347474,
      "num_tokens": 41045730.0,
      "step": 20270
    },
    {
      "entropy": 6.190864324569702,
      "epoch": 1.7718255702176002,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004686904281958315,
      "loss": 5.8781,
      "mean_token_accuracy": 0.146004269272089,
      "num_tokens": 41057921.0,
      "step": 20275
    },
    {
      "entropy": 6.196317291259765,
      "epoch": 1.772262518570305,
      "grad_norm": 2.375,
      "learning_rate": 0.00046867456954688847,
      "loss": 5.87,
      "mean_token_accuracy": 0.16148425191640853,
      "num_tokens": 41068879.0,
      "step": 20280
    },
    {
      "entropy": 6.098105239868164,
      "epoch": 1.7726994669230098,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004686587071831657,
      "loss": 5.8766,
      "mean_token_accuracy": 0.15764899104833602,
      "num_tokens": 41079443.0,
      "step": 20285
    },
    {
      "entropy": 6.100146055221558,
      "epoch": 1.7731364152757143,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004686428411049675,
      "loss": 5.8421,
      "mean_token_accuracy": 0.15223202258348464,
      "num_tokens": 41089932.0,
      "step": 20290
    },
    {
      "entropy": 6.190094470977783,
      "epoch": 1.7735733636284192,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004686269713125981,
      "loss": 5.8346,
      "mean_token_accuracy": 0.15577563941478728,
      "num_tokens": 41100412.0,
      "step": 20295
    },
    {
      "entropy": 6.152592897415161,
      "epoch": 1.7740103119811237,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004686110978063619,
      "loss": 5.8058,
      "mean_token_accuracy": 0.15591662526130676,
      "num_tokens": 41110830.0,
      "step": 20300
    },
    {
      "entropy": 6.08135666847229,
      "epoch": 1.7744472603338286,
      "grad_norm": 2.5,
      "learning_rate": 0.0004685952205865634,
      "loss": 5.7821,
      "mean_token_accuracy": 0.15901214629411697,
      "num_tokens": 41120662.0,
      "step": 20305
    },
    {
      "entropy": 6.03677077293396,
      "epoch": 1.7748842086865333,
      "grad_norm": 3.375,
      "learning_rate": 0.00046857933965350714,
      "loss": 5.7475,
      "mean_token_accuracy": 0.16290881484746933,
      "num_tokens": 41130523.0,
      "step": 20310
    },
    {
      "entropy": 6.153215599060059,
      "epoch": 1.775321157039238,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004685634550074977,
      "loss": 5.8722,
      "mean_token_accuracy": 0.1608460485935211,
      "num_tokens": 41141659.0,
      "step": 20315
    },
    {
      "entropy": 6.190534353256226,
      "epoch": 1.7757581053919427,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004685475666488398,
      "loss": 5.9061,
      "mean_token_accuracy": 0.1462700441479683,
      "num_tokens": 41151038.0,
      "step": 20320
    },
    {
      "entropy": 6.113929414749146,
      "epoch": 1.7761950537446474,
      "grad_norm": 2.25,
      "learning_rate": 0.00046853167457783815,
      "loss": 5.8543,
      "mean_token_accuracy": 0.16006929129362107,
      "num_tokens": 41161888.0,
      "step": 20325
    },
    {
      "entropy": 6.189467811584473,
      "epoch": 1.776632002097352,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004685157787947975,
      "loss": 5.8947,
      "mean_token_accuracy": 0.1520603619515896,
      "num_tokens": 41172722.0,
      "step": 20330
    },
    {
      "entropy": 6.119915103912353,
      "epoch": 1.7770689504500568,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004684998793000229,
      "loss": 5.7469,
      "mean_token_accuracy": 0.16104741841554643,
      "num_tokens": 41181716.0,
      "step": 20335
    },
    {
      "entropy": 6.0875872611999515,
      "epoch": 1.7775058988027617,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004684839760938193,
      "loss": 5.849,
      "mean_token_accuracy": 0.15372057110071183,
      "num_tokens": 41192903.0,
      "step": 20340
    },
    {
      "entropy": 6.08648681640625,
      "epoch": 1.7779428471554661,
      "grad_norm": 3.03125,
      "learning_rate": 0.00046846806917649145,
      "loss": 5.8064,
      "mean_token_accuracy": 0.1587863400578499,
      "num_tokens": 41203283.0,
      "step": 20345
    },
    {
      "entropy": 6.144145202636719,
      "epoch": 1.778379795508171,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004684521585483447,
      "loss": 5.8143,
      "mean_token_accuracy": 0.1550808683037758,
      "num_tokens": 41213615.0,
      "step": 20350
    },
    {
      "entropy": 6.129676866531372,
      "epoch": 1.7788167438608755,
      "grad_norm": 2.53125,
      "learning_rate": 0.00046843624420968435,
      "loss": 5.7218,
      "mean_token_accuracy": 0.1703510418534279,
      "num_tokens": 41222705.0,
      "step": 20355
    },
    {
      "entropy": 6.0233738899230955,
      "epoch": 1.7792536922135804,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004684203261608153,
      "loss": 5.7619,
      "mean_token_accuracy": 0.1642073407769203,
      "num_tokens": 41233201.0,
      "step": 20360
    },
    {
      "entropy": 6.09271092414856,
      "epoch": 1.7796906405662851,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004684044044020431,
      "loss": 5.8156,
      "mean_token_accuracy": 0.1556283265352249,
      "num_tokens": 41243088.0,
      "step": 20365
    },
    {
      "entropy": 6.176855421066284,
      "epoch": 1.7801275889189898,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004683884789336731,
      "loss": 5.8311,
      "mean_token_accuracy": 0.1534987837076187,
      "num_tokens": 41253009.0,
      "step": 20370
    },
    {
      "entropy": 6.109649515151977,
      "epoch": 1.7805645372716945,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046837254975601073,
      "loss": 5.8156,
      "mean_token_accuracy": 0.16142655462026595,
      "num_tokens": 41262209.0,
      "step": 20375
    },
    {
      "entropy": 6.1149022579193115,
      "epoch": 1.7810014856243992,
      "grad_norm": 2.5,
      "learning_rate": 0.00046835661686936154,
      "loss": 5.8162,
      "mean_token_accuracy": 0.16219808012247086,
      "num_tokens": 41272762.0,
      "step": 20380
    },
    {
      "entropy": 6.1016394138336185,
      "epoch": 1.7814384339771039,
      "grad_norm": 2.671875,
      "learning_rate": 0.00046834068027403115,
      "loss": 5.8904,
      "mean_token_accuracy": 0.1537313625216484,
      "num_tokens": 41284306.0,
      "step": 20385
    },
    {
      "entropy": 6.140647554397583,
      "epoch": 1.7818753823298086,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046832473997032514,
      "loss": 5.8428,
      "mean_token_accuracy": 0.15506634265184402,
      "num_tokens": 41295211.0,
      "step": 20390
    },
    {
      "entropy": 6.09630970954895,
      "epoch": 1.7823123306825135,
      "grad_norm": 2.125,
      "learning_rate": 0.0004683087959585493,
      "loss": 5.6874,
      "mean_token_accuracy": 0.1630358673632145,
      "num_tokens": 41303773.0,
      "step": 20395
    },
    {
      "entropy": 6.034410905838013,
      "epoch": 1.782749279035218,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004682928482390095,
      "loss": 5.7906,
      "mean_token_accuracy": 0.16355063617229462,
      "num_tokens": 41313536.0,
      "step": 20400
    },
    {
      "entropy": 6.161115884780884,
      "epoch": 1.7831862273879229,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046827689681201164,
      "loss": 5.8195,
      "mean_token_accuracy": 0.1601507380604744,
      "num_tokens": 41323831.0,
      "step": 20405
    },
    {
      "entropy": 6.147857761383056,
      "epoch": 1.7836231757406273,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004682609416778615,
      "loss": 5.808,
      "mean_token_accuracy": 0.16318029016256333,
      "num_tokens": 41334295.0,
      "step": 20410
    },
    {
      "entropy": 6.079022693634033,
      "epoch": 1.7840601240933323,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004682449828368653,
      "loss": 5.7991,
      "mean_token_accuracy": 0.15694925785064698,
      "num_tokens": 41344672.0,
      "step": 20415
    },
    {
      "entropy": 6.1396078109741214,
      "epoch": 1.7844970724460367,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004682290202893291,
      "loss": 5.8472,
      "mean_token_accuracy": 0.15283788591623307,
      "num_tokens": 41355108.0,
      "step": 20420
    },
    {
      "entropy": 6.275498867034912,
      "epoch": 1.7849340207987416,
      "grad_norm": 1.875,
      "learning_rate": 0.00046821305403555905,
      "loss": 5.9388,
      "mean_token_accuracy": 0.1496388629078865,
      "num_tokens": 41366277.0,
      "step": 20425
    },
    {
      "entropy": 6.143874597549439,
      "epoch": 1.7853709691514463,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00046819708407586135,
      "loss": 5.7921,
      "mean_token_accuracy": 0.16322183907032012,
      "num_tokens": 41374757.0,
      "step": 20430
    },
    {
      "entropy": 6.081591987609864,
      "epoch": 1.785807917504151,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004681811104105423,
      "loss": 5.7979,
      "mean_token_accuracy": 0.16126659363508225,
      "num_tokens": 41384883.0,
      "step": 20435
    },
    {
      "entropy": 6.1282867908477785,
      "epoch": 1.7862448658568557,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004681651330399084,
      "loss": 5.8241,
      "mean_token_accuracy": 0.15773673504590988,
      "num_tokens": 41396492.0,
      "step": 20440
    },
    {
      "entropy": 6.1116548538208,
      "epoch": 1.7866818142095604,
      "grad_norm": 2.078125,
      "learning_rate": 0.000468149151964266,
      "loss": 5.7903,
      "mean_token_accuracy": 0.15950769037008286,
      "num_tokens": 41406217.0,
      "step": 20445
    },
    {
      "entropy": 6.08352632522583,
      "epoch": 1.787118762562265,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004681331671839217,
      "loss": 5.716,
      "mean_token_accuracy": 0.16734521239995956,
      "num_tokens": 41415895.0,
      "step": 20450
    },
    {
      "entropy": 6.1378966808319095,
      "epoch": 1.7875557109149698,
      "grad_norm": 2.546875,
      "learning_rate": 0.00046811717869918196,
      "loss": 5.8459,
      "mean_token_accuracy": 0.15817746967077256,
      "num_tokens": 41426021.0,
      "step": 20455
    },
    {
      "entropy": 6.145797538757324,
      "epoch": 1.7879926592676747,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004681011865103536,
      "loss": 5.7852,
      "mean_token_accuracy": 0.15867676734924316,
      "num_tokens": 41434656.0,
      "step": 20460
    },
    {
      "entropy": 6.0983133792877195,
      "epoch": 1.7884296076203792,
      "grad_norm": 2.734375,
      "learning_rate": 0.00046808519061774324,
      "loss": 5.8081,
      "mean_token_accuracy": 0.1602679818868637,
      "num_tokens": 41443940.0,
      "step": 20465
    },
    {
      "entropy": 6.095771837234497,
      "epoch": 1.788866555973084,
      "grad_norm": 3.015625,
      "learning_rate": 0.00046806919102165785,
      "loss": 5.8759,
      "mean_token_accuracy": 0.15755071491003036,
      "num_tokens": 41454907.0,
      "step": 20470
    },
    {
      "entropy": 6.12061333656311,
      "epoch": 1.7893035043257886,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004680531877224042,
      "loss": 5.7317,
      "mean_token_accuracy": 0.16473085135221482,
      "num_tokens": 41464248.0,
      "step": 20475
    },
    {
      "entropy": 6.107315635681152,
      "epoch": 1.7897404526784935,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004680371807202892,
      "loss": 5.7533,
      "mean_token_accuracy": 0.16742003262042998,
      "num_tokens": 41475192.0,
      "step": 20480
    },
    {
      "entropy": 6.112410163879394,
      "epoch": 1.7901774010311982,
      "grad_norm": 3.125,
      "learning_rate": 0.00046802117001561996,
      "loss": 5.8535,
      "mean_token_accuracy": 0.14819685891270637,
      "num_tokens": 41484753.0,
      "step": 20485
    },
    {
      "entropy": 6.100326347351074,
      "epoch": 1.7906143493839028,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046800515560870346,
      "loss": 5.9182,
      "mean_token_accuracy": 0.14722870737314225,
      "num_tokens": 41495885.0,
      "step": 20490
    },
    {
      "entropy": 6.2265296459198,
      "epoch": 1.7910512977366075,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00046798913749984694,
      "loss": 5.8236,
      "mean_token_accuracy": 0.1564646005630493,
      "num_tokens": 41506330.0,
      "step": 20495
    },
    {
      "entropy": 6.090691995620728,
      "epoch": 1.7914882460893122,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004679731156893577,
      "loss": 5.8006,
      "mean_token_accuracy": 0.16281561702489852,
      "num_tokens": 41515850.0,
      "step": 20500
    },
    {
      "entropy": 6.097841072082519,
      "epoch": 1.791925194442017,
      "grad_norm": 7.53125,
      "learning_rate": 0.000467957090177543,
      "loss": 5.8068,
      "mean_token_accuracy": 0.16149195283651352,
      "num_tokens": 41525910.0,
      "step": 20505
    },
    {
      "entropy": 6.12894926071167,
      "epoch": 1.7923621427947216,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046794106096471015,
      "loss": 5.7294,
      "mean_token_accuracy": 0.16208693981170655,
      "num_tokens": 41536417.0,
      "step": 20510
    },
    {
      "entropy": 6.085343742370606,
      "epoch": 1.7927990911474265,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004679250280511667,
      "loss": 5.8062,
      "mean_token_accuracy": 0.15372299104928971,
      "num_tokens": 41547215.0,
      "step": 20515
    },
    {
      "entropy": 6.101957893371582,
      "epoch": 1.793236039500131,
      "grad_norm": 2.0,
      "learning_rate": 0.0004679089914372201,
      "loss": 5.8324,
      "mean_token_accuracy": 0.1555679500102997,
      "num_tokens": 41556036.0,
      "step": 20520
    },
    {
      "entropy": 6.112474632263184,
      "epoch": 1.793672987852836,
      "grad_norm": 1.890625,
      "learning_rate": 0.00046789295112317796,
      "loss": 5.8195,
      "mean_token_accuracy": 0.16118495017290116,
      "num_tokens": 41565534.0,
      "step": 20525
    },
    {
      "entropy": 6.118927335739135,
      "epoch": 1.7941099362055404,
      "grad_norm": 2.125,
      "learning_rate": 0.0004678769071093479,
      "loss": 5.7762,
      "mean_token_accuracy": 0.15942887365818023,
      "num_tokens": 41576150.0,
      "step": 20530
    },
    {
      "entropy": 6.055995988845825,
      "epoch": 1.7945468845582453,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00046786085939603785,
      "loss": 5.7603,
      "mean_token_accuracy": 0.16656645238399506,
      "num_tokens": 41585248.0,
      "step": 20535
    },
    {
      "entropy": 6.102273845672608,
      "epoch": 1.7949838329109498,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004678448079835552,
      "loss": 5.8275,
      "mean_token_accuracy": 0.15687010139226915,
      "num_tokens": 41595215.0,
      "step": 20540
    },
    {
      "entropy": 6.20481424331665,
      "epoch": 1.7954207812636547,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004678287528722083,
      "loss": 5.8759,
      "mean_token_accuracy": 0.1566680058836937,
      "num_tokens": 41605268.0,
      "step": 20545
    },
    {
      "entropy": 6.0769875049591064,
      "epoch": 1.7958577296163594,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004678126940623048,
      "loss": 5.7155,
      "mean_token_accuracy": 0.16630046516656877,
      "num_tokens": 41614554.0,
      "step": 20550
    },
    {
      "entropy": 6.060003280639648,
      "epoch": 1.796294677969064,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046779663155415287,
      "loss": 5.8423,
      "mean_token_accuracy": 0.16168198585510254,
      "num_tokens": 41624839.0,
      "step": 20555
    },
    {
      "entropy": 6.028343343734742,
      "epoch": 1.7967316263217687,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004677805653480604,
      "loss": 5.6969,
      "mean_token_accuracy": 0.17468472868204116,
      "num_tokens": 41634577.0,
      "step": 20560
    },
    {
      "entropy": 6.154319858551025,
      "epoch": 1.7971685746744734,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046776449544433575,
      "loss": 5.8416,
      "mean_token_accuracy": 0.15375510305166246,
      "num_tokens": 41645213.0,
      "step": 20565
    },
    {
      "entropy": 6.152465438842773,
      "epoch": 1.7976055230271784,
      "grad_norm": 2.09375,
      "learning_rate": 0.000467748421843287,
      "loss": 5.7128,
      "mean_token_accuracy": 0.16934482455253602,
      "num_tokens": 41655160.0,
      "step": 20570
    },
    {
      "entropy": 6.109637594223022,
      "epoch": 1.7980424713798828,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00046773234454522265,
      "loss": 5.7927,
      "mean_token_accuracy": 0.15572577118873596,
      "num_tokens": 41665861.0,
      "step": 20575
    },
    {
      "entropy": 6.1086390018463135,
      "epoch": 1.7984794197325877,
      "grad_norm": 2.65625,
      "learning_rate": 0.00046771626355045085,
      "loss": 5.8102,
      "mean_token_accuracy": 0.16062305569648744,
      "num_tokens": 41676699.0,
      "step": 20580
    },
    {
      "entropy": 6.127663087844849,
      "epoch": 1.7989163680852922,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004677001788592801,
      "loss": 5.839,
      "mean_token_accuracy": 0.16055029779672622,
      "num_tokens": 41688162.0,
      "step": 20585
    },
    {
      "entropy": 6.151435804367066,
      "epoch": 1.7993533164379971,
      "grad_norm": 2.28125,
      "learning_rate": 0.000467684090472019,
      "loss": 5.8337,
      "mean_token_accuracy": 0.15665507316589355,
      "num_tokens": 41698088.0,
      "step": 20590
    },
    {
      "entropy": 6.134163761138916,
      "epoch": 1.7997902647907016,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004676679983889761,
      "loss": 5.7751,
      "mean_token_accuracy": 0.16672883033752442,
      "num_tokens": 41707446.0,
      "step": 20595
    },
    {
      "entropy": 6.117214727401733,
      "epoch": 1.8002272131434065,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046765190261046,
      "loss": 5.8269,
      "mean_token_accuracy": 0.15792782604694366,
      "num_tokens": 41716388.0,
      "step": 20600
    },
    {
      "entropy": 6.152116441726685,
      "epoch": 1.8006641614961112,
      "grad_norm": 2.0,
      "learning_rate": 0.0004676358031367795,
      "loss": 5.8478,
      "mean_token_accuracy": 0.161505389213562,
      "num_tokens": 41726328.0,
      "step": 20605
    },
    {
      "entropy": 6.130058240890503,
      "epoch": 1.8011011098488159,
      "grad_norm": 3.125,
      "learning_rate": 0.00046761969996824326,
      "loss": 5.786,
      "mean_token_accuracy": 0.15552452951669693,
      "num_tokens": 41736128.0,
      "step": 20610
    },
    {
      "entropy": 6.059102201461792,
      "epoch": 1.8015380582015206,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004676035931051604,
      "loss": 5.7392,
      "mean_token_accuracy": 0.1600651666522026,
      "num_tokens": 41746316.0,
      "step": 20615
    },
    {
      "entropy": 6.154334640502929,
      "epoch": 1.8019750065542253,
      "grad_norm": 3.5,
      "learning_rate": 0.00046758748254783957,
      "loss": 5.7031,
      "mean_token_accuracy": 0.16528104543685912,
      "num_tokens": 41756015.0,
      "step": 20620
    },
    {
      "entropy": 6.1261743068695065,
      "epoch": 1.80241195490693,
      "grad_norm": 3.109375,
      "learning_rate": 0.00046757136829658996,
      "loss": 5.7524,
      "mean_token_accuracy": 0.16703375279903412,
      "num_tokens": 41766360.0,
      "step": 20625
    },
    {
      "entropy": 6.138262987136841,
      "epoch": 1.8028489032596346,
      "grad_norm": 3.578125,
      "learning_rate": 0.00046755525035172056,
      "loss": 5.8596,
      "mean_token_accuracy": 0.155458702147007,
      "num_tokens": 41776640.0,
      "step": 20630
    },
    {
      "entropy": 6.113243770599365,
      "epoch": 1.8032858516123396,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004675391287135406,
      "loss": 5.8368,
      "mean_token_accuracy": 0.16024687588214875,
      "num_tokens": 41787329.0,
      "step": 20635
    },
    {
      "entropy": 6.127006244659424,
      "epoch": 1.803722799965044,
      "grad_norm": 6.3125,
      "learning_rate": 0.00046752300338235924,
      "loss": 5.8445,
      "mean_token_accuracy": 0.15542975589632987,
      "num_tokens": 41797890.0,
      "step": 20640
    },
    {
      "entropy": 6.138513469696045,
      "epoch": 1.804159748317749,
      "grad_norm": 2.375,
      "learning_rate": 0.0004675068743584858,
      "loss": 5.84,
      "mean_token_accuracy": 0.15307705402374266,
      "num_tokens": 41808845.0,
      "step": 20645
    },
    {
      "entropy": 6.190570974349976,
      "epoch": 1.8045966966704534,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004674907416422297,
      "loss": 5.8173,
      "mean_token_accuracy": 0.15928134173154831,
      "num_tokens": 41819499.0,
      "step": 20650
    },
    {
      "entropy": 6.091172552108764,
      "epoch": 1.8050336450231583,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004674746052339002,
      "loss": 5.8281,
      "mean_token_accuracy": 0.15936562418937683,
      "num_tokens": 41827998.0,
      "step": 20655
    },
    {
      "entropy": 6.114912319183349,
      "epoch": 1.805470593375863,
      "grad_norm": 2.4375,
      "learning_rate": 0.00046745846513380704,
      "loss": 5.8685,
      "mean_token_accuracy": 0.15339112058281898,
      "num_tokens": 41839610.0,
      "step": 20660
    },
    {
      "entropy": 6.187399244308471,
      "epoch": 1.8059075417285677,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046744232134225954,
      "loss": 5.7874,
      "mean_token_accuracy": 0.1547248400747776,
      "num_tokens": 41850111.0,
      "step": 20665
    },
    {
      "entropy": 6.133647871017456,
      "epoch": 1.8063444900812724,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004674261738595676,
      "loss": 5.7727,
      "mean_token_accuracy": 0.15688270926475525,
      "num_tokens": 41859843.0,
      "step": 20670
    },
    {
      "entropy": 6.102827501296997,
      "epoch": 1.806781438433977,
      "grad_norm": 1.796875,
      "learning_rate": 0.00046741002268604073,
      "loss": 5.7885,
      "mean_token_accuracy": 0.1591956213116646,
      "num_tokens": 41868864.0,
      "step": 20675
    },
    {
      "entropy": 6.1106727600097654,
      "epoch": 1.8072183867866818,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00046739386782198883,
      "loss": 5.8242,
      "mean_token_accuracy": 0.15824078321456908,
      "num_tokens": 41878274.0,
      "step": 20680
    },
    {
      "entropy": 6.124302911758423,
      "epoch": 1.8076553351393865,
      "grad_norm": 1.890625,
      "learning_rate": 0.00046737770926772175,
      "loss": 5.7774,
      "mean_token_accuracy": 0.15494089871644973,
      "num_tokens": 41888005.0,
      "step": 20685
    },
    {
      "entropy": 6.178764772415161,
      "epoch": 1.8080922834920914,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004673615470235494,
      "loss": 5.8843,
      "mean_token_accuracy": 0.15514364540576936,
      "num_tokens": 41897415.0,
      "step": 20690
    },
    {
      "entropy": 6.115621042251587,
      "epoch": 1.8085292318447959,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004673453810897817,
      "loss": 5.7573,
      "mean_token_accuracy": 0.1636564776301384,
      "num_tokens": 41907021.0,
      "step": 20695
    },
    {
      "entropy": 6.111355829238891,
      "epoch": 1.8089661801975008,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004673292114667289,
      "loss": 5.8536,
      "mean_token_accuracy": 0.152622027695179,
      "num_tokens": 41918358.0,
      "step": 20700
    },
    {
      "entropy": 6.199451732635498,
      "epoch": 1.8094031285502052,
      "grad_norm": 1.96875,
      "learning_rate": 0.00046731303815470104,
      "loss": 5.8645,
      "mean_token_accuracy": 0.14563099816441535,
      "num_tokens": 41929311.0,
      "step": 20705
    },
    {
      "entropy": 6.1681983947753904,
      "epoch": 1.8098400769029102,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046729686115400826,
      "loss": 5.8174,
      "mean_token_accuracy": 0.15910544097423554,
      "num_tokens": 41939623.0,
      "step": 20710
    },
    {
      "entropy": 6.062235879898071,
      "epoch": 1.8102770252556146,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004672806804649611,
      "loss": 5.79,
      "mean_token_accuracy": 0.16066670417785645,
      "num_tokens": 41949650.0,
      "step": 20715
    },
    {
      "entropy": 6.136530256271362,
      "epoch": 1.8107139736083195,
      "grad_norm": 3.453125,
      "learning_rate": 0.00046726449608786957,
      "loss": 5.7672,
      "mean_token_accuracy": 0.15900637358427047,
      "num_tokens": 41960262.0,
      "step": 20720
    },
    {
      "entropy": 6.169749546051025,
      "epoch": 1.8111509219610242,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004672483080230443,
      "loss": 5.7968,
      "mean_token_accuracy": 0.15270697921514512,
      "num_tokens": 41969874.0,
      "step": 20725
    },
    {
      "entropy": 6.083888387680053,
      "epoch": 1.811587870313729,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046723211627079577,
      "loss": 5.7977,
      "mean_token_accuracy": 0.15930332541465758,
      "num_tokens": 41978810.0,
      "step": 20730
    },
    {
      "entropy": 6.055484867095947,
      "epoch": 1.8120248186664336,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004672159208314344,
      "loss": 5.7535,
      "mean_token_accuracy": 0.15917766243219375,
      "num_tokens": 41988785.0,
      "step": 20735
    },
    {
      "entropy": 6.085396146774292,
      "epoch": 1.8124617670191383,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004671997217052711,
      "loss": 5.8294,
      "mean_token_accuracy": 0.16241493076086044,
      "num_tokens": 41998698.0,
      "step": 20740
    },
    {
      "entropy": 6.107514905929565,
      "epoch": 1.812898715371843,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004671835188926164,
      "loss": 5.7815,
      "mean_token_accuracy": 0.1644461750984192,
      "num_tokens": 42009082.0,
      "step": 20745
    },
    {
      "entropy": 6.1661248207092285,
      "epoch": 1.8133356637245477,
      "grad_norm": 2.0,
      "learning_rate": 0.0004671673123937811,
      "loss": 5.8053,
      "mean_token_accuracy": 0.156510129570961,
      "num_tokens": 42019960.0,
      "step": 20750
    },
    {
      "entropy": 6.071886777877808,
      "epoch": 1.8137726120772526,
      "grad_norm": 2.09375,
      "learning_rate": 0.000467151102209076,
      "loss": 5.7171,
      "mean_token_accuracy": 0.16496340185403824,
      "num_tokens": 42028799.0,
      "step": 20755
    },
    {
      "entropy": 6.058169794082642,
      "epoch": 1.814209560429957,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046713488833881214,
      "loss": 5.8008,
      "mean_token_accuracy": 0.15486937165260314,
      "num_tokens": 42038967.0,
      "step": 20760
    },
    {
      "entropy": 6.094454908370972,
      "epoch": 1.814646508782662,
      "grad_norm": 2.625,
      "learning_rate": 0.00046711867078330047,
      "loss": 5.765,
      "mean_token_accuracy": 0.16041286438703536,
      "num_tokens": 42049982.0,
      "step": 20765
    },
    {
      "entropy": 6.12947187423706,
      "epoch": 1.8150834571353665,
      "grad_norm": 3.578125,
      "learning_rate": 0.00046710244954285184,
      "loss": 5.8831,
      "mean_token_accuracy": 0.15631230175495148,
      "num_tokens": 42059835.0,
      "step": 20770
    },
    {
      "entropy": 6.086891078948975,
      "epoch": 1.8155204054880714,
      "grad_norm": 1.78125,
      "learning_rate": 0.00046708622461777774,
      "loss": 5.6918,
      "mean_token_accuracy": 0.16795631349086762,
      "num_tokens": 42069339.0,
      "step": 20775
    },
    {
      "entropy": 6.078075170516968,
      "epoch": 1.815957353840776,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004670699960083892,
      "loss": 5.7284,
      "mean_token_accuracy": 0.16038838326931,
      "num_tokens": 42080618.0,
      "step": 20780
    },
    {
      "entropy": 6.147924709320068,
      "epoch": 1.8163943021934807,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004670537637149974,
      "loss": 5.8712,
      "mean_token_accuracy": 0.15019579082727433,
      "num_tokens": 42091458.0,
      "step": 20785
    },
    {
      "entropy": 6.135207605361939,
      "epoch": 1.8168312505461854,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046703752773791387,
      "loss": 5.8832,
      "mean_token_accuracy": 0.15260590761899948,
      "num_tokens": 42102515.0,
      "step": 20790
    },
    {
      "entropy": 6.1127361297607425,
      "epoch": 1.8172681988988901,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004670212880774499,
      "loss": 5.79,
      "mean_token_accuracy": 0.1580725595355034,
      "num_tokens": 42113853.0,
      "step": 20795
    },
    {
      "entropy": 6.13453516960144,
      "epoch": 1.8177051472515948,
      "grad_norm": 1.78125,
      "learning_rate": 0.000467005044733917,
      "loss": 5.8315,
      "mean_token_accuracy": 0.15106360316276551,
      "num_tokens": 42123101.0,
      "step": 20800
    },
    {
      "entropy": 6.1749011993408205,
      "epoch": 1.8181420956042995,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004669887977076267,
      "loss": 5.8111,
      "mean_token_accuracy": 0.15376880913972854,
      "num_tokens": 42133051.0,
      "step": 20805
    },
    {
      "entropy": 6.115439319610596,
      "epoch": 1.8185790439570044,
      "grad_norm": 1.8125,
      "learning_rate": 0.00046697254699889064,
      "loss": 5.8066,
      "mean_token_accuracy": 0.1609758272767067,
      "num_tokens": 42142705.0,
      "step": 20810
    },
    {
      "entropy": 6.093760919570923,
      "epoch": 1.819015992309709,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046695629260802064,
      "loss": 5.7881,
      "mean_token_accuracy": 0.1603650778532028,
      "num_tokens": 42152399.0,
      "step": 20815
    },
    {
      "entropy": 6.1325829982757565,
      "epoch": 1.8194529406624138,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004669400345353283,
      "loss": 5.8116,
      "mean_token_accuracy": 0.15808191448450087,
      "num_tokens": 42162651.0,
      "step": 20820
    },
    {
      "entropy": 6.026057291030884,
      "epoch": 1.8198898890151183,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004669237727811255,
      "loss": 5.7018,
      "mean_token_accuracy": 0.16397202312946318,
      "num_tokens": 42172595.0,
      "step": 20825
    },
    {
      "entropy": 6.091785955429077,
      "epoch": 1.8203268373678232,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004669075073457242,
      "loss": 5.7844,
      "mean_token_accuracy": 0.1637825071811676,
      "num_tokens": 42182631.0,
      "step": 20830
    },
    {
      "entropy": 6.128258085250854,
      "epoch": 1.8207637857205279,
      "grad_norm": 2.765625,
      "learning_rate": 0.00046689123822943643,
      "loss": 5.7818,
      "mean_token_accuracy": 0.15594326704740524,
      "num_tokens": 42192743.0,
      "step": 20835
    },
    {
      "entropy": 6.117905712127685,
      "epoch": 1.8212007340732326,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004668749654325741,
      "loss": 5.7913,
      "mean_token_accuracy": 0.15570884943008423,
      "num_tokens": 42203090.0,
      "step": 20840
    },
    {
      "entropy": 6.155374908447266,
      "epoch": 1.8216376824259373,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004668586889554493,
      "loss": 5.8801,
      "mean_token_accuracy": 0.15288415253162385,
      "num_tokens": 42213341.0,
      "step": 20845
    },
    {
      "entropy": 6.152459764480591,
      "epoch": 1.822074630778642,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004668424087983744,
      "loss": 5.8583,
      "mean_token_accuracy": 0.16066688150167466,
      "num_tokens": 42222862.0,
      "step": 20850
    },
    {
      "entropy": 6.159763431549072,
      "epoch": 1.8225115791313466,
      "grad_norm": 3.25,
      "learning_rate": 0.00046682612496166156,
      "loss": 5.854,
      "mean_token_accuracy": 0.15694592669606208,
      "num_tokens": 42234045.0,
      "step": 20855
    },
    {
      "entropy": 6.1904944896698,
      "epoch": 1.8229485274840513,
      "grad_norm": 2.703125,
      "learning_rate": 0.00046680983744562315,
      "loss": 5.783,
      "mean_token_accuracy": 0.16031327098608017,
      "num_tokens": 42244093.0,
      "step": 20860
    },
    {
      "entropy": 6.055976724624633,
      "epoch": 1.8233854758367563,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004667935462505716,
      "loss": 5.7981,
      "mean_token_accuracy": 0.16109249144792556,
      "num_tokens": 42254865.0,
      "step": 20865
    },
    {
      "entropy": 6.09691743850708,
      "epoch": 1.8238224241894607,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046677725137681924,
      "loss": 5.744,
      "mean_token_accuracy": 0.1622224047780037,
      "num_tokens": 42264565.0,
      "step": 20870
    },
    {
      "entropy": 6.1953661918640135,
      "epoch": 1.8242593725421656,
      "grad_norm": 1.765625,
      "learning_rate": 0.00046676095282467873,
      "loss": 5.8307,
      "mean_token_accuracy": 0.15550556927919387,
      "num_tokens": 42274289.0,
      "step": 20875
    },
    {
      "entropy": 6.1334230422973635,
      "epoch": 1.82469632089487,
      "grad_norm": 2.5,
      "learning_rate": 0.0004667446505944627,
      "loss": 5.8113,
      "mean_token_accuracy": 0.15679647624492646,
      "num_tokens": 42283855.0,
      "step": 20880
    },
    {
      "entropy": 5.984568786621094,
      "epoch": 1.825133269247575,
      "grad_norm": 5.0625,
      "learning_rate": 0.00046672834468648375,
      "loss": 5.83,
      "mean_token_accuracy": 0.16409749537706375,
      "num_tokens": 42294871.0,
      "step": 20885
    },
    {
      "entropy": 6.168867206573486,
      "epoch": 1.8255702176002795,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004667120351010547,
      "loss": 5.7963,
      "mean_token_accuracy": 0.16068321764469146,
      "num_tokens": 42305315.0,
      "step": 20890
    },
    {
      "entropy": 6.183169364929199,
      "epoch": 1.8260071659529844,
      "grad_norm": 1.890625,
      "learning_rate": 0.00046669572183848833,
      "loss": 5.7879,
      "mean_token_accuracy": 0.1644870012998581,
      "num_tokens": 42316893.0,
      "step": 20895
    },
    {
      "entropy": 6.064131784439087,
      "epoch": 1.826444114305689,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046667940489909766,
      "loss": 5.7142,
      "mean_token_accuracy": 0.16768437176942824,
      "num_tokens": 42326784.0,
      "step": 20900
    },
    {
      "entropy": 6.046210527420044,
      "epoch": 1.8268810626583938,
      "grad_norm": 1.6875,
      "learning_rate": 0.00046666308428319547,
      "loss": 5.7691,
      "mean_token_accuracy": 0.16883353739976883,
      "num_tokens": 42336283.0,
      "step": 20905
    },
    {
      "entropy": 6.078384971618652,
      "epoch": 1.8273180110110985,
      "grad_norm": 1.859375,
      "learning_rate": 0.00046664675999109494,
      "loss": 5.7902,
      "mean_token_accuracy": 0.16064900606870652,
      "num_tokens": 42346933.0,
      "step": 20910
    },
    {
      "entropy": 6.100211954116821,
      "epoch": 1.8277549593638032,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004666304320231091,
      "loss": 5.8441,
      "mean_token_accuracy": 0.15444888770580292,
      "num_tokens": 42357616.0,
      "step": 20915
    },
    {
      "entropy": 6.153164720535278,
      "epoch": 1.8281919077165079,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004666141003795512,
      "loss": 5.8095,
      "mean_token_accuracy": 0.16054720282554627,
      "num_tokens": 42367566.0,
      "step": 20920
    },
    {
      "entropy": 6.156742668151855,
      "epoch": 1.8286288560692125,
      "grad_norm": 1.84375,
      "learning_rate": 0.00046659776506073446,
      "loss": 5.8082,
      "mean_token_accuracy": 0.16173860877752305,
      "num_tokens": 42377412.0,
      "step": 20925
    },
    {
      "entropy": 6.144735383987427,
      "epoch": 1.8290658044219175,
      "grad_norm": 2.953125,
      "learning_rate": 0.00046658142606697216,
      "loss": 5.7777,
      "mean_token_accuracy": 0.1658511847257614,
      "num_tokens": 42387952.0,
      "step": 20930
    },
    {
      "entropy": 6.1396565437316895,
      "epoch": 1.829502752774622,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004665650833985777,
      "loss": 5.8192,
      "mean_token_accuracy": 0.15561270862817764,
      "num_tokens": 42398488.0,
      "step": 20935
    },
    {
      "entropy": 6.138120889663696,
      "epoch": 1.8299397011273268,
      "grad_norm": 1.75,
      "learning_rate": 0.00046654873705586464,
      "loss": 5.762,
      "mean_token_accuracy": 0.16394922137260437,
      "num_tokens": 42409137.0,
      "step": 20940
    },
    {
      "entropy": 6.157368946075439,
      "epoch": 1.8303766494800313,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004665323870391464,
      "loss": 5.8707,
      "mean_token_accuracy": 0.1497701436281204,
      "num_tokens": 42419867.0,
      "step": 20945
    },
    {
      "entropy": 6.118807220458985,
      "epoch": 1.8308135978327362,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004665160333487365,
      "loss": 5.8183,
      "mean_token_accuracy": 0.15170530527830123,
      "num_tokens": 42429590.0,
      "step": 20950
    },
    {
      "entropy": 6.175193643569946,
      "epoch": 1.831250546185441,
      "grad_norm": 2.125,
      "learning_rate": 0.0004664996759849489,
      "loss": 5.7813,
      "mean_token_accuracy": 0.16057996898889543,
      "num_tokens": 42439777.0,
      "step": 20955
    },
    {
      "entropy": 6.063419008255005,
      "epoch": 1.8316874945381456,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004664833149480971,
      "loss": 5.7779,
      "mean_token_accuracy": 0.16476726680994033,
      "num_tokens": 42449843.0,
      "step": 20960
    },
    {
      "entropy": 6.064124917984008,
      "epoch": 1.8321244428908503,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046646695023849506,
      "loss": 5.6485,
      "mean_token_accuracy": 0.17267682552337646,
      "num_tokens": 42459056.0,
      "step": 20965
    },
    {
      "entropy": 6.121998119354248,
      "epoch": 1.832561391243555,
      "grad_norm": 2.90625,
      "learning_rate": 0.00046645058185645644,
      "loss": 5.7092,
      "mean_token_accuracy": 0.16060147136449815,
      "num_tokens": 42468946.0,
      "step": 20970
    },
    {
      "entropy": 6.054912042617798,
      "epoch": 1.8329983395962597,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004664342098022954,
      "loss": 5.7746,
      "mean_token_accuracy": 0.16255973130464554,
      "num_tokens": 42478894.0,
      "step": 20975
    },
    {
      "entropy": 6.12795844078064,
      "epoch": 1.8334352879489644,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004664178340763259,
      "loss": 5.7703,
      "mean_token_accuracy": 0.16182252168655395,
      "num_tokens": 42490124.0,
      "step": 20980
    },
    {
      "entropy": 6.119140577316284,
      "epoch": 1.8338722363016693,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004664014546788621,
      "loss": 5.7913,
      "mean_token_accuracy": 0.15724663734436034,
      "num_tokens": 42499554.0,
      "step": 20985
    },
    {
      "entropy": 6.027507019042969,
      "epoch": 1.8343091846543738,
      "grad_norm": 3.828125,
      "learning_rate": 0.00046638507161021807,
      "loss": 5.675,
      "mean_token_accuracy": 0.1705660715699196,
      "num_tokens": 42510364.0,
      "step": 20990
    },
    {
      "entropy": 6.135519075393677,
      "epoch": 1.8347461330070787,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00046636868487070804,
      "loss": 5.8733,
      "mean_token_accuracy": 0.14906709194183348,
      "num_tokens": 42521671.0,
      "step": 20995
    },
    {
      "entropy": 6.136737632751465,
      "epoch": 1.8351830813597831,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00046635229446064647,
      "loss": 5.766,
      "mean_token_accuracy": 0.1624959036707878,
      "num_tokens": 42532784.0,
      "step": 21000
    },
    {
      "epoch": 1.8351830813597831,
      "eval_entropy": 5.872524485929187,
      "eval_loss": 5.83701753616333,
      "eval_mean_token_accuracy": 0.1655561091253135,
      "eval_num_tokens": 42532784.0,
      "eval_runtime": 28.1492,
      "eval_samples_per_second": 1314.921,
      "eval_steps_per_second": 164.374,
      "step": 21000
    },
    {
      "entropy": 6.001714563369751,
      "epoch": 1.835620029712488,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004663359003803475,
      "loss": 5.6816,
      "mean_token_accuracy": 0.17239667177200318,
      "num_tokens": 42542502.0,
      "step": 21005
    },
    {
      "entropy": 6.030185127258301,
      "epoch": 1.8360569780651927,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004663195026301258,
      "loss": 5.7289,
      "mean_token_accuracy": 0.15943232625722886,
      "num_tokens": 42551837.0,
      "step": 21010
    },
    {
      "entropy": 6.11195559501648,
      "epoch": 1.8364939264178974,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046630310121029573,
      "loss": 5.7594,
      "mean_token_accuracy": 0.16124747544527054,
      "num_tokens": 42560537.0,
      "step": 21015
    },
    {
      "entropy": 6.0969810962677,
      "epoch": 1.8369308747706021,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046628669612117194,
      "loss": 5.778,
      "mean_token_accuracy": 0.15909681916236879,
      "num_tokens": 42570386.0,
      "step": 21020
    },
    {
      "entropy": 6.122919607162475,
      "epoch": 1.8373678231233068,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046627028736306907,
      "loss": 5.8347,
      "mean_token_accuracy": 0.1570967212319374,
      "num_tokens": 42579860.0,
      "step": 21025
    },
    {
      "entropy": 6.175709533691406,
      "epoch": 1.8378047714760115,
      "grad_norm": 1.875,
      "learning_rate": 0.00046625387493630187,
      "loss": 5.8028,
      "mean_token_accuracy": 0.15672231912612916,
      "num_tokens": 42590320.0,
      "step": 21030
    },
    {
      "entropy": 6.111327171325684,
      "epoch": 1.8382417198287162,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004662374588411852,
      "loss": 5.7707,
      "mean_token_accuracy": 0.16084133237600326,
      "num_tokens": 42600331.0,
      "step": 21035
    },
    {
      "entropy": 6.076353120803833,
      "epoch": 1.8386786681814211,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046622103907803384,
      "loss": 5.8081,
      "mean_token_accuracy": 0.16115770787000655,
      "num_tokens": 42610860.0,
      "step": 21040
    },
    {
      "entropy": 6.073611211776734,
      "epoch": 1.8391156165341256,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046620461564716265,
      "loss": 5.834,
      "mean_token_accuracy": 0.154962195456028,
      "num_tokens": 42620305.0,
      "step": 21045
    },
    {
      "entropy": 6.0908531665802,
      "epoch": 1.8395525648868305,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004661881885488869,
      "loss": 5.7311,
      "mean_token_accuracy": 0.16491455286741258,
      "num_tokens": 42629667.0,
      "step": 21050
    },
    {
      "entropy": 6.084347772598266,
      "epoch": 1.839989513239535,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004661717577835214,
      "loss": 5.8231,
      "mean_token_accuracy": 0.16295826137065889,
      "num_tokens": 42640116.0,
      "step": 21055
    },
    {
      "entropy": 6.076373815536499,
      "epoch": 1.8404264615922399,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004661553233513815,
      "loss": 5.7508,
      "mean_token_accuracy": 0.1641483947634697,
      "num_tokens": 42650014.0,
      "step": 21060
    },
    {
      "entropy": 6.100523805618286,
      "epoch": 1.8408634099449444,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004661388852527822,
      "loss": 5.8319,
      "mean_token_accuracy": 0.1542024865746498,
      "num_tokens": 42661280.0,
      "step": 21065
    },
    {
      "entropy": 6.070639038085938,
      "epoch": 1.8413003582976493,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046612244348803906,
      "loss": 5.8505,
      "mean_token_accuracy": 0.1605138137936592,
      "num_tokens": 42671530.0,
      "step": 21070
    },
    {
      "entropy": 6.173719787597657,
      "epoch": 1.841737306650354,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004661059980574672,
      "loss": 5.8482,
      "mean_token_accuracy": 0.15914203524589537,
      "num_tokens": 42682738.0,
      "step": 21075
    },
    {
      "entropy": 6.095088386535645,
      "epoch": 1.8421742550030586,
      "grad_norm": 2.5,
      "learning_rate": 0.0004660895489613822,
      "loss": 5.8152,
      "mean_token_accuracy": 0.15877957344055177,
      "num_tokens": 42694271.0,
      "step": 21080
    },
    {
      "entropy": 6.118168592453003,
      "epoch": 1.8426112033557633,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004660730962000996,
      "loss": 5.8607,
      "mean_token_accuracy": 0.15424942672252656,
      "num_tokens": 42704896.0,
      "step": 21085
    },
    {
      "entropy": 6.079190492630005,
      "epoch": 1.843048151708468,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004660566397739348,
      "loss": 5.7702,
      "mean_token_accuracy": 0.15847987532615662,
      "num_tokens": 42715506.0,
      "step": 21090
    },
    {
      "entropy": 6.122553777694702,
      "epoch": 1.8434851000611727,
      "grad_norm": 2.96875,
      "learning_rate": 0.00046604017968320357,
      "loss": 5.7575,
      "mean_token_accuracy": 0.17038540840148925,
      "num_tokens": 42725428.0,
      "step": 21095
    },
    {
      "entropy": 6.076107358932495,
      "epoch": 1.8439220484138774,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004660237159282216,
      "loss": 5.7607,
      "mean_token_accuracy": 0.161995729804039,
      "num_tokens": 42735105.0,
      "step": 21100
    },
    {
      "entropy": 6.054056119918823,
      "epoch": 1.8443589967665823,
      "grad_norm": 3.34375,
      "learning_rate": 0.00046600724850930466,
      "loss": 5.7013,
      "mean_token_accuracy": 0.1656486764550209,
      "num_tokens": 42744191.0,
      "step": 21105
    },
    {
      "entropy": 6.120818662643432,
      "epoch": 1.8447959451192868,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004659907774267686,
      "loss": 5.8868,
      "mean_token_accuracy": 0.1558373227715492,
      "num_tokens": 42755416.0,
      "step": 21110
    },
    {
      "entropy": 6.097904634475708,
      "epoch": 1.8452328934719917,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00046597430268092933,
      "loss": 5.7004,
      "mean_token_accuracy": 0.17004013657569886,
      "num_tokens": 42764974.0,
      "step": 21115
    },
    {
      "entropy": 6.081737613677978,
      "epoch": 1.8456698418246962,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00046595782427210285,
      "loss": 5.6842,
      "mean_token_accuracy": 0.16898720413446428,
      "num_tokens": 42775080.0,
      "step": 21120
    },
    {
      "entropy": 6.066785049438477,
      "epoch": 1.846106790177401,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004659413422006053,
      "loss": 5.7636,
      "mean_token_accuracy": 0.15952033847570418,
      "num_tokens": 42784333.0,
      "step": 21125
    },
    {
      "entropy": 6.119672441482544,
      "epoch": 1.8465437385301058,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004659248564667527,
      "loss": 5.7855,
      "mean_token_accuracy": 0.1606551542878151,
      "num_tokens": 42792912.0,
      "step": 21130
    },
    {
      "entropy": 6.091796588897705,
      "epoch": 1.8469806868828105,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046590836707086135,
      "loss": 5.8013,
      "mean_token_accuracy": 0.16150100231170655,
      "num_tokens": 42803812.0,
      "step": 21135
    },
    {
      "entropy": 6.1871438980102536,
      "epoch": 1.8474176352355152,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046589187401324753,
      "loss": 5.8619,
      "mean_token_accuracy": 0.16030937507748605,
      "num_tokens": 42815172.0,
      "step": 21140
    },
    {
      "entropy": 6.179558563232422,
      "epoch": 1.8478545835882199,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004658753772942274,
      "loss": 5.7577,
      "mean_token_accuracy": 0.16022829115390777,
      "num_tokens": 42824958.0,
      "step": 21145
    },
    {
      "entropy": 6.053794527053833,
      "epoch": 1.8482915319409245,
      "grad_norm": 2.34375,
      "learning_rate": 0.00046585887691411764,
      "loss": 5.7968,
      "mean_token_accuracy": 0.16131065040826797,
      "num_tokens": 42834940.0,
      "step": 21150
    },
    {
      "entropy": 6.0853846073150635,
      "epoch": 1.8487284802936292,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004658423728732346,
      "loss": 5.7275,
      "mean_token_accuracy": 0.16349667310714722,
      "num_tokens": 42845102.0,
      "step": 21155
    },
    {
      "entropy": 6.071463108062744,
      "epoch": 1.8491654286463342,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004658258651718948,
      "loss": 5.7282,
      "mean_token_accuracy": 0.164603154361248,
      "num_tokens": 42853876.0,
      "step": 21160
    },
    {
      "entropy": 6.156432676315307,
      "epoch": 1.8496023769990386,
      "grad_norm": 3.6875,
      "learning_rate": 0.00046580935381041496,
      "loss": 5.8968,
      "mean_token_accuracy": 0.1499750331044197,
      "num_tokens": 42863592.0,
      "step": 21165
    },
    {
      "entropy": 6.157015085220337,
      "epoch": 1.8500393253517435,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004657928387891117,
      "loss": 5.7241,
      "mean_token_accuracy": 0.17081764042377473,
      "num_tokens": 42873478.0,
      "step": 21170
    },
    {
      "entropy": 6.084035778045655,
      "epoch": 1.850476273704448,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046577632010830184,
      "loss": 5.7093,
      "mean_token_accuracy": 0.16472101509571074,
      "num_tokens": 42882591.0,
      "step": 21175
    },
    {
      "entropy": 6.015787267684937,
      "epoch": 1.850913222057153,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004657597977683022,
      "loss": 5.6812,
      "mean_token_accuracy": 0.16612837463617325,
      "num_tokens": 42892391.0,
      "step": 21180
    },
    {
      "entropy": 6.0931525230407715,
      "epoch": 1.8513501704098576,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004657432717694297,
      "loss": 5.7632,
      "mean_token_accuracy": 0.1639419302344322,
      "num_tokens": 42902839.0,
      "step": 21185
    },
    {
      "entropy": 6.112688970565796,
      "epoch": 1.8517871187625623,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004657267421120012,
      "loss": 5.8698,
      "mean_token_accuracy": 0.15854010432958604,
      "num_tokens": 42912404.0,
      "step": 21190
    },
    {
      "entropy": 6.107426929473877,
      "epoch": 1.852224067115267,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004657102087963339,
      "loss": 5.7747,
      "mean_token_accuracy": 0.16196314841508866,
      "num_tokens": 42922939.0,
      "step": 21195
    },
    {
      "entropy": 6.0321714878082275,
      "epoch": 1.8526610154679717,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00046569367182274494,
      "loss": 5.7742,
      "mean_token_accuracy": 0.16651603281497956,
      "num_tokens": 42932095.0,
      "step": 21200
    },
    {
      "entropy": 6.121813583374023,
      "epoch": 1.8530979638206764,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004656771311915513,
      "loss": 5.7733,
      "mean_token_accuracy": 0.1605008438229561,
      "num_tokens": 42941910.0,
      "step": 21205
    },
    {
      "entropy": 6.171591567993164,
      "epoch": 1.853534912173381,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004656605869030704,
      "loss": 5.7855,
      "mean_token_accuracy": 0.16009848415851594,
      "num_tokens": 42952869.0,
      "step": 21210
    },
    {
      "entropy": 6.154482841491699,
      "epoch": 1.853971860526086,
      "grad_norm": 2.453125,
      "learning_rate": 0.00046564403895761966,
      "loss": 5.896,
      "mean_token_accuracy": 0.1588836908340454,
      "num_tokens": 42963300.0,
      "step": 21215
    },
    {
      "entropy": 6.122915315628052,
      "epoch": 1.8544088088787904,
      "grad_norm": 2.125,
      "learning_rate": 0.0004656274873555162,
      "loss": 5.7355,
      "mean_token_accuracy": 0.16273196637630463,
      "num_tokens": 42973505.0,
      "step": 21220
    },
    {
      "entropy": 6.051207113265991,
      "epoch": 1.8548457572314954,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004656109320970778,
      "loss": 5.779,
      "mean_token_accuracy": 0.1615119308233261,
      "num_tokens": 42982813.0,
      "step": 21225
    },
    {
      "entropy": 6.019395589828491,
      "epoch": 1.8552827055841998,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004655943731826217,
      "loss": 5.8028,
      "mean_token_accuracy": 0.15874289721250534,
      "num_tokens": 42991792.0,
      "step": 21230
    },
    {
      "entropy": 6.096352910995483,
      "epoch": 1.8557196539369047,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004655778106124658,
      "loss": 5.7304,
      "mean_token_accuracy": 0.16833239644765854,
      "num_tokens": 43002191.0,
      "step": 21235
    },
    {
      "entropy": 6.086009311676025,
      "epoch": 1.8561566022896092,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004655612443869275,
      "loss": 5.744,
      "mean_token_accuracy": 0.16930028200149536,
      "num_tokens": 43013771.0,
      "step": 21240
    },
    {
      "entropy": 6.077953433990478,
      "epoch": 1.8565935506423141,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004655446745063247,
      "loss": 5.843,
      "mean_token_accuracy": 0.15259860903024675,
      "num_tokens": 43023982.0,
      "step": 21245
    },
    {
      "entropy": 6.130091714859009,
      "epoch": 1.8570304989950188,
      "grad_norm": 1.875,
      "learning_rate": 0.0004655281009709753,
      "loss": 5.8691,
      "mean_token_accuracy": 0.15205853879451753,
      "num_tokens": 43034827.0,
      "step": 21250
    },
    {
      "entropy": 6.140582513809204,
      "epoch": 1.8574674473477235,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046551152378119696,
      "loss": 5.7378,
      "mean_token_accuracy": 0.16446263194084168,
      "num_tokens": 43044811.0,
      "step": 21255
    },
    {
      "entropy": 6.077371311187744,
      "epoch": 1.8579043957004282,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004654949429373078,
      "loss": 5.7865,
      "mean_token_accuracy": 0.15672390013933182,
      "num_tokens": 43054513.0,
      "step": 21260
    },
    {
      "entropy": 6.073606014251709,
      "epoch": 1.858341344053133,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004654783584396258,
      "loss": 5.8187,
      "mean_token_accuracy": 0.1587653249502182,
      "num_tokens": 43064356.0,
      "step": 21265
    },
    {
      "entropy": 6.148980712890625,
      "epoch": 1.8587782924058376,
      "grad_norm": 2.375,
      "learning_rate": 0.00046546177028846915,
      "loss": 5.675,
      "mean_token_accuracy": 0.1662023887038231,
      "num_tokens": 43074663.0,
      "step": 21270
    },
    {
      "entropy": 6.077380275726318,
      "epoch": 1.8592152407585423,
      "grad_norm": 1.875,
      "learning_rate": 0.0004654451784841559,
      "loss": 5.765,
      "mean_token_accuracy": 0.1643494337797165,
      "num_tokens": 43084542.0,
      "step": 21275
    },
    {
      "entropy": 5.970224666595459,
      "epoch": 1.8596521891112472,
      "grad_norm": 1.71875,
      "learning_rate": 0.00046542858302700443,
      "loss": 5.5945,
      "mean_token_accuracy": 0.17556772977113724,
      "num_tokens": 43093864.0,
      "step": 21280
    },
    {
      "entropy": 6.049560499191284,
      "epoch": 1.8600891374639517,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046541198391733286,
      "loss": 5.7402,
      "mean_token_accuracy": 0.1662487953901291,
      "num_tokens": 43104541.0,
      "step": 21285
    },
    {
      "entropy": 6.105762624740601,
      "epoch": 1.8605260858166566,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004653953811554596,
      "loss": 5.7757,
      "mean_token_accuracy": 0.1604187607765198,
      "num_tokens": 43114378.0,
      "step": 21290
    },
    {
      "entropy": 6.1775867462158205,
      "epoch": 1.860963034169361,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004653787747417033,
      "loss": 5.9595,
      "mean_token_accuracy": 0.15377915054559707,
      "num_tokens": 43125503.0,
      "step": 21295
    },
    {
      "entropy": 6.1538002490997314,
      "epoch": 1.861399982522066,
      "grad_norm": 3.015625,
      "learning_rate": 0.00046536216467638227,
      "loss": 5.7502,
      "mean_token_accuracy": 0.16760197132825852,
      "num_tokens": 43135034.0,
      "step": 21300
    },
    {
      "entropy": 6.025908184051514,
      "epoch": 1.8618369308747706,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004653455509598152,
      "loss": 5.8023,
      "mean_token_accuracy": 0.1602514922618866,
      "num_tokens": 43145359.0,
      "step": 21305
    },
    {
      "entropy": 6.0600872993469235,
      "epoch": 1.8622738792274753,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004653289335923206,
      "loss": 5.7018,
      "mean_token_accuracy": 0.17040675580501558,
      "num_tokens": 43154995.0,
      "step": 21310
    },
    {
      "entropy": 6.106922912597656,
      "epoch": 1.86271082758018,
      "grad_norm": 1.875,
      "learning_rate": 0.0004653123125742175,
      "loss": 5.7621,
      "mean_token_accuracy": 0.15993706732988358,
      "num_tokens": 43164703.0,
      "step": 21315
    },
    {
      "entropy": 6.076250743865967,
      "epoch": 1.8631477759328847,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004652956879058245,
      "loss": 5.7762,
      "mean_token_accuracy": 0.16136355698108673,
      "num_tokens": 43174193.0,
      "step": 21320
    },
    {
      "entropy": 6.0683828830719,
      "epoch": 1.8635847242855894,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046527905958746043,
      "loss": 5.8031,
      "mean_token_accuracy": 0.16810499280691146,
      "num_tokens": 43184236.0,
      "step": 21325
    },
    {
      "entropy": 6.1298089027404785,
      "epoch": 1.864021672638294,
      "grad_norm": 1.734375,
      "learning_rate": 0.00046526242761944434,
      "loss": 5.8101,
      "mean_token_accuracy": 0.16728546619415283,
      "num_tokens": 43194421.0,
      "step": 21330
    },
    {
      "entropy": 6.111288022994995,
      "epoch": 1.864458620990999,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00046524579200209513,
      "loss": 5.822,
      "mean_token_accuracy": 0.16491037160158156,
      "num_tokens": 43205200.0,
      "step": 21335
    },
    {
      "entropy": 6.034941101074219,
      "epoch": 1.8648955693437035,
      "grad_norm": 1.8125,
      "learning_rate": 0.00046522915273573194,
      "loss": 5.739,
      "mean_token_accuracy": 0.1665284067392349,
      "num_tokens": 43215415.0,
      "step": 21340
    },
    {
      "entropy": 6.102831220626831,
      "epoch": 1.8653325176964084,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004652125098206739,
      "loss": 5.7925,
      "mean_token_accuracy": 0.15773145258426666,
      "num_tokens": 43225648.0,
      "step": 21345
    },
    {
      "entropy": 6.1530920505523685,
      "epoch": 1.8657694660491129,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046519586325724033,
      "loss": 5.7696,
      "mean_token_accuracy": 0.15194757133722306,
      "num_tokens": 43236358.0,
      "step": 21350
    },
    {
      "entropy": 6.160669374465942,
      "epoch": 1.8662064144018178,
      "grad_norm": 2.0625,
      "learning_rate": 0.00046517921304575045,
      "loss": 5.808,
      "mean_token_accuracy": 0.1583761528134346,
      "num_tokens": 43247022.0,
      "step": 21355
    },
    {
      "entropy": 6.153940439224243,
      "epoch": 1.8666433627545223,
      "grad_norm": 1.859375,
      "learning_rate": 0.00046516255918652356,
      "loss": 5.8619,
      "mean_token_accuracy": 0.1575590193271637,
      "num_tokens": 43258249.0,
      "step": 21360
    },
    {
      "entropy": 6.113579177856446,
      "epoch": 1.8670803111072272,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004651459016798791,
      "loss": 5.7962,
      "mean_token_accuracy": 0.15957245528697966,
      "num_tokens": 43268997.0,
      "step": 21365
    },
    {
      "entropy": 6.088466453552246,
      "epoch": 1.8675172594599319,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004651292405261368,
      "loss": 5.757,
      "mean_token_accuracy": 0.16015819758176802,
      "num_tokens": 43279165.0,
      "step": 21370
    },
    {
      "entropy": 6.054091310501098,
      "epoch": 1.8679542078126365,
      "grad_norm": 1.84375,
      "learning_rate": 0.00046511257572561584,
      "loss": 5.8049,
      "mean_token_accuracy": 0.16256335228681565,
      "num_tokens": 43289778.0,
      "step": 21375
    },
    {
      "entropy": 6.049788475036621,
      "epoch": 1.8683911561653412,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004650959072786362,
      "loss": 5.7251,
      "mean_token_accuracy": 0.15973448157310485,
      "num_tokens": 43299865.0,
      "step": 21380
    },
    {
      "entropy": 6.033264207839966,
      "epoch": 1.868828104518046,
      "grad_norm": 1.5,
      "learning_rate": 0.0004650792351855173,
      "loss": 5.7462,
      "mean_token_accuracy": 0.16247664242982865,
      "num_tokens": 43309224.0,
      "step": 21385
    },
    {
      "entropy": 6.14097638130188,
      "epoch": 1.8692650528707508,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004650625594465792,
      "loss": 5.7206,
      "mean_token_accuracy": 0.1524226486682892,
      "num_tokens": 43320001.0,
      "step": 21390
    },
    {
      "entropy": 6.105134677886963,
      "epoch": 1.8697020012234553,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004650458800621415,
      "loss": 5.7225,
      "mean_token_accuracy": 0.16701753586530685,
      "num_tokens": 43329790.0,
      "step": 21395
    },
    {
      "entropy": 6.131563997268676,
      "epoch": 1.8701389495761602,
      "grad_norm": 1.875,
      "learning_rate": 0.0004650291970325244,
      "loss": 5.8502,
      "mean_token_accuracy": 0.16006935983896256,
      "num_tokens": 43340394.0,
      "step": 21400
    },
    {
      "entropy": 6.110442113876343,
      "epoch": 1.8705758979288647,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004650125103580476,
      "loss": 5.8139,
      "mean_token_accuracy": 0.15924681350588799,
      "num_tokens": 43351404.0,
      "step": 21405
    },
    {
      "entropy": 6.004949522018433,
      "epoch": 1.8710128462815696,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004649958200390313,
      "loss": 5.6082,
      "mean_token_accuracy": 0.17720636129379272,
      "num_tokens": 43361017.0,
      "step": 21410
    },
    {
      "entropy": 6.0817413330078125,
      "epoch": 1.871449794634274,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004649791260757956,
      "loss": 5.8435,
      "mean_token_accuracy": 0.158952596783638,
      "num_tokens": 43371680.0,
      "step": 21415
    },
    {
      "entropy": 6.133879566192627,
      "epoch": 1.871886742986979,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004649624284686608,
      "loss": 5.8236,
      "mean_token_accuracy": 0.16215385794639586,
      "num_tokens": 43381908.0,
      "step": 21420
    },
    {
      "entropy": 6.236153602600098,
      "epoch": 1.8723236913396837,
      "grad_norm": 1.5625,
      "learning_rate": 0.000464945727217947,
      "loss": 5.8851,
      "mean_token_accuracy": 0.15022221505641936,
      "num_tokens": 43393234.0,
      "step": 21425
    },
    {
      "entropy": 6.143374443054199,
      "epoch": 1.8727606396923884,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046492902232397464,
      "loss": 5.8752,
      "mean_token_accuracy": 0.15742940008640288,
      "num_tokens": 43404065.0,
      "step": 21430
    },
    {
      "entropy": 6.106721258163452,
      "epoch": 1.873197588045093,
      "grad_norm": 2.515625,
      "learning_rate": 0.00046491231378706415,
      "loss": 5.7859,
      "mean_token_accuracy": 0.15866263434290886,
      "num_tokens": 43414417.0,
      "step": 21435
    },
    {
      "entropy": 6.148135137557984,
      "epoch": 1.8736345363977978,
      "grad_norm": 1.46875,
      "learning_rate": 0.000464895601607536,
      "loss": 5.7954,
      "mean_token_accuracy": 0.1533399850130081,
      "num_tokens": 43424265.0,
      "step": 21440
    },
    {
      "entropy": 6.065779399871826,
      "epoch": 1.8740714847505024,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004648788857857106,
      "loss": 5.706,
      "mean_token_accuracy": 0.16649597585201265,
      "num_tokens": 43434699.0,
      "step": 21445
    },
    {
      "entropy": 6.131023359298706,
      "epoch": 1.8745084331032071,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00046486216632190874,
      "loss": 5.8037,
      "mean_token_accuracy": 0.15973794907331468,
      "num_tokens": 43444113.0,
      "step": 21450
    },
    {
      "entropy": 6.091975164413452,
      "epoch": 1.874945381455912,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00046484544321645103,
      "loss": 5.7666,
      "mean_token_accuracy": 0.1634966641664505,
      "num_tokens": 43454256.0,
      "step": 21455
    },
    {
      "entropy": 6.1085569858551025,
      "epoch": 1.8753823298086165,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004648287164696583,
      "loss": 5.7744,
      "mean_token_accuracy": 0.16158985793590547,
      "num_tokens": 43463182.0,
      "step": 21460
    },
    {
      "entropy": 6.113115978240967,
      "epoch": 1.8758192781613214,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004648119860818512,
      "loss": 5.7396,
      "mean_token_accuracy": 0.1646423101425171,
      "num_tokens": 43473326.0,
      "step": 21465
    },
    {
      "entropy": 6.089161014556884,
      "epoch": 1.876256226514026,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00046479525205335084,
      "loss": 5.8175,
      "mean_token_accuracy": 0.16231722384691238,
      "num_tokens": 43483210.0,
      "step": 21470
    },
    {
      "entropy": 6.108362913131714,
      "epoch": 1.8766931748667308,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004647785143844781,
      "loss": 5.7697,
      "mean_token_accuracy": 0.1638277381658554,
      "num_tokens": 43493834.0,
      "step": 21475
    },
    {
      "entropy": 6.115188646316528,
      "epoch": 1.8771301232194355,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046476177307555394,
      "loss": 5.7832,
      "mean_token_accuracy": 0.1598975405097008,
      "num_tokens": 43504522.0,
      "step": 21480
    },
    {
      "entropy": 6.115261745452881,
      "epoch": 1.8775670715721402,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046474502812689954,
      "loss": 5.8632,
      "mean_token_accuracy": 0.15327217429876328,
      "num_tokens": 43514950.0,
      "step": 21485
    },
    {
      "entropy": 6.168273019790649,
      "epoch": 1.878004019924845,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004647282795388361,
      "loss": 5.7969,
      "mean_token_accuracy": 0.16333367824554443,
      "num_tokens": 43526463.0,
      "step": 21490
    },
    {
      "entropy": 6.09530873298645,
      "epoch": 1.8784409682775496,
      "grad_norm": 2.125,
      "learning_rate": 0.0004647115273116849,
      "loss": 5.7846,
      "mean_token_accuracy": 0.1630472406744957,
      "num_tokens": 43536104.0,
      "step": 21495
    },
    {
      "entropy": 6.1933815479278564,
      "epoch": 1.8788779166302543,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00046469477144576716,
      "loss": 5.938,
      "mean_token_accuracy": 0.15442835092544555,
      "num_tokens": 43547863.0,
      "step": 21500
    },
    {
      "entropy": 6.174981117248535,
      "epoch": 1.879314864982959,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004646780119414043,
      "loss": 5.7779,
      "mean_token_accuracy": 0.1558986186981201,
      "num_tokens": 43558024.0,
      "step": 21505
    },
    {
      "entropy": 6.122202968597412,
      "epoch": 1.8797518133356639,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004646612487989178,
      "loss": 5.8132,
      "mean_token_accuracy": 0.15924556851387023,
      "num_tokens": 43567464.0,
      "step": 21510
    },
    {
      "entropy": 6.052398824691773,
      "epoch": 1.8801887616883683,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004646444820186291,
      "loss": 5.7066,
      "mean_token_accuracy": 0.1668059855699539,
      "num_tokens": 43576826.0,
      "step": 21515
    },
    {
      "entropy": 6.069525861740113,
      "epoch": 1.8806257100410733,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046462771160085995,
      "loss": 5.7102,
      "mean_token_accuracy": 0.17070818245410918,
      "num_tokens": 43586106.0,
      "step": 21520
    },
    {
      "entropy": 6.085967350006103,
      "epoch": 1.8810626583937777,
      "grad_norm": 1.828125,
      "learning_rate": 0.00046461093754593185,
      "loss": 5.8524,
      "mean_token_accuracy": 0.15460758060216903,
      "num_tokens": 43597077.0,
      "step": 21525
    },
    {
      "entropy": 6.192723274230957,
      "epoch": 1.8814996067464826,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004645941598541667,
      "loss": 5.8841,
      "mean_token_accuracy": 0.15249804630875588,
      "num_tokens": 43608773.0,
      "step": 21530
    },
    {
      "entropy": 6.205351161956787,
      "epoch": 1.8819365550991871,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004645773785258862,
      "loss": 5.7716,
      "mean_token_accuracy": 0.16561608612537385,
      "num_tokens": 43618017.0,
      "step": 21535
    },
    {
      "entropy": 6.031742143630981,
      "epoch": 1.882373503451892,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004645605935614122,
      "loss": 5.7837,
      "mean_token_accuracy": 0.15772687867283822,
      "num_tokens": 43627865.0,
      "step": 21540
    },
    {
      "entropy": 6.1055442810058596,
      "epoch": 1.8828104518045967,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046454380496106665,
      "loss": 5.7868,
      "mean_token_accuracy": 0.15888897180557252,
      "num_tokens": 43637772.0,
      "step": 21545
    },
    {
      "entropy": 6.071532964706421,
      "epoch": 1.8832474001573014,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00046452701272517167,
      "loss": 5.7141,
      "mean_token_accuracy": 0.16597330570220947,
      "num_tokens": 43648242.0,
      "step": 21550
    },
    {
      "entropy": 6.0739952564239506,
      "epoch": 1.883684348510006,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046451021685404927,
      "loss": 5.8152,
      "mean_token_accuracy": 0.16316785514354706,
      "num_tokens": 43658298.0,
      "step": 21555
    },
    {
      "entropy": 6.122341632843018,
      "epoch": 1.8841212968627108,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00046449341734802155,
      "loss": 5.7951,
      "mean_token_accuracy": 0.16252297759056092,
      "num_tokens": 43667792.0,
      "step": 21560
    },
    {
      "entropy": 6.155367374420166,
      "epoch": 1.8845582452154155,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004644766142074108,
      "loss": 5.7379,
      "mean_token_accuracy": 0.16779206693172455,
      "num_tokens": 43678571.0,
      "step": 21565
    },
    {
      "entropy": 6.091327095031739,
      "epoch": 1.8849951935681202,
      "grad_norm": 1.875,
      "learning_rate": 0.00046445980743253936,
      "loss": 5.833,
      "mean_token_accuracy": 0.15783566683530809,
      "num_tokens": 43690851.0,
      "step": 21570
    },
    {
      "entropy": 6.060152864456176,
      "epoch": 1.885432141920825,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046444299702372955,
      "loss": 5.7635,
      "mean_token_accuracy": 0.16465977877378463,
      "num_tokens": 43700960.0,
      "step": 21575
    },
    {
      "entropy": 6.1534586429595945,
      "epoch": 1.8858690902735296,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004644261829813038,
      "loss": 5.7582,
      "mean_token_accuracy": 0.1612315744161606,
      "num_tokens": 43710466.0,
      "step": 21580
    },
    {
      "entropy": 6.13233790397644,
      "epoch": 1.8863060386262345,
      "grad_norm": 1.5625,
      "learning_rate": 0.00046440936530558455,
      "loss": 5.8313,
      "mean_token_accuracy": 0.1589461714029312,
      "num_tokens": 43720261.0,
      "step": 21585
    },
    {
      "entropy": 6.12819995880127,
      "epoch": 1.886742986978939,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004643925439968944,
      "loss": 5.8268,
      "mean_token_accuracy": 0.15618243664503098,
      "num_tokens": 43731045.0,
      "step": 21590
    },
    {
      "entropy": 6.116936445236206,
      "epoch": 1.8871799353316439,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046437571905555604,
      "loss": 5.7959,
      "mean_token_accuracy": 0.16024910882115365,
      "num_tokens": 43741873.0,
      "step": 21595
    },
    {
      "entropy": 6.087710285186768,
      "epoch": 1.8876168836843485,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046435889048189216,
      "loss": 5.8458,
      "mean_token_accuracy": 0.15889556109905242,
      "num_tokens": 43751473.0,
      "step": 21600
    },
    {
      "entropy": 6.193316888809204,
      "epoch": 1.8880538320370532,
      "grad_norm": 1.609375,
      "learning_rate": 0.00046434205827622555,
      "loss": 5.8163,
      "mean_token_accuracy": 0.16316343396902083,
      "num_tokens": 43762517.0,
      "step": 21605
    },
    {
      "entropy": 6.141143369674682,
      "epoch": 1.888490780389758,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004643252224388791,
      "loss": 5.8621,
      "mean_token_accuracy": 0.15329610407352448,
      "num_tokens": 43772491.0,
      "step": 21610
    },
    {
      "entropy": 6.117872858047486,
      "epoch": 1.8889277287424626,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00046430838297017556,
      "loss": 5.7847,
      "mean_token_accuracy": 0.15975019782781602,
      "num_tokens": 43781646.0,
      "step": 21615
    },
    {
      "entropy": 6.133139276504517,
      "epoch": 1.8893646770951673,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00046429153987043815,
      "loss": 5.7148,
      "mean_token_accuracy": 0.16329576820135117,
      "num_tokens": 43792020.0,
      "step": 21620
    },
    {
      "entropy": 6.008482980728149,
      "epoch": 1.889801625447872,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004642746931399898,
      "loss": 5.716,
      "mean_token_accuracy": 0.1673393204808235,
      "num_tokens": 43801638.0,
      "step": 21625
    },
    {
      "entropy": 6.16087965965271,
      "epoch": 1.890238573800577,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00046425784277915363,
      "loss": 5.8647,
      "mean_token_accuracy": 0.15411169528961183,
      "num_tokens": 43811475.0,
      "step": 21630
    },
    {
      "entropy": 6.137119913101197,
      "epoch": 1.8906755221532814,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004642409887882528,
      "loss": 5.77,
      "mean_token_accuracy": 0.15606426447629929,
      "num_tokens": 43820811.0,
      "step": 21635
    },
    {
      "entropy": 6.13551697731018,
      "epoch": 1.8911124705059863,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004642241311676107,
      "loss": 5.8625,
      "mean_token_accuracy": 0.14912659749388696,
      "num_tokens": 43830804.0,
      "step": 21640
    },
    {
      "entropy": 6.0241631984710695,
      "epoch": 1.8915494188586908,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004642072699175506,
      "loss": 5.6519,
      "mean_token_accuracy": 0.1744380623102188,
      "num_tokens": 43840156.0,
      "step": 21645
    },
    {
      "entropy": 6.089308834075927,
      "epoch": 1.8919863672113957,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000464190405038396,
      "loss": 5.7481,
      "mean_token_accuracy": 0.16399511843919753,
      "num_tokens": 43849622.0,
      "step": 21650
    },
    {
      "entropy": 6.1434211254119875,
      "epoch": 1.8924233155641004,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004641735365304703,
      "loss": 5.8234,
      "mean_token_accuracy": 0.15403493195772172,
      "num_tokens": 43860184.0,
      "step": 21655
    },
    {
      "entropy": 5.990828561782837,
      "epoch": 1.892860263916805,
      "grad_norm": 1.671875,
      "learning_rate": 0.00046415666439409694,
      "loss": 5.6656,
      "mean_token_accuracy": 0.1788930132985115,
      "num_tokens": 43870157.0,
      "step": 21660
    },
    {
      "entropy": 6.091065073013306,
      "epoch": 1.8932972122695098,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046413978862959964,
      "loss": 5.7412,
      "mean_token_accuracy": 0.15863099843263626,
      "num_tokens": 43879867.0,
      "step": 21665
    },
    {
      "entropy": 6.069249963760376,
      "epoch": 1.8937341606222144,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004641229092373022,
      "loss": 5.7647,
      "mean_token_accuracy": 0.16559610068798064,
      "num_tokens": 43890314.0,
      "step": 21670
    },
    {
      "entropy": 6.145177793502808,
      "epoch": 1.8941711089749191,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004641060262175282,
      "loss": 5.8196,
      "mean_token_accuracy": 0.15627628266811372,
      "num_tokens": 43901075.0,
      "step": 21675
    },
    {
      "entropy": 6.1337498188018795,
      "epoch": 1.8946080573276238,
      "grad_norm": 2.46875,
      "learning_rate": 0.00046408913957060144,
      "loss": 5.8333,
      "mean_token_accuracy": 0.15639868527650833,
      "num_tokens": 43911777.0,
      "step": 21680
    },
    {
      "entropy": 6.115855550765991,
      "epoch": 1.8950450056803287,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046407224929684597,
      "loss": 5.7203,
      "mean_token_accuracy": 0.1649986445903778,
      "num_tokens": 43921901.0,
      "step": 21685
    },
    {
      "entropy": 6.075071477890015,
      "epoch": 1.8954819540330332,
      "grad_norm": 2.25,
      "learning_rate": 0.0004640553553965857,
      "loss": 5.8108,
      "mean_token_accuracy": 0.1567857265472412,
      "num_tokens": 43931142.0,
      "step": 21690
    },
    {
      "entropy": 6.104142475128174,
      "epoch": 1.8959189023857381,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004640384578701447,
      "loss": 5.764,
      "mean_token_accuracy": 0.16318970173597336,
      "num_tokens": 43940642.0,
      "step": 21695
    },
    {
      "entropy": 6.1351892948150635,
      "epoch": 1.8963558507384426,
      "grad_norm": 1.90625,
      "learning_rate": 0.00046402155671784696,
      "loss": 5.8041,
      "mean_token_accuracy": 0.1547599896788597,
      "num_tokens": 43950808.0,
      "step": 21700
    },
    {
      "entropy": 6.200026941299439,
      "epoch": 1.8967927990911475,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046400465194001674,
      "loss": 5.9107,
      "mean_token_accuracy": 0.15347514301538467,
      "num_tokens": 43961349.0,
      "step": 21705
    },
    {
      "entropy": 6.1740327835083,
      "epoch": 1.897229747443852,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046398774353697826,
      "loss": 5.7471,
      "mean_token_accuracy": 0.16320045441389083,
      "num_tokens": 43970072.0,
      "step": 21710
    },
    {
      "entropy": 6.117642736434936,
      "epoch": 1.897666695796557,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004639708315090558,
      "loss": 5.8372,
      "mean_token_accuracy": 0.15671235620975493,
      "num_tokens": 43980076.0,
      "step": 21715
    },
    {
      "entropy": 6.046651220321655,
      "epoch": 1.8981036441492616,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004639539158565738,
      "loss": 5.6621,
      "mean_token_accuracy": 0.17301945835351945,
      "num_tokens": 43989876.0,
      "step": 21720
    },
    {
      "entropy": 6.134624671936035,
      "epoch": 1.8985405925019663,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004639369965798567,
      "loss": 5.8577,
      "mean_token_accuracy": 0.15149969905614852,
      "num_tokens": 44000920.0,
      "step": 21725
    },
    {
      "entropy": 6.070005464553833,
      "epoch": 1.898977540854671,
      "grad_norm": 1.84375,
      "learning_rate": 0.00046392007367922894,
      "loss": 5.6958,
      "mean_token_accuracy": 0.16323314607143402,
      "num_tokens": 44011539.0,
      "step": 21730
    },
    {
      "entropy": 6.081657457351684,
      "epoch": 1.8994144892073757,
      "grad_norm": 1.625,
      "learning_rate": 0.0004639031471550152,
      "loss": 5.7792,
      "mean_token_accuracy": 0.16068895757198334,
      "num_tokens": 44022624.0,
      "step": 21735
    },
    {
      "entropy": 6.061473178863525,
      "epoch": 1.8998514375600803,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00046388621700754014,
      "loss": 5.7505,
      "mean_token_accuracy": 0.16548720896244049,
      "num_tokens": 44033105.0,
      "step": 21740
    },
    {
      "entropy": 6.091430807113648,
      "epoch": 1.900288385912785,
      "grad_norm": 1.765625,
      "learning_rate": 0.00046386928323712847,
      "loss": 5.7542,
      "mean_token_accuracy": 0.15974067151546478,
      "num_tokens": 44042805.0,
      "step": 21745
    },
    {
      "entropy": 6.058866834640503,
      "epoch": 1.90072533426549,
      "grad_norm": 1.625,
      "learning_rate": 0.000463852345844105,
      "loss": 5.7115,
      "mean_token_accuracy": 0.17621221542358398,
      "num_tokens": 44052471.0,
      "step": 21750
    },
    {
      "entropy": 6.091930866241455,
      "epoch": 1.9011622826181944,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004638354048287945,
      "loss": 5.8158,
      "mean_token_accuracy": 0.15638008564710618,
      "num_tokens": 44063358.0,
      "step": 21755
    },
    {
      "entropy": 6.085887861251831,
      "epoch": 1.9015992309708993,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046381846019152194,
      "loss": 5.796,
      "mean_token_accuracy": 0.15400446727871894,
      "num_tokens": 44073598.0,
      "step": 21760
    },
    {
      "entropy": 6.159935569763183,
      "epoch": 1.9020361793236038,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046380151193261247,
      "loss": 5.8308,
      "mean_token_accuracy": 0.16072463542222976,
      "num_tokens": 44084752.0,
      "step": 21765
    },
    {
      "entropy": 6.071489953994751,
      "epoch": 1.9024731276763087,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004637845600523911,
      "loss": 5.7435,
      "mean_token_accuracy": 0.16512414067983627,
      "num_tokens": 44094857.0,
      "step": 21770
    },
    {
      "entropy": 6.161677265167237,
      "epoch": 1.9029100760290134,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004637676045511829,
      "loss": 5.8178,
      "mean_token_accuracy": 0.15860415995121002,
      "num_tokens": 44105700.0,
      "step": 21775
    },
    {
      "entropy": 6.13016905784607,
      "epoch": 1.903347024381718,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004637506454293131,
      "loss": 5.7662,
      "mean_token_accuracy": 0.15920145064592361,
      "num_tokens": 44116588.0,
      "step": 21780
    },
    {
      "entropy": 6.101988935470581,
      "epoch": 1.9037839727344228,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004637336826871071,
      "loss": 5.7924,
      "mean_token_accuracy": 0.16125551909208297,
      "num_tokens": 44126948.0,
      "step": 21785
    },
    {
      "entropy": 6.15350866317749,
      "epoch": 1.9042209210871275,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046371671632489017,
      "loss": 5.839,
      "mean_token_accuracy": 0.15830574482679366,
      "num_tokens": 44136239.0,
      "step": 21790
    },
    {
      "entropy": 6.075198459625244,
      "epoch": 1.9046578694398322,
      "grad_norm": 1.5,
      "learning_rate": 0.00046369974634298766,
      "loss": 5.6734,
      "mean_token_accuracy": 0.16757322400808333,
      "num_tokens": 44146314.0,
      "step": 21795
    },
    {
      "entropy": 6.046156692504883,
      "epoch": 1.9050948177925369,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004636827727417252,
      "loss": 5.8147,
      "mean_token_accuracy": 0.16116336286067962,
      "num_tokens": 44155767.0,
      "step": 21800
    },
    {
      "entropy": 6.151869487762451,
      "epoch": 1.9055317661452418,
      "grad_norm": 2.125,
      "learning_rate": 0.00046366579552142823,
      "loss": 5.8194,
      "mean_token_accuracy": 0.1616693302989006,
      "num_tokens": 44165830.0,
      "step": 21805
    },
    {
      "entropy": 6.054533433914185,
      "epoch": 1.9059687144979462,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004636488146824225,
      "loss": 5.6629,
      "mean_token_accuracy": 0.1686060532927513,
      "num_tokens": 44175526.0,
      "step": 21810
    },
    {
      "entropy": 6.112943506240844,
      "epoch": 1.9064056628506512,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004636318302250336,
      "loss": 5.8039,
      "mean_token_accuracy": 0.1601998284459114,
      "num_tokens": 44185950.0,
      "step": 21815
    },
    {
      "entropy": 6.158704853057861,
      "epoch": 1.9068426112033556,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004636148421495874,
      "loss": 5.726,
      "mean_token_accuracy": 0.16330220252275468,
      "num_tokens": 44196518.0,
      "step": 21820
    },
    {
      "entropy": 6.146063709259034,
      "epoch": 1.9072795595560605,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046359785045640967,
      "loss": 5.8727,
      "mean_token_accuracy": 0.15643210709095,
      "num_tokens": 44207709.0,
      "step": 21825
    },
    {
      "entropy": 6.132011985778808,
      "epoch": 1.9077165079087652,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004635808551458264,
      "loss": 5.8266,
      "mean_token_accuracy": 0.15349278450012208,
      "num_tokens": 44218311.0,
      "step": 21830
    },
    {
      "entropy": 6.173475646972657,
      "epoch": 1.90815345626147,
      "grad_norm": 2.640625,
      "learning_rate": 0.00046356385621816343,
      "loss": 5.7735,
      "mean_token_accuracy": 0.1582676112651825,
      "num_tokens": 44229059.0,
      "step": 21835
    },
    {
      "entropy": 6.112079620361328,
      "epoch": 1.9085904046141746,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004635468536737469,
      "loss": 5.7736,
      "mean_token_accuracy": 0.15432663410902023,
      "num_tokens": 44239254.0,
      "step": 21840
    },
    {
      "entropy": 6.075419807434082,
      "epoch": 1.9090273529668793,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00046352984751290297,
      "loss": 5.6596,
      "mean_token_accuracy": 0.16641340553760528,
      "num_tokens": 44248465.0,
      "step": 21845
    },
    {
      "entropy": 6.057498264312744,
      "epoch": 1.909464301319584,
      "grad_norm": 1.921875,
      "learning_rate": 0.00046351283773595777,
      "loss": 5.7708,
      "mean_token_accuracy": 0.15571195632219315,
      "num_tokens": 44259121.0,
      "step": 21850
    },
    {
      "entropy": 6.111133337020874,
      "epoch": 1.9099012496722887,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046349582434323757,
      "loss": 5.7668,
      "mean_token_accuracy": 0.16955541223287582,
      "num_tokens": 44269291.0,
      "step": 21855
    },
    {
      "entropy": 6.098084545135498,
      "epoch": 1.9103381980249936,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00046347880733506867,
      "loss": 5.7714,
      "mean_token_accuracy": 0.15944246351718902,
      "num_tokens": 44280985.0,
      "step": 21860
    },
    {
      "entropy": 6.06198902130127,
      "epoch": 1.910775146377698,
      "grad_norm": 2.703125,
      "learning_rate": 0.00046346178671177746,
      "loss": 5.7601,
      "mean_token_accuracy": 0.16336750686168672,
      "num_tokens": 44290370.0,
      "step": 21865
    },
    {
      "entropy": 6.039211130142212,
      "epoch": 1.911212094730403,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00046344476247369044,
      "loss": 5.7615,
      "mean_token_accuracy": 0.1621663585305214,
      "num_tokens": 44300670.0,
      "step": 21870
    },
    {
      "entropy": 6.116272878646851,
      "epoch": 1.9116490430831075,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046342773462113414,
      "loss": 5.7776,
      "mean_token_accuracy": 0.1633043810725212,
      "num_tokens": 44311578.0,
      "step": 21875
    },
    {
      "entropy": 6.065890216827393,
      "epoch": 1.9120859914358124,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046341070315443516,
      "loss": 5.6987,
      "mean_token_accuracy": 0.165633362531662,
      "num_tokens": 44321692.0,
      "step": 21880
    },
    {
      "entropy": 6.035878038406372,
      "epoch": 1.9125229397885168,
      "grad_norm": 2.265625,
      "learning_rate": 0.00046339366807392017,
      "loss": 5.7403,
      "mean_token_accuracy": 0.1676764264702797,
      "num_tokens": 44331468.0,
      "step": 21885
    },
    {
      "entropy": 6.077155733108521,
      "epoch": 1.9129598881412218,
      "grad_norm": 2.71875,
      "learning_rate": 0.00046337662937991585,
      "loss": 5.7885,
      "mean_token_accuracy": 0.162793393433094,
      "num_tokens": 44341491.0,
      "step": 21890
    },
    {
      "entropy": 6.080180263519287,
      "epoch": 1.9133968364939264,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004633595870727492,
      "loss": 5.8517,
      "mean_token_accuracy": 0.15628968328237533,
      "num_tokens": 44351568.0,
      "step": 21895
    },
    {
      "entropy": 6.20089864730835,
      "epoch": 1.9138337848466311,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046334254115274685,
      "loss": 5.8463,
      "mean_token_accuracy": 0.16026855558156966,
      "num_tokens": 44362009.0,
      "step": 21900
    },
    {
      "entropy": 6.106795310974121,
      "epoch": 1.9142707331993358,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004633254916202359,
      "loss": 5.7243,
      "mean_token_accuracy": 0.16975750923156738,
      "num_tokens": 44372671.0,
      "step": 21905
    },
    {
      "entropy": 6.062414455413818,
      "epoch": 1.9147076815520405,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004633084384755434,
      "loss": 5.7323,
      "mean_token_accuracy": 0.1693192332983017,
      "num_tokens": 44382658.0,
      "step": 21910
    },
    {
      "entropy": 6.11456561088562,
      "epoch": 1.9151446299047452,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004632913817189963,
      "loss": 5.7505,
      "mean_token_accuracy": 0.1679163783788681,
      "num_tokens": 44392090.0,
      "step": 21915
    },
    {
      "entropy": 6.126551771163941,
      "epoch": 1.91558157825745,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004632743213509219,
      "loss": 5.8389,
      "mean_token_accuracy": 0.15136836916208268,
      "num_tokens": 44402896.0,
      "step": 21920
    },
    {
      "entropy": 6.179987525939941,
      "epoch": 1.9160185266101548,
      "grad_norm": 1.71875,
      "learning_rate": 0.00046325725737164734,
      "loss": 5.8052,
      "mean_token_accuracy": 0.16037040799856186,
      "num_tokens": 44413700.0,
      "step": 21925
    },
    {
      "entropy": 6.075572443008423,
      "epoch": 1.9164554749628593,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004632401897814999,
      "loss": 5.6993,
      "mean_token_accuracy": 0.17137067019939423,
      "num_tokens": 44423442.0,
      "step": 21930
    },
    {
      "entropy": 5.992259120941162,
      "epoch": 1.9168924233155642,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004632231185808071,
      "loss": 5.7597,
      "mean_token_accuracy": 0.16723961085081102,
      "num_tokens": 44434197.0,
      "step": 21935
    },
    {
      "entropy": 6.1544146060943605,
      "epoch": 1.9173293716682687,
      "grad_norm": 1.828125,
      "learning_rate": 0.00046320604376989604,
      "loss": 5.7511,
      "mean_token_accuracy": 0.16442470997571945,
      "num_tokens": 44444483.0,
      "step": 21940
    },
    {
      "entropy": 6.105314159393311,
      "epoch": 1.9177663200209736,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004631889653490946,
      "loss": 5.7787,
      "mean_token_accuracy": 0.16129601895809173,
      "num_tokens": 44456303.0,
      "step": 21945
    },
    {
      "entropy": 6.056937217712402,
      "epoch": 1.9182032683736783,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046317188331873024,
      "loss": 5.8132,
      "mean_token_accuracy": 0.1640782356262207,
      "num_tokens": 44467314.0,
      "step": 21950
    },
    {
      "entropy": 6.0934751510620115,
      "epoch": 1.918640216726383,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004631547976791305,
      "loss": 5.7651,
      "mean_token_accuracy": 0.16648557633161545,
      "num_tokens": 44476466.0,
      "step": 21955
    },
    {
      "entropy": 6.093678617477417,
      "epoch": 1.9190771650790877,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004631377084306231,
      "loss": 5.771,
      "mean_token_accuracy": 0.16118590384721757,
      "num_tokens": 44486247.0,
      "step": 21960
    },
    {
      "entropy": 6.09524884223938,
      "epoch": 1.9195141134317923,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00046312061557353585,
      "loss": 5.7681,
      "mean_token_accuracy": 0.16380950808525085,
      "num_tokens": 44497166.0,
      "step": 21965
    },
    {
      "entropy": 6.1303809642791744,
      "epoch": 1.919951061784497,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004631035191081967,
      "loss": 5.7964,
      "mean_token_accuracy": 0.15627716928720475,
      "num_tokens": 44507677.0,
      "step": 21970
    },
    {
      "entropy": 6.066018390655517,
      "epoch": 1.9203880101372017,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004630864190349334,
      "loss": 5.7365,
      "mean_token_accuracy": 0.16496471911668778,
      "num_tokens": 44517392.0,
      "step": 21975
    },
    {
      "entropy": 6.209807062149048,
      "epoch": 1.9208249584899066,
      "grad_norm": 1.890625,
      "learning_rate": 0.00046306931535407416,
      "loss": 5.7997,
      "mean_token_accuracy": 0.16254190653562545,
      "num_tokens": 44528914.0,
      "step": 21980
    },
    {
      "entropy": 6.18611707687378,
      "epoch": 1.9212619068426111,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00046305220806594683,
      "loss": 5.8791,
      "mean_token_accuracy": 0.15022816956043245,
      "num_tokens": 44539419.0,
      "step": 21985
    },
    {
      "entropy": 6.1022804260253904,
      "epoch": 1.921698855195316,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004630350971708797,
      "loss": 5.7539,
      "mean_token_accuracy": 0.167881940305233,
      "num_tokens": 44548704.0,
      "step": 21990
    },
    {
      "entropy": 6.136773204803466,
      "epoch": 1.9221358035480205,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00046301798266920076,
      "loss": 5.7557,
      "mean_token_accuracy": 0.16054175198078155,
      "num_tokens": 44559110.0,
      "step": 21995
    },
    {
      "entropy": 6.132760620117187,
      "epoch": 1.9225727519007254,
      "grad_norm": 2.125,
      "learning_rate": 0.0004630008645612384,
      "loss": 5.839,
      "mean_token_accuracy": 0.15543888211250306,
      "num_tokens": 44570414.0,
      "step": 22000
    },
    {
      "entropy": 6.137655019760132,
      "epoch": 1.92300970025343,
      "grad_norm": 1.796875,
      "learning_rate": 0.00046298374284732096,
      "loss": 5.8916,
      "mean_token_accuracy": 0.1546381965279579,
      "num_tokens": 44580194.0,
      "step": 22005
    },
    {
      "entropy": 6.0906298637390135,
      "epoch": 1.9234466486061348,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004629666175277768,
      "loss": 5.7563,
      "mean_token_accuracy": 0.16483185887336732,
      "num_tokens": 44590174.0,
      "step": 22010
    },
    {
      "entropy": 6.135630750656128,
      "epoch": 1.9238835969588395,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004629494886029345,
      "loss": 5.8523,
      "mean_token_accuracy": 0.15479106307029725,
      "num_tokens": 44600940.0,
      "step": 22015
    },
    {
      "entropy": 6.151897048950195,
      "epoch": 1.9243205453115442,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004629323560731225,
      "loss": 5.8285,
      "mean_token_accuracy": 0.1601486660540104,
      "num_tokens": 44610852.0,
      "step": 22020
    },
    {
      "entropy": 6.171092224121094,
      "epoch": 1.9247574936642489,
      "grad_norm": 1.984375,
      "learning_rate": 0.00046291521993866943,
      "loss": 5.8613,
      "mean_token_accuracy": 0.15577651411294938,
      "num_tokens": 44621479.0,
      "step": 22025
    },
    {
      "entropy": 6.133688259124756,
      "epoch": 1.9251944420169536,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000462898080199904,
      "loss": 5.7861,
      "mean_token_accuracy": 0.1618320792913437,
      "num_tokens": 44631838.0,
      "step": 22030
    },
    {
      "entropy": 6.052557134628296,
      "epoch": 1.9256313903696585,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004628809368571549,
      "loss": 5.7538,
      "mean_token_accuracy": 0.15958594679832458,
      "num_tokens": 44643103.0,
      "step": 22035
    },
    {
      "entropy": 6.153899765014648,
      "epoch": 1.926068338722363,
      "grad_norm": 3.265625,
      "learning_rate": 0.00046286378991075095,
      "loss": 5.9029,
      "mean_token_accuracy": 0.15257836729288102,
      "num_tokens": 44653659.0,
      "step": 22040
    },
    {
      "entropy": 6.1691399097442625,
      "epoch": 1.9265052870750679,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004628466393610211,
      "loss": 5.766,
      "mean_token_accuracy": 0.15681518167257308,
      "num_tokens": 44665064.0,
      "step": 22045
    },
    {
      "entropy": 6.10623140335083,
      "epoch": 1.9269422354277723,
      "grad_norm": 2.828125,
      "learning_rate": 0.00046282948520829433,
      "loss": 5.7447,
      "mean_token_accuracy": 0.1657790094614029,
      "num_tokens": 44674256.0,
      "step": 22050
    },
    {
      "entropy": 6.026172685623169,
      "epoch": 1.9273791837804772,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004628123274528996,
      "loss": 5.7162,
      "mean_token_accuracy": 0.16533030867576598,
      "num_tokens": 44683715.0,
      "step": 22055
    },
    {
      "entropy": 6.157532691955566,
      "epoch": 1.9278161321331817,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046279516609516594,
      "loss": 5.8866,
      "mean_token_accuracy": 0.15124964565038682,
      "num_tokens": 44694637.0,
      "step": 22060
    },
    {
      "entropy": 6.183302307128907,
      "epoch": 1.9282530804858866,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004627780011354226,
      "loss": 5.7752,
      "mean_token_accuracy": 0.1666324704885483,
      "num_tokens": 44705307.0,
      "step": 22065
    },
    {
      "entropy": 6.138606357574463,
      "epoch": 1.9286900288385913,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004627608325739989,
      "loss": 5.7971,
      "mean_token_accuracy": 0.15701915323734283,
      "num_tokens": 44715293.0,
      "step": 22070
    },
    {
      "entropy": 6.119704675674439,
      "epoch": 1.929126977191296,
      "grad_norm": 2.21875,
      "learning_rate": 0.000462743660411224,
      "loss": 5.8792,
      "mean_token_accuracy": 0.14809797406196595,
      "num_tokens": 44726406.0,
      "step": 22075
    },
    {
      "entropy": 6.172959232330323,
      "epoch": 1.9295639255440007,
      "grad_norm": 2.359375,
      "learning_rate": 0.00046272648464742733,
      "loss": 5.7899,
      "mean_token_accuracy": 0.16341413408517838,
      "num_tokens": 44737166.0,
      "step": 22080
    },
    {
      "entropy": 6.07903470993042,
      "epoch": 1.9300008738967054,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00046270930528293826,
      "loss": 5.6842,
      "mean_token_accuracy": 0.16352759152650834,
      "num_tokens": 44746693.0,
      "step": 22085
    },
    {
      "entropy": 6.089733457565307,
      "epoch": 1.93043782224941,
      "grad_norm": 2.375,
      "learning_rate": 0.0004626921223180864,
      "loss": 5.7309,
      "mean_token_accuracy": 0.16394534558057786,
      "num_tokens": 44757499.0,
      "step": 22090
    },
    {
      "entropy": 6.103984308242798,
      "epoch": 1.9308747706021148,
      "grad_norm": 1.875,
      "learning_rate": 0.00046267493575320127,
      "loss": 5.7675,
      "mean_token_accuracy": 0.1596020832657814,
      "num_tokens": 44768112.0,
      "step": 22095
    },
    {
      "entropy": 6.061951589584351,
      "epoch": 1.9313117189548197,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004626577455886125,
      "loss": 5.7065,
      "mean_token_accuracy": 0.15942114293575288,
      "num_tokens": 44778317.0,
      "step": 22100
    },
    {
      "entropy": 6.122918176651001,
      "epoch": 1.9317486673075241,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004626405518246499,
      "loss": 5.8382,
      "mean_token_accuracy": 0.1519172042608261,
      "num_tokens": 44790483.0,
      "step": 22105
    },
    {
      "entropy": 6.082853364944458,
      "epoch": 1.932185615660229,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00046262335446164315,
      "loss": 5.711,
      "mean_token_accuracy": 0.17167649865150453,
      "num_tokens": 44799891.0,
      "step": 22110
    },
    {
      "entropy": 6.0961119651794435,
      "epoch": 1.9326225640129335,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004626061534999222,
      "loss": 5.8418,
      "mean_token_accuracy": 0.15664188414812089,
      "num_tokens": 44810728.0,
      "step": 22115
    },
    {
      "entropy": 6.140141582489013,
      "epoch": 1.9330595123656384,
      "grad_norm": 3.015625,
      "learning_rate": 0.00046258894893981686,
      "loss": 5.8229,
      "mean_token_accuracy": 0.15805481970310212,
      "num_tokens": 44820597.0,
      "step": 22120
    },
    {
      "entropy": 6.135374450683594,
      "epoch": 1.9334964607183431,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004625717407816572,
      "loss": 5.8083,
      "mean_token_accuracy": 0.15865223109722137,
      "num_tokens": 44832570.0,
      "step": 22125
    },
    {
      "entropy": 6.144927835464477,
      "epoch": 1.9339334090710478,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004625545290257733,
      "loss": 5.8247,
      "mean_token_accuracy": 0.15994991064071656,
      "num_tokens": 44841936.0,
      "step": 22130
    },
    {
      "entropy": 6.047894287109375,
      "epoch": 1.9343703574237525,
      "grad_norm": 1.6875,
      "learning_rate": 0.00046253731367249527,
      "loss": 5.7289,
      "mean_token_accuracy": 0.1726667985320091,
      "num_tokens": 44852329.0,
      "step": 22135
    },
    {
      "entropy": 6.107885837554932,
      "epoch": 1.9348073057764572,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004625200947221532,
      "loss": 5.7717,
      "mean_token_accuracy": 0.15672435835003853,
      "num_tokens": 44862183.0,
      "step": 22140
    },
    {
      "entropy": 6.126272773742675,
      "epoch": 1.935244254129162,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004625028721750776,
      "loss": 5.7984,
      "mean_token_accuracy": 0.15121425986289977,
      "num_tokens": 44872712.0,
      "step": 22145
    },
    {
      "entropy": 6.0679951190948485,
      "epoch": 1.9356812024818666,
      "grad_norm": 4.3125,
      "learning_rate": 0.00046248564603159855,
      "loss": 5.8154,
      "mean_token_accuracy": 0.16205878630280496,
      "num_tokens": 44884423.0,
      "step": 22150
    },
    {
      "entropy": 6.13159065246582,
      "epoch": 1.9361181508345715,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004624684162920466,
      "loss": 5.7664,
      "mean_token_accuracy": 0.1572164088487625,
      "num_tokens": 44894556.0,
      "step": 22155
    },
    {
      "entropy": 6.050223350524902,
      "epoch": 1.936555099187276,
      "grad_norm": 2.375,
      "learning_rate": 0.00046245118295675213,
      "loss": 5.7279,
      "mean_token_accuracy": 0.16140951067209244,
      "num_tokens": 44905208.0,
      "step": 22160
    },
    {
      "entropy": 6.0173999786376955,
      "epoch": 1.9369920475399809,
      "grad_norm": 1.734375,
      "learning_rate": 0.00046243394602604585,
      "loss": 5.7428,
      "mean_token_accuracy": 0.16098992079496383,
      "num_tokens": 44915674.0,
      "step": 22165
    },
    {
      "entropy": 6.103794288635254,
      "epoch": 1.9374289958926854,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004624167055002582,
      "loss": 5.7829,
      "mean_token_accuracy": 0.16225987672805786,
      "num_tokens": 44926382.0,
      "step": 22170
    },
    {
      "entropy": 6.1749267578125,
      "epoch": 1.9378659442453903,
      "grad_norm": 1.921875,
      "learning_rate": 0.00046239946137971997,
      "loss": 5.7029,
      "mean_token_accuracy": 0.1779550462961197,
      "num_tokens": 44935963.0,
      "step": 22175
    },
    {
      "entropy": 6.185078477859497,
      "epoch": 1.9383028925980947,
      "grad_norm": 2.125,
      "learning_rate": 0.00046238221366476183,
      "loss": 5.8228,
      "mean_token_accuracy": 0.15476867109537124,
      "num_tokens": 44946312.0,
      "step": 22180
    },
    {
      "entropy": 6.025645685195923,
      "epoch": 1.9387398409507997,
      "grad_norm": 1.953125,
      "learning_rate": 0.00046236496235571475,
      "loss": 5.7414,
      "mean_token_accuracy": 0.16805175542831421,
      "num_tokens": 44956213.0,
      "step": 22185
    },
    {
      "entropy": 6.088394355773926,
      "epoch": 1.9391767893035043,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004623477074529094,
      "loss": 5.9029,
      "mean_token_accuracy": 0.14928019493818284,
      "num_tokens": 44966575.0,
      "step": 22190
    },
    {
      "entropy": 6.167176914215088,
      "epoch": 1.939613737656209,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004623304489566769,
      "loss": 5.8145,
      "mean_token_accuracy": 0.16052188724279404,
      "num_tokens": 44976353.0,
      "step": 22195
    },
    {
      "entropy": 6.1288879871368405,
      "epoch": 1.9400506860089137,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00046231318686734826,
      "loss": 5.7632,
      "mean_token_accuracy": 0.16271302103996277,
      "num_tokens": 44987046.0,
      "step": 22200
    },
    {
      "entropy": 6.136386585235596,
      "epoch": 1.9404876343616184,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004622959211852545,
      "loss": 5.7442,
      "mean_token_accuracy": 0.16020454317331315,
      "num_tokens": 44996606.0,
      "step": 22205
    },
    {
      "entropy": 6.164762878417969,
      "epoch": 1.9409245827143233,
      "grad_norm": 1.59375,
      "learning_rate": 0.00046227865191072684,
      "loss": 5.8292,
      "mean_token_accuracy": 0.15318823009729385,
      "num_tokens": 45006895.0,
      "step": 22210
    },
    {
      "entropy": 6.089016675949097,
      "epoch": 1.9413615310670278,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004622613790440965,
      "loss": 5.7215,
      "mean_token_accuracy": 0.16618750244379044,
      "num_tokens": 45017075.0,
      "step": 22215
    },
    {
      "entropy": 6.106326627731323,
      "epoch": 1.9417984794197327,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004622441025856948,
      "loss": 5.7597,
      "mean_token_accuracy": 0.16013819575309754,
      "num_tokens": 45027331.0,
      "step": 22220
    },
    {
      "entropy": 6.048409366607666,
      "epoch": 1.9422354277724372,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004622268225358531,
      "loss": 5.7623,
      "mean_token_accuracy": 0.15767046213150024,
      "num_tokens": 45037623.0,
      "step": 22225
    },
    {
      "entropy": 6.154530954360962,
      "epoch": 1.942672376125142,
      "grad_norm": 1.71875,
      "learning_rate": 0.00046220953889490287,
      "loss": 5.8148,
      "mean_token_accuracy": 0.15700974613428115,
      "num_tokens": 45048191.0,
      "step": 22230
    },
    {
      "entropy": 6.166302013397217,
      "epoch": 1.9431093244778466,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046219225166317554,
      "loss": 5.8067,
      "mean_token_accuracy": 0.16180141717195512,
      "num_tokens": 45057906.0,
      "step": 22235
    },
    {
      "entropy": 6.042593765258789,
      "epoch": 1.9435462728305515,
      "grad_norm": 1.96875,
      "learning_rate": 0.00046217496084100267,
      "loss": 5.786,
      "mean_token_accuracy": 0.17028869241476058,
      "num_tokens": 45067842.0,
      "step": 22240
    },
    {
      "entropy": 6.095809459686279,
      "epoch": 1.9439832211832562,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00046215766642871607,
      "loss": 5.755,
      "mean_token_accuracy": 0.16260093897581102,
      "num_tokens": 45077388.0,
      "step": 22245
    },
    {
      "entropy": 6.035816383361817,
      "epoch": 1.9444201695359609,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046214036842664736,
      "loss": 5.655,
      "mean_token_accuracy": 0.1734872505068779,
      "num_tokens": 45086001.0,
      "step": 22250
    },
    {
      "entropy": 6.083702135086059,
      "epoch": 1.9448571178886656,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004621230668351283,
      "loss": 5.7287,
      "mean_token_accuracy": 0.17018088400363923,
      "num_tokens": 45096203.0,
      "step": 22255
    },
    {
      "entropy": 6.057285928726197,
      "epoch": 1.9452940662413702,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00046210576165449076,
      "loss": 5.7092,
      "mean_token_accuracy": 0.16466959565877914,
      "num_tokens": 45104997.0,
      "step": 22260
    },
    {
      "entropy": 6.063611841201782,
      "epoch": 1.945731014594075,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00046208845288506667,
      "loss": 5.7631,
      "mean_token_accuracy": 0.16488433480262757,
      "num_tokens": 45114150.0,
      "step": 22265
    },
    {
      "entropy": 6.014978456497192,
      "epoch": 1.9461679629467796,
      "grad_norm": 1.8671875,
      "learning_rate": 0.000462071140527188,
      "loss": 5.7522,
      "mean_token_accuracy": 0.17244884371757507,
      "num_tokens": 45124668.0,
      "step": 22270
    },
    {
      "entropy": 6.107720232009887,
      "epoch": 1.9466049112994845,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004620538245811869,
      "loss": 5.7958,
      "mean_token_accuracy": 0.1599855564534664,
      "num_tokens": 45135164.0,
      "step": 22275
    },
    {
      "entropy": 6.043752145767212,
      "epoch": 1.947041859652189,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046203650504739543,
      "loss": 5.6194,
      "mean_token_accuracy": 0.1780226856470108,
      "num_tokens": 45144453.0,
      "step": 22280
    },
    {
      "entropy": 6.052005290985107,
      "epoch": 1.947478808004894,
      "grad_norm": 1.5,
      "learning_rate": 0.00046201918192614573,
      "loss": 5.8103,
      "mean_token_accuracy": 0.15398862659931184,
      "num_tokens": 45154557.0,
      "step": 22285
    },
    {
      "entropy": 6.110474729537964,
      "epoch": 1.9479157563575984,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004620018552177701,
      "loss": 5.7486,
      "mean_token_accuracy": 0.16008991450071336,
      "num_tokens": 45164938.0,
      "step": 22290
    },
    {
      "entropy": 6.1742030620574955,
      "epoch": 1.9483527047103033,
      "grad_norm": 2.375,
      "learning_rate": 0.00046198452492260096,
      "loss": 5.9106,
      "mean_token_accuracy": 0.14635787010192872,
      "num_tokens": 45175823.0,
      "step": 22295
    },
    {
      "entropy": 6.004029273986816,
      "epoch": 1.948789653063008,
      "grad_norm": 1.625,
      "learning_rate": 0.0004619671910409706,
      "loss": 5.6838,
      "mean_token_accuracy": 0.18009693026542664,
      "num_tokens": 45185589.0,
      "step": 22300
    },
    {
      "entropy": 6.081045484542846,
      "epoch": 1.9492266014157127,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004619498535732116,
      "loss": 5.8373,
      "mean_token_accuracy": 0.16004773527383803,
      "num_tokens": 45195654.0,
      "step": 22305
    },
    {
      "entropy": 6.224378442764282,
      "epoch": 1.9496635497684174,
      "grad_norm": 1.65625,
      "learning_rate": 0.00046193251251965646,
      "loss": 5.8538,
      "mean_token_accuracy": 0.15393090844154358,
      "num_tokens": 45205824.0,
      "step": 22310
    },
    {
      "entropy": 6.152261972427368,
      "epoch": 1.950100498121122,
      "grad_norm": 1.875,
      "learning_rate": 0.0004619151678806377,
      "loss": 5.7822,
      "mean_token_accuracy": 0.16502559334039688,
      "num_tokens": 45215228.0,
      "step": 22315
    },
    {
      "entropy": 6.054143190383911,
      "epoch": 1.9505374464738268,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004618978196564882,
      "loss": 5.6763,
      "mean_token_accuracy": 0.1652392104268074,
      "num_tokens": 45225043.0,
      "step": 22320
    },
    {
      "entropy": 6.106949758529663,
      "epoch": 1.9509743948265315,
      "grad_norm": 2.125,
      "learning_rate": 0.0004618804678475405,
      "loss": 5.7892,
      "mean_token_accuracy": 0.15750280022621155,
      "num_tokens": 45234837.0,
      "step": 22325
    },
    {
      "entropy": 6.088729429244995,
      "epoch": 1.9514113431792364,
      "grad_norm": 1.5,
      "learning_rate": 0.00046186311245412746,
      "loss": 5.6486,
      "mean_token_accuracy": 0.1685670033097267,
      "num_tokens": 45243540.0,
      "step": 22330
    },
    {
      "entropy": 6.06598916053772,
      "epoch": 1.9518482915319408,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004618457534765821,
      "loss": 5.7409,
      "mean_token_accuracy": 0.16102411597967148,
      "num_tokens": 45253053.0,
      "step": 22335
    },
    {
      "entropy": 6.039823532104492,
      "epoch": 1.9522852398846458,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004618283909152372,
      "loss": 5.6778,
      "mean_token_accuracy": 0.16668858975172043,
      "num_tokens": 45262973.0,
      "step": 22340
    },
    {
      "entropy": 6.012207794189453,
      "epoch": 1.9527221882373502,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004618110247704259,
      "loss": 5.7012,
      "mean_token_accuracy": 0.17069580554962158,
      "num_tokens": 45272951.0,
      "step": 22345
    },
    {
      "entropy": 6.1274542808532715,
      "epoch": 1.9531591365900551,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00046179365504248125,
      "loss": 5.8297,
      "mean_token_accuracy": 0.16335624307394028,
      "num_tokens": 45283596.0,
      "step": 22350
    },
    {
      "entropy": 6.15087480545044,
      "epoch": 1.9535960849427596,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046177628173173647,
      "loss": 5.6647,
      "mean_token_accuracy": 0.17282828837633132,
      "num_tokens": 45294712.0,
      "step": 22355
    },
    {
      "entropy": 6.044143486022949,
      "epoch": 1.9540330332954645,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004617589048385247,
      "loss": 5.7772,
      "mean_token_accuracy": 0.15929984599351882,
      "num_tokens": 45305681.0,
      "step": 22360
    },
    {
      "entropy": 6.037092018127441,
      "epoch": 1.9544699816481692,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004617415243631793,
      "loss": 5.687,
      "mean_token_accuracy": 0.1705254763364792,
      "num_tokens": 45315332.0,
      "step": 22365
    },
    {
      "entropy": 6.0008275508880615,
      "epoch": 1.954906930000874,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00046172414030603356,
      "loss": 5.686,
      "mean_token_accuracy": 0.16844792813062667,
      "num_tokens": 45324526.0,
      "step": 22370
    },
    {
      "entropy": 6.124578809738159,
      "epoch": 1.9553438783535786,
      "grad_norm": 2.03125,
      "learning_rate": 0.000461706752667421,
      "loss": 5.74,
      "mean_token_accuracy": 0.16463537961244584,
      "num_tokens": 45333989.0,
      "step": 22375
    },
    {
      "entropy": 6.078993129730224,
      "epoch": 1.9557808267062833,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004616893614476751,
      "loss": 5.6996,
      "mean_token_accuracy": 0.16351039856672286,
      "num_tokens": 45344545.0,
      "step": 22380
    },
    {
      "entropy": 6.087796258926391,
      "epoch": 1.956217775058988,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00046167196664712945,
      "loss": 5.7627,
      "mean_token_accuracy": 0.15941549688577653,
      "num_tokens": 45354125.0,
      "step": 22385
    },
    {
      "entropy": 6.169203567504883,
      "epoch": 1.9566547234116927,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046165456826611763,
      "loss": 5.8143,
      "mean_token_accuracy": 0.15324278473854064,
      "num_tokens": 45364577.0,
      "step": 22390
    },
    {
      "entropy": 6.090648746490478,
      "epoch": 1.9570916717643976,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046163716630497336,
      "loss": 5.7967,
      "mean_token_accuracy": 0.16701366007328033,
      "num_tokens": 45374655.0,
      "step": 22395
    },
    {
      "entropy": 6.078590822219849,
      "epoch": 1.957528620117102,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046161976076403047,
      "loss": 5.7557,
      "mean_token_accuracy": 0.15947119295597076,
      "num_tokens": 45384383.0,
      "step": 22400
    },
    {
      "entropy": 6.065071249008179,
      "epoch": 1.957965568469807,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004616023516436228,
      "loss": 5.7547,
      "mean_token_accuracy": 0.16258560568094255,
      "num_tokens": 45394645.0,
      "step": 22405
    },
    {
      "entropy": 6.08292064666748,
      "epoch": 1.9584025168225114,
      "grad_norm": 1.90625,
      "learning_rate": 0.00046158493894408423,
      "loss": 5.705,
      "mean_token_accuracy": 0.16788882613182068,
      "num_tokens": 45404661.0,
      "step": 22410
    },
    {
      "entropy": 6.104598617553711,
      "epoch": 1.9588394651752163,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004615675226657487,
      "loss": 5.8396,
      "mean_token_accuracy": 0.16154442280530928,
      "num_tokens": 45413915.0,
      "step": 22415
    },
    {
      "entropy": 6.174936199188233,
      "epoch": 1.959276413527921,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004615501028089504,
      "loss": 5.7615,
      "mean_token_accuracy": 0.15446306020021439,
      "num_tokens": 45424019.0,
      "step": 22420
    },
    {
      "entropy": 6.078040218353271,
      "epoch": 1.9597133618806257,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004615326793740233,
      "loss": 5.7303,
      "mean_token_accuracy": 0.16038698703050613,
      "num_tokens": 45435087.0,
      "step": 22425
    },
    {
      "entropy": 6.121442127227783,
      "epoch": 1.9601503102333304,
      "grad_norm": 1.65625,
      "learning_rate": 0.00046151525236130174,
      "loss": 5.8559,
      "mean_token_accuracy": 0.15621094703674315,
      "num_tokens": 45446363.0,
      "step": 22430
    },
    {
      "entropy": 6.139925813674926,
      "epoch": 1.960587258586035,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004614978217711199,
      "loss": 5.7449,
      "mean_token_accuracy": 0.16370547115802764,
      "num_tokens": 45456268.0,
      "step": 22435
    },
    {
      "entropy": 6.010792684555054,
      "epoch": 1.9610242069387398,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00046148038760381217,
      "loss": 5.6924,
      "mean_token_accuracy": 0.1684295803308487,
      "num_tokens": 45466335.0,
      "step": 22440
    },
    {
      "entropy": 6.058513259887695,
      "epoch": 1.9614611552914445,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004614629498597128,
      "loss": 5.75,
      "mean_token_accuracy": 0.15821104794740676,
      "num_tokens": 45476128.0,
      "step": 22445
    },
    {
      "entropy": 6.108731317520141,
      "epoch": 1.9618981036441494,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004614455085391564,
      "loss": 5.6937,
      "mean_token_accuracy": 0.1670481577515602,
      "num_tokens": 45485986.0,
      "step": 22450
    },
    {
      "entropy": 6.0783614158630375,
      "epoch": 1.9623350519968539,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004614280636424774,
      "loss": 5.7597,
      "mean_token_accuracy": 0.1610029548406601,
      "num_tokens": 45497270.0,
      "step": 22455
    },
    {
      "entropy": 6.043559408187866,
      "epoch": 1.9627720003495588,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004614106151700105,
      "loss": 5.7707,
      "mean_token_accuracy": 0.1638267770409584,
      "num_tokens": 45507415.0,
      "step": 22460
    },
    {
      "entropy": 6.08746829032898,
      "epoch": 1.9632089487022633,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004613931631220903,
      "loss": 5.777,
      "mean_token_accuracy": 0.15974549800157548,
      "num_tokens": 45518025.0,
      "step": 22465
    },
    {
      "entropy": 6.088878536224366,
      "epoch": 1.9636458970549682,
      "grad_norm": 2.125,
      "learning_rate": 0.00046137570749905157,
      "loss": 5.8089,
      "mean_token_accuracy": 0.16160180121660234,
      "num_tokens": 45528834.0,
      "step": 22470
    },
    {
      "entropy": 6.095954418182373,
      "epoch": 1.9640828454076729,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004613582483012291,
      "loss": 5.6949,
      "mean_token_accuracy": 0.1664995864033699,
      "num_tokens": 45538543.0,
      "step": 22475
    },
    {
      "entropy": 6.080070495605469,
      "epoch": 1.9645197937603776,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004613407855289578,
      "loss": 5.8234,
      "mean_token_accuracy": 0.15997837632894515,
      "num_tokens": 45549298.0,
      "step": 22480
    },
    {
      "entropy": 6.132661151885986,
      "epoch": 1.9649567421130822,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004613233191825726,
      "loss": 5.824,
      "mean_token_accuracy": 0.15820150077342987,
      "num_tokens": 45560064.0,
      "step": 22485
    },
    {
      "entropy": 6.107428407669067,
      "epoch": 1.965393690465787,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004613058492624084,
      "loss": 5.661,
      "mean_token_accuracy": 0.17348057627677918,
      "num_tokens": 45569780.0,
      "step": 22490
    },
    {
      "entropy": 6.015264892578125,
      "epoch": 1.9658306388184916,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004612883757688005,
      "loss": 5.662,
      "mean_token_accuracy": 0.16760581582784653,
      "num_tokens": 45580256.0,
      "step": 22495
    },
    {
      "entropy": 6.070045757293701,
      "epoch": 1.9662675871711963,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046127089870208395,
      "loss": 5.773,
      "mean_token_accuracy": 0.1602664217352867,
      "num_tokens": 45590615.0,
      "step": 22500
    },
    {
      "entropy": 6.013664817810058,
      "epoch": 1.9667045355239012,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004612534180625939,
      "loss": 5.6844,
      "mean_token_accuracy": 0.16667721122503282,
      "num_tokens": 45600215.0,
      "step": 22505
    },
    {
      "entropy": 6.047497797012329,
      "epoch": 1.9671414838766057,
      "grad_norm": 1.71875,
      "learning_rate": 0.00046123593385066574,
      "loss": 5.763,
      "mean_token_accuracy": 0.1648322120308876,
      "num_tokens": 45610484.0,
      "step": 22510
    },
    {
      "entropy": 6.0978326320648195,
      "epoch": 1.9675784322293106,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046121844606663475,
      "loss": 5.7714,
      "mean_token_accuracy": 0.16370593756437302,
      "num_tokens": 45620528.0,
      "step": 22515
    },
    {
      "entropy": 6.110642433166504,
      "epoch": 1.968015380582015,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004612009547108364,
      "loss": 5.7288,
      "mean_token_accuracy": 0.16233521252870559,
      "num_tokens": 45629985.0,
      "step": 22520
    },
    {
      "entropy": 6.108655595779419,
      "epoch": 1.96845232893472,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046118345978360615,
      "loss": 5.7839,
      "mean_token_accuracy": 0.15828921049833297,
      "num_tokens": 45639750.0,
      "step": 22525
    },
    {
      "entropy": 6.0946649551391605,
      "epoch": 1.9688892772874245,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004611659612852796,
      "loss": 5.7466,
      "mean_token_accuracy": 0.16023516207933425,
      "num_tokens": 45650779.0,
      "step": 22530
    },
    {
      "entropy": 6.082919454574585,
      "epoch": 1.9693262256401294,
      "grad_norm": 2.09375,
      "learning_rate": 0.00046114845921619234,
      "loss": 5.7282,
      "mean_token_accuracy": 0.15885045230388642,
      "num_tokens": 45660363.0,
      "step": 22535
    },
    {
      "entropy": 6.00978102684021,
      "epoch": 1.969763173992834,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004611309535766801,
      "loss": 5.8069,
      "mean_token_accuracy": 0.15965428948402405,
      "num_tokens": 45670474.0,
      "step": 22540
    },
    {
      "entropy": 6.0597669124603275,
      "epoch": 1.9702001223455388,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004611134443670787,
      "loss": 5.6507,
      "mean_token_accuracy": 0.17145250290632247,
      "num_tokens": 45680157.0,
      "step": 22545
    },
    {
      "entropy": 6.08842625617981,
      "epoch": 1.9706370706982435,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004610959315877239,
      "loss": 5.751,
      "mean_token_accuracy": 0.1684378519654274,
      "num_tokens": 45689340.0,
      "step": 22550
    },
    {
      "entropy": 6.128485441207886,
      "epoch": 1.9710740190509481,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00046107841523895153,
      "loss": 5.7752,
      "mean_token_accuracy": 0.15523731857538223,
      "num_tokens": 45699453.0,
      "step": 22555
    },
    {
      "entropy": 6.096054935455323,
      "epoch": 1.9715109674036528,
      "grad_norm": 2.609375,
      "learning_rate": 0.00046106089532109776,
      "loss": 5.7349,
      "mean_token_accuracy": 0.16418419331312178,
      "num_tokens": 45711513.0,
      "step": 22560
    },
    {
      "entropy": 6.109231472015381,
      "epoch": 1.9719479157563575,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004610433718344985,
      "loss": 5.8233,
      "mean_token_accuracy": 0.15598211586475372,
      "num_tokens": 45722939.0,
      "step": 22565
    },
    {
      "entropy": 6.08656940460205,
      "epoch": 1.9723848641090624,
      "grad_norm": 7.0,
      "learning_rate": 0.00046102584477948985,
      "loss": 5.7673,
      "mean_token_accuracy": 0.16265629678964616,
      "num_tokens": 45733099.0,
      "step": 22570
    },
    {
      "entropy": 6.0580028057098385,
      "epoch": 1.972821812461767,
      "grad_norm": 1.75,
      "learning_rate": 0.0004610083141564079,
      "loss": 5.7982,
      "mean_token_accuracy": 0.16258619874715804,
      "num_tokens": 45743161.0,
      "step": 22575
    },
    {
      "entropy": 6.101008176803589,
      "epoch": 1.9732587608144718,
      "grad_norm": 3.21875,
      "learning_rate": 0.00046099077996558923,
      "loss": 5.8406,
      "mean_token_accuracy": 0.15620536208152772,
      "num_tokens": 45752903.0,
      "step": 22580
    },
    {
      "entropy": 6.050364065170288,
      "epoch": 1.9736957091671763,
      "grad_norm": 1.75,
      "learning_rate": 0.00046097324220736993,
      "loss": 5.6534,
      "mean_token_accuracy": 0.17089722901582718,
      "num_tokens": 45763360.0,
      "step": 22585
    },
    {
      "entropy": 6.032467699050903,
      "epoch": 1.9741326575198812,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004609557008820863,
      "loss": 5.6654,
      "mean_token_accuracy": 0.17225194722414017,
      "num_tokens": 45773025.0,
      "step": 22590
    },
    {
      "entropy": 6.051274156570434,
      "epoch": 1.974569605872586,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004609381559900749,
      "loss": 5.6483,
      "mean_token_accuracy": 0.17039043009281157,
      "num_tokens": 45782844.0,
      "step": 22595
    },
    {
      "entropy": 6.080592012405395,
      "epoch": 1.9750065542252906,
      "grad_norm": 1.765625,
      "learning_rate": 0.00046092060753167224,
      "loss": 5.7429,
      "mean_token_accuracy": 0.1633959725499153,
      "num_tokens": 45793058.0,
      "step": 22600
    },
    {
      "entropy": 6.146870756149292,
      "epoch": 1.9754435025779953,
      "grad_norm": 2.0,
      "learning_rate": 0.0004609030555072149,
      "loss": 5.7488,
      "mean_token_accuracy": 0.16555430144071578,
      "num_tokens": 45802865.0,
      "step": 22605
    },
    {
      "entropy": 6.021728563308716,
      "epoch": 1.9758804509307,
      "grad_norm": 1.984375,
      "learning_rate": 0.00046088549991703953,
      "loss": 5.6796,
      "mean_token_accuracy": 0.17114602625370026,
      "num_tokens": 45812981.0,
      "step": 22610
    },
    {
      "entropy": 5.992774152755738,
      "epoch": 1.9763173992834047,
      "grad_norm": 1.8359375,
      "learning_rate": 0.000460867940761483,
      "loss": 5.6414,
      "mean_token_accuracy": 0.16828957051038743,
      "num_tokens": 45822477.0,
      "step": 22615
    },
    {
      "entropy": 6.0608161926269535,
      "epoch": 1.9767543476361094,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004608503780408818,
      "loss": 5.8274,
      "mean_token_accuracy": 0.16007512658834458,
      "num_tokens": 45833642.0,
      "step": 22620
    },
    {
      "entropy": 6.150693464279175,
      "epoch": 1.9771912959888143,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004608328117555731,
      "loss": 5.7456,
      "mean_token_accuracy": 0.16341430246829985,
      "num_tokens": 45843079.0,
      "step": 22625
    },
    {
      "entropy": 6.133130121231079,
      "epoch": 1.9776282443415187,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004608152419058936,
      "loss": 5.7701,
      "mean_token_accuracy": 0.15971323549747468,
      "num_tokens": 45852987.0,
      "step": 22630
    },
    {
      "entropy": 6.064407205581665,
      "epoch": 1.9780651926942237,
      "grad_norm": 1.828125,
      "learning_rate": 0.00046079766849218043,
      "loss": 5.7593,
      "mean_token_accuracy": 0.15862571820616722,
      "num_tokens": 45863148.0,
      "step": 22635
    },
    {
      "entropy": 6.027590322494507,
      "epoch": 1.9785021410469281,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046078009151477063,
      "loss": 5.7462,
      "mean_token_accuracy": 0.16334546506404876,
      "num_tokens": 45873430.0,
      "step": 22640
    },
    {
      "entropy": 6.179696798324585,
      "epoch": 1.978939089399633,
      "grad_norm": 1.78125,
      "learning_rate": 0.00046076251097400126,
      "loss": 5.7364,
      "mean_token_accuracy": 0.16242606043815613,
      "num_tokens": 45883169.0,
      "step": 22645
    },
    {
      "entropy": 6.124985265731811,
      "epoch": 1.9793760377523377,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046074492687020967,
      "loss": 5.7931,
      "mean_token_accuracy": 0.15960925668478013,
      "num_tokens": 45893948.0,
      "step": 22650
    },
    {
      "entropy": 6.035357713699341,
      "epoch": 1.9798129861050424,
      "grad_norm": 5.0,
      "learning_rate": 0.000460727339203733,
      "loss": 5.7715,
      "mean_token_accuracy": 0.16885453313589097,
      "num_tokens": 45904887.0,
      "step": 22655
    },
    {
      "entropy": 6.154988956451416,
      "epoch": 1.980249934457747,
      "grad_norm": 1.6875,
      "learning_rate": 0.00046070974797490875,
      "loss": 5.7522,
      "mean_token_accuracy": 0.15924702286720277,
      "num_tokens": 45914845.0,
      "step": 22660
    },
    {
      "entropy": 6.077184677124023,
      "epoch": 1.9806868828104518,
      "grad_norm": 1.375,
      "learning_rate": 0.0004606921531840741,
      "loss": 5.7589,
      "mean_token_accuracy": 0.16273245662450792,
      "num_tokens": 45925104.0,
      "step": 22665
    },
    {
      "entropy": 6.027268171310425,
      "epoch": 1.9811238311631565,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004606745548315667,
      "loss": 5.6869,
      "mean_token_accuracy": 0.16162634491920472,
      "num_tokens": 45934934.0,
      "step": 22670
    },
    {
      "entropy": 6.052566766738892,
      "epoch": 1.9815607795158612,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004606569529177241,
      "loss": 5.7381,
      "mean_token_accuracy": 0.1706233039498329,
      "num_tokens": 45945449.0,
      "step": 22675
    },
    {
      "entropy": 6.133802318572998,
      "epoch": 1.981997727868566,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004606393474428838,
      "loss": 5.7663,
      "mean_token_accuracy": 0.15987107455730437,
      "num_tokens": 45954897.0,
      "step": 22680
    },
    {
      "entropy": 6.129783630371094,
      "epoch": 1.9824346762212706,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046062173840738356,
      "loss": 5.7109,
      "mean_token_accuracy": 0.16761244982481002,
      "num_tokens": 45965085.0,
      "step": 22685
    },
    {
      "entropy": 6.036872482299804,
      "epoch": 1.9828716245739755,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004606041258115611,
      "loss": 5.755,
      "mean_token_accuracy": 0.1720885679125786,
      "num_tokens": 45974610.0,
      "step": 22690
    },
    {
      "entropy": 6.090267038345337,
      "epoch": 1.98330857292668,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004605865096557542,
      "loss": 5.7394,
      "mean_token_accuracy": 0.15660437941551208,
      "num_tokens": 45984922.0,
      "step": 22695
    },
    {
      "entropy": 6.058012390136719,
      "epoch": 1.9837455212793849,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004605688899403009,
      "loss": 5.7222,
      "mean_token_accuracy": 0.16665259152650833,
      "num_tokens": 45994591.0,
      "step": 22700
    },
    {
      "entropy": 6.101830530166626,
      "epoch": 1.9841824696320893,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046055126666553906,
      "loss": 5.8529,
      "mean_token_accuracy": 0.1552139014005661,
      "num_tokens": 46005291.0,
      "step": 22705
    },
    {
      "entropy": 6.057538318634033,
      "epoch": 1.9846194179847942,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046053363983180657,
      "loss": 5.7111,
      "mean_token_accuracy": 0.16418332904577254,
      "num_tokens": 46014970.0,
      "step": 22710
    },
    {
      "entropy": 6.101908826828003,
      "epoch": 1.985056366337499,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046051600943944175,
      "loss": 5.8046,
      "mean_token_accuracy": 0.16192987263202668,
      "num_tokens": 46025042.0,
      "step": 22715
    },
    {
      "entropy": 6.136908483505249,
      "epoch": 1.9854933146902036,
      "grad_norm": 1.65625,
      "learning_rate": 0.00046049837548878264,
      "loss": 5.8356,
      "mean_token_accuracy": 0.15601289868354798,
      "num_tokens": 46035440.0,
      "step": 22720
    },
    {
      "entropy": 6.054912137985229,
      "epoch": 1.9859302630429083,
      "grad_norm": 1.828125,
      "learning_rate": 0.00046048073798016746,
      "loss": 5.7234,
      "mean_token_accuracy": 0.16993111222982407,
      "num_tokens": 46045344.0,
      "step": 22725
    },
    {
      "entropy": 6.106076860427857,
      "epoch": 1.986367211395613,
      "grad_norm": 2.671875,
      "learning_rate": 0.00046046309691393455,
      "loss": 5.7977,
      "mean_token_accuracy": 0.15871866792440414,
      "num_tokens": 46055054.0,
      "step": 22730
    },
    {
      "entropy": 6.1511735916137695,
      "epoch": 1.9868041597483177,
      "grad_norm": 1.734375,
      "learning_rate": 0.00046044545229042226,
      "loss": 5.7672,
      "mean_token_accuracy": 0.1637808784842491,
      "num_tokens": 46063835.0,
      "step": 22735
    },
    {
      "entropy": 6.089233827590943,
      "epoch": 1.9872411081010224,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046042780410996906,
      "loss": 5.7862,
      "mean_token_accuracy": 0.15857492685317992,
      "num_tokens": 46074256.0,
      "step": 22740
    },
    {
      "entropy": 6.159700202941894,
      "epoch": 1.9876780564537273,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00046041015237291346,
      "loss": 5.8596,
      "mean_token_accuracy": 0.16098454594612122,
      "num_tokens": 46085563.0,
      "step": 22745
    },
    {
      "entropy": 6.108792495727539,
      "epoch": 1.9881150048064318,
      "grad_norm": 1.84375,
      "learning_rate": 0.00046039249707959386,
      "loss": 5.7488,
      "mean_token_accuracy": 0.1651621237397194,
      "num_tokens": 46094956.0,
      "step": 22750
    },
    {
      "entropy": 6.0901154518127445,
      "epoch": 1.9885519531591367,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046037483823034906,
      "loss": 5.7746,
      "mean_token_accuracy": 0.1657084971666336,
      "num_tokens": 46105292.0,
      "step": 22755
    },
    {
      "entropy": 6.127906608581543,
      "epoch": 1.9889889015118412,
      "grad_norm": 2.5,
      "learning_rate": 0.0004603571758255177,
      "loss": 5.9035,
      "mean_token_accuracy": 0.14976495653390884,
      "num_tokens": 46117254.0,
      "step": 22760
    },
    {
      "entropy": 6.124961233139038,
      "epoch": 1.989425849864546,
      "grad_norm": 1.734375,
      "learning_rate": 0.00046033950986543865,
      "loss": 5.8186,
      "mean_token_accuracy": 0.15687327533960344,
      "num_tokens": 46127846.0,
      "step": 22765
    },
    {
      "entropy": 6.124219799041748,
      "epoch": 1.9898627982172508,
      "grad_norm": 2.46875,
      "learning_rate": 0.00046032184035045066,
      "loss": 5.8089,
      "mean_token_accuracy": 0.1607054889202118,
      "num_tokens": 46138476.0,
      "step": 22770
    },
    {
      "entropy": 6.068752098083496,
      "epoch": 1.9902997465699555,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046030416728089267,
      "loss": 5.6931,
      "mean_token_accuracy": 0.1698729455471039,
      "num_tokens": 46147789.0,
      "step": 22775
    },
    {
      "entropy": 6.074982070922852,
      "epoch": 1.9907366949226601,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046028649065710367,
      "loss": 5.8308,
      "mean_token_accuracy": 0.1555711016058922,
      "num_tokens": 46158484.0,
      "step": 22780
    },
    {
      "entropy": 6.163306713104248,
      "epoch": 1.9911736432753648,
      "grad_norm": 2.375,
      "learning_rate": 0.00046026881047942273,
      "loss": 5.7813,
      "mean_token_accuracy": 0.16297018826007842,
      "num_tokens": 46168062.0,
      "step": 22785
    },
    {
      "entropy": 6.129098892211914,
      "epoch": 1.9916105916280695,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00046025112674818883,
      "loss": 5.7527,
      "mean_token_accuracy": 0.16226703375577928,
      "num_tokens": 46178710.0,
      "step": 22790
    },
    {
      "entropy": 6.049213647842407,
      "epoch": 1.9920475399807742,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00046023343946374127,
      "loss": 5.7877,
      "mean_token_accuracy": 0.15767476558685303,
      "num_tokens": 46188728.0,
      "step": 22795
    },
    {
      "entropy": 6.08474178314209,
      "epoch": 1.9924844883334791,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046021574862641925,
      "loss": 5.7621,
      "mean_token_accuracy": 0.16782106459140778,
      "num_tokens": 46199716.0,
      "step": 22800
    },
    {
      "entropy": 6.172091722488403,
      "epoch": 1.9929214366861836,
      "grad_norm": 3.625,
      "learning_rate": 0.00046019805423656217,
      "loss": 5.828,
      "mean_token_accuracy": 0.15575862526893616,
      "num_tokens": 46210885.0,
      "step": 22805
    },
    {
      "entropy": 6.101210308074951,
      "epoch": 1.9933583850388885,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004601803562945094,
      "loss": 5.7657,
      "mean_token_accuracy": 0.17010775953531265,
      "num_tokens": 46220879.0,
      "step": 22810
    },
    {
      "entropy": 6.022843599319458,
      "epoch": 1.993795333391593,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004601626548006003,
      "loss": 5.772,
      "mean_token_accuracy": 0.15980671793222428,
      "num_tokens": 46230420.0,
      "step": 22815
    },
    {
      "entropy": 6.0228776931762695,
      "epoch": 1.994232281744298,
      "grad_norm": 2.75,
      "learning_rate": 0.0004601449497551745,
      "loss": 5.7366,
      "mean_token_accuracy": 0.16613879501819612,
      "num_tokens": 46240548.0,
      "step": 22820
    },
    {
      "entropy": 6.093614006042481,
      "epoch": 1.9946692300970026,
      "grad_norm": 1.859375,
      "learning_rate": 0.00046012724115857153,
      "loss": 5.7503,
      "mean_token_accuracy": 0.16345759183168412,
      "num_tokens": 46250484.0,
      "step": 22825
    },
    {
      "entropy": 6.166223239898682,
      "epoch": 1.9951061784497073,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046010952901113107,
      "loss": 5.7805,
      "mean_token_accuracy": 0.16283805817365646,
      "num_tokens": 46261184.0,
      "step": 22830
    },
    {
      "entropy": 6.033894300460815,
      "epoch": 1.995543126802412,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004600918133131928,
      "loss": 5.7523,
      "mean_token_accuracy": 0.16789272278547288,
      "num_tokens": 46271301.0,
      "step": 22835
    },
    {
      "entropy": 6.0566002368927006,
      "epoch": 1.9959800751551167,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046007409406509663,
      "loss": 5.7786,
      "mean_token_accuracy": 0.1589366152882576,
      "num_tokens": 46281201.0,
      "step": 22840
    },
    {
      "entropy": 6.167833709716797,
      "epoch": 1.9964170235078214,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004600563712671823,
      "loss": 5.8103,
      "mean_token_accuracy": 0.16274736523628236,
      "num_tokens": 46291969.0,
      "step": 22845
    },
    {
      "entropy": 5.974885654449463,
      "epoch": 1.996853971860526,
      "grad_norm": 1.546875,
      "learning_rate": 0.00046003864491978984,
      "loss": 5.576,
      "mean_token_accuracy": 0.18469550013542174,
      "num_tokens": 46301308.0,
      "step": 22850
    },
    {
      "entropy": 6.055268573760986,
      "epoch": 1.997290920213231,
      "grad_norm": 2.734375,
      "learning_rate": 0.00046002091502325926,
      "loss": 5.7594,
      "mean_token_accuracy": 0.1599859654903412,
      "num_tokens": 46311564.0,
      "step": 22855
    },
    {
      "entropy": 6.0955170631408695,
      "epoch": 1.9977278685659354,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004600031815779306,
      "loss": 5.7331,
      "mean_token_accuracy": 0.1684732586145401,
      "num_tokens": 46321843.0,
      "step": 22860
    },
    {
      "entropy": 6.014130544662476,
      "epoch": 1.9981648169186403,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00045998544458414385,
      "loss": 5.6801,
      "mean_token_accuracy": 0.16574269235134126,
      "num_tokens": 46331274.0,
      "step": 22865
    },
    {
      "entropy": 6.106042432785034,
      "epoch": 1.9986017652713448,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045996770404223937,
      "loss": 5.8089,
      "mean_token_accuracy": 0.15510234460234643,
      "num_tokens": 46341385.0,
      "step": 22870
    },
    {
      "entropy": 6.086014080047607,
      "epoch": 1.9990387136240497,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045994995995255746,
      "loss": 5.731,
      "mean_token_accuracy": 0.16630140393972398,
      "num_tokens": 46350385.0,
      "step": 22875
    },
    {
      "entropy": 5.97367525100708,
      "epoch": 1.9994756619767542,
      "grad_norm": 2.28125,
      "learning_rate": 0.00045993221231543837,
      "loss": 5.6535,
      "mean_token_accuracy": 0.1656590521335602,
      "num_tokens": 46361111.0,
      "step": 22880
    },
    {
      "entropy": 6.071186161041259,
      "epoch": 1.999912610329459,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004599144611312226,
      "loss": 5.7964,
      "mean_token_accuracy": 0.15679125338792801,
      "num_tokens": 46371316.0,
      "step": 22885
    },
    {
      "entropy": 6.139257192611694,
      "epoch": 2.0003495586821636,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00045989670640025047,
      "loss": 5.7494,
      "mean_token_accuracy": 0.16184891909360885,
      "num_tokens": 46381151.0,
      "step": 22890
    },
    {
      "entropy": 6.166263628005981,
      "epoch": 2.0007865070348685,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00045987894812286276,
      "loss": 5.7051,
      "mean_token_accuracy": 0.16529084295034407,
      "num_tokens": 46391420.0,
      "step": 22895
    },
    {
      "entropy": 6.02949857711792,
      "epoch": 2.0012234553875734,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045986118629939984,
      "loss": 5.6896,
      "mean_token_accuracy": 0.16630890518426894,
      "num_tokens": 46401575.0,
      "step": 22900
    },
    {
      "entropy": 6.022871971130371,
      "epoch": 2.001660403740278,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004598434209302025,
      "loss": 5.6581,
      "mean_token_accuracy": 0.17053814977407455,
      "num_tokens": 46411275.0,
      "step": 22905
    },
    {
      "entropy": 6.100580167770386,
      "epoch": 2.002097352092983,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004598256520156116,
      "loss": 5.7062,
      "mean_token_accuracy": 0.17140741348266603,
      "num_tokens": 46421179.0,
      "step": 22910
    },
    {
      "entropy": 6.096224737167359,
      "epoch": 2.0025343004456873,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004598078795559678,
      "loss": 5.7545,
      "mean_token_accuracy": 0.1623886078596115,
      "num_tokens": 46431865.0,
      "step": 22915
    },
    {
      "entropy": 6.084360790252686,
      "epoch": 2.002971248798392,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045979010355161196,
      "loss": 5.8373,
      "mean_token_accuracy": 0.1521059677004814,
      "num_tokens": 46443398.0,
      "step": 22920
    },
    {
      "entropy": 6.185590171813965,
      "epoch": 2.0034081971510966,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004597723240028852,
      "loss": 5.8235,
      "mean_token_accuracy": 0.15559173375368118,
      "num_tokens": 46454818.0,
      "step": 22925
    },
    {
      "entropy": 6.185563611984253,
      "epoch": 2.0038451455038016,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004597545409101283,
      "loss": 5.7664,
      "mean_token_accuracy": 0.16048292070627213,
      "num_tokens": 46465231.0,
      "step": 22930
    },
    {
      "entropy": 6.140785217285156,
      "epoch": 2.004282093856506,
      "grad_norm": 2.75,
      "learning_rate": 0.0004597367542736826,
      "loss": 5.7349,
      "mean_token_accuracy": 0.16676682531833648,
      "num_tokens": 46476323.0,
      "step": 22935
    },
    {
      "entropy": 5.994490337371826,
      "epoch": 2.004719042209211,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004597189640938892,
      "loss": 5.7037,
      "mean_token_accuracy": 0.16294893473386765,
      "num_tokens": 46486573.0,
      "step": 22940
    },
    {
      "entropy": 5.981491661071777,
      "epoch": 2.0051559905619154,
      "grad_norm": 1.5,
      "learning_rate": 0.0004597011703710892,
      "loss": 5.6807,
      "mean_token_accuracy": 0.16890868693590164,
      "num_tokens": 46497779.0,
      "step": 22945
    },
    {
      "entropy": 6.134116888046265,
      "epoch": 2.0055929389146203,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000459683373105624,
      "loss": 5.7574,
      "mean_token_accuracy": 0.15806543976068496,
      "num_tokens": 46509074.0,
      "step": 22950
    },
    {
      "entropy": 6.028037261962891,
      "epoch": 2.006029887267325,
      "grad_norm": 1.75,
      "learning_rate": 0.00045966557229783493,
      "loss": 5.5772,
      "mean_token_accuracy": 0.17580075114965438,
      "num_tokens": 46519292.0,
      "step": 22955
    },
    {
      "entropy": 6.004097414016724,
      "epoch": 2.0064668356200297,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004596477679480634,
      "loss": 5.6663,
      "mean_token_accuracy": 0.16862158328294755,
      "num_tokens": 46529689.0,
      "step": 22960
    },
    {
      "entropy": 6.093133163452149,
      "epoch": 2.0069037839727346,
      "grad_norm": 1.46875,
      "learning_rate": 0.00045962996005665096,
      "loss": 5.7382,
      "mean_token_accuracy": 0.15880288779735566,
      "num_tokens": 46538616.0,
      "step": 22965
    },
    {
      "entropy": 6.008227252960205,
      "epoch": 2.007340732325439,
      "grad_norm": 2.28125,
      "learning_rate": 0.00045961214862393914,
      "loss": 5.6155,
      "mean_token_accuracy": 0.1764351323246956,
      "num_tokens": 46548372.0,
      "step": 22970
    },
    {
      "entropy": 6.008027601242065,
      "epoch": 2.007777680678144,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004595943336502695,
      "loss": 5.7047,
      "mean_token_accuracy": 0.16187805980443953,
      "num_tokens": 46558991.0,
      "step": 22975
    },
    {
      "entropy": 6.071149826049805,
      "epoch": 2.0082146290308485,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004595765151359839,
      "loss": 5.7111,
      "mean_token_accuracy": 0.1643735498189926,
      "num_tokens": 46569531.0,
      "step": 22980
    },
    {
      "entropy": 6.174090480804443,
      "epoch": 2.0086515773835534,
      "grad_norm": 1.8828125,
      "learning_rate": 0.000459558693081424,
      "loss": 5.7367,
      "mean_token_accuracy": 0.1662942185997963,
      "num_tokens": 46579774.0,
      "step": 22985
    },
    {
      "entropy": 6.083174753189087,
      "epoch": 2.009088525736258,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004595408674869317,
      "loss": 5.7152,
      "mean_token_accuracy": 0.16373969614505768,
      "num_tokens": 46589810.0,
      "step": 22990
    },
    {
      "entropy": 6.035809803009033,
      "epoch": 2.0095254740889628,
      "grad_norm": 1.875,
      "learning_rate": 0.0004595230383528488,
      "loss": 5.6863,
      "mean_token_accuracy": 0.16751551181077956,
      "num_tokens": 46599347.0,
      "step": 22995
    },
    {
      "entropy": 6.203772163391113,
      "epoch": 2.0099624224416672,
      "grad_norm": 1.515625,
      "learning_rate": 0.00045950520567951734,
      "loss": 5.7993,
      "mean_token_accuracy": 0.15256014913320542,
      "num_tokens": 46609926.0,
      "step": 23000
    },
    {
      "entropy": 6.050665950775146,
      "epoch": 2.010399370794372,
      "grad_norm": 2.875,
      "learning_rate": 0.0004594873694672794,
      "loss": 5.7239,
      "mean_token_accuracy": 0.16721907258033752,
      "num_tokens": 46620736.0,
      "step": 23005
    },
    {
      "entropy": 6.019650316238403,
      "epoch": 2.0108363191470766,
      "grad_norm": 3.03125,
      "learning_rate": 0.00045946952971647696,
      "loss": 5.6689,
      "mean_token_accuracy": 0.1644389048218727,
      "num_tokens": 46630713.0,
      "step": 23010
    },
    {
      "entropy": 6.048619413375855,
      "epoch": 2.0112732674997815,
      "grad_norm": 1.625,
      "learning_rate": 0.0004594516864274524,
      "loss": 5.7105,
      "mean_token_accuracy": 0.16697389483451844,
      "num_tokens": 46641197.0,
      "step": 23015
    },
    {
      "entropy": 5.983325099945068,
      "epoch": 2.0117102158524864,
      "grad_norm": 1.5,
      "learning_rate": 0.00045943383960054775,
      "loss": 5.5401,
      "mean_token_accuracy": 0.1900812044739723,
      "num_tokens": 46651325.0,
      "step": 23020
    },
    {
      "entropy": 6.034312343597412,
      "epoch": 2.012147164205191,
      "grad_norm": 2.703125,
      "learning_rate": 0.00045941598923610543,
      "loss": 5.7248,
      "mean_token_accuracy": 0.16668762415647506,
      "num_tokens": 46662100.0,
      "step": 23025
    },
    {
      "entropy": 6.055118465423584,
      "epoch": 2.012584112557896,
      "grad_norm": 2.453125,
      "learning_rate": 0.00045939813533446783,
      "loss": 5.6997,
      "mean_token_accuracy": 0.1677699625492096,
      "num_tokens": 46671460.0,
      "step": 23030
    },
    {
      "entropy": 6.10442419052124,
      "epoch": 2.0130210609106003,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004593802778959773,
      "loss": 5.7771,
      "mean_token_accuracy": 0.16210529804229737,
      "num_tokens": 46681835.0,
      "step": 23035
    },
    {
      "entropy": 6.009375858306885,
      "epoch": 2.013458009263305,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004593624169209765,
      "loss": 5.7021,
      "mean_token_accuracy": 0.1659682109951973,
      "num_tokens": 46691976.0,
      "step": 23040
    },
    {
      "entropy": 6.026082229614258,
      "epoch": 2.0138949576160097,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045934455240980793,
      "loss": 5.6975,
      "mean_token_accuracy": 0.16736950874328613,
      "num_tokens": 46702589.0,
      "step": 23045
    },
    {
      "entropy": 6.125696897506714,
      "epoch": 2.0143319059687146,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004593266843628143,
      "loss": 5.7965,
      "mean_token_accuracy": 0.15735709220170974,
      "num_tokens": 46712167.0,
      "step": 23050
    },
    {
      "entropy": 6.095148038864136,
      "epoch": 2.014768854321419,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004593088127803382,
      "loss": 5.729,
      "mean_token_accuracy": 0.1662774220108986,
      "num_tokens": 46722684.0,
      "step": 23055
    },
    {
      "entropy": 6.062026834487915,
      "epoch": 2.015205802674124,
      "grad_norm": 1.984375,
      "learning_rate": 0.00045929093766272253,
      "loss": 5.7493,
      "mean_token_accuracy": 0.16227701902389527,
      "num_tokens": 46733856.0,
      "step": 23060
    },
    {
      "entropy": 6.131888437271118,
      "epoch": 2.0156427510268284,
      "grad_norm": 2.5,
      "learning_rate": 0.0004592730590103101,
      "loss": 5.7793,
      "mean_token_accuracy": 0.15947109386324881,
      "num_tokens": 46744210.0,
      "step": 23065
    },
    {
      "entropy": 6.125852060317993,
      "epoch": 2.0160796993795334,
      "grad_norm": 1.625,
      "learning_rate": 0.0004592551768234438,
      "loss": 5.6938,
      "mean_token_accuracy": 0.1635714963078499,
      "num_tokens": 46754611.0,
      "step": 23070
    },
    {
      "entropy": 6.097046375274658,
      "epoch": 2.0165166477322383,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004592372911024667,
      "loss": 5.7837,
      "mean_token_accuracy": 0.1567164808511734,
      "num_tokens": 46765647.0,
      "step": 23075
    },
    {
      "entropy": 6.0518913745880125,
      "epoch": 2.0169535960849427,
      "grad_norm": 2.28125,
      "learning_rate": 0.00045921940184772184,
      "loss": 5.7554,
      "mean_token_accuracy": 0.16240618526935577,
      "num_tokens": 46776052.0,
      "step": 23080
    },
    {
      "entropy": 6.099923467636108,
      "epoch": 2.0173905444376476,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004592015090595523,
      "loss": 5.6957,
      "mean_token_accuracy": 0.16792017519474028,
      "num_tokens": 46786255.0,
      "step": 23085
    },
    {
      "entropy": 6.007470226287841,
      "epoch": 2.017827492790352,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004591836127383013,
      "loss": 5.7015,
      "mean_token_accuracy": 0.16948963403701783,
      "num_tokens": 46795958.0,
      "step": 23090
    },
    {
      "entropy": 6.074303436279297,
      "epoch": 2.018264441143057,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045916571288431206,
      "loss": 5.7167,
      "mean_token_accuracy": 0.15861888974905014,
      "num_tokens": 46806508.0,
      "step": 23095
    },
    {
      "entropy": 6.062554836273193,
      "epoch": 2.0187013894957615,
      "grad_norm": 1.625,
      "learning_rate": 0.00045914780949792797,
      "loss": 5.7834,
      "mean_token_accuracy": 0.15887600183486938,
      "num_tokens": 46816868.0,
      "step": 23100
    },
    {
      "entropy": 6.064577198028564,
      "epoch": 2.0191383378484664,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004591299025794924,
      "loss": 5.7351,
      "mean_token_accuracy": 0.16613228917121886,
      "num_tokens": 46827591.0,
      "step": 23105
    },
    {
      "entropy": 6.077753210067749,
      "epoch": 2.019575286201171,
      "grad_norm": 2.0,
      "learning_rate": 0.0004591119921293488,
      "loss": 5.6828,
      "mean_token_accuracy": 0.16564187109470369,
      "num_tokens": 46837394.0,
      "step": 23110
    },
    {
      "entropy": 6.085086154937744,
      "epoch": 2.020012234553876,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00045909407814784074,
      "loss": 5.6827,
      "mean_token_accuracy": 0.1691093310713768,
      "num_tokens": 46847881.0,
      "step": 23115
    },
    {
      "entropy": 6.055296897888184,
      "epoch": 2.0204491829065803,
      "grad_norm": 1.546875,
      "learning_rate": 0.00045907616063531185,
      "loss": 5.7185,
      "mean_token_accuracy": 0.16428267508745192,
      "num_tokens": 46857728.0,
      "step": 23120
    },
    {
      "entropy": 6.112531900405884,
      "epoch": 2.020886131259285,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004590582395921057,
      "loss": 5.7573,
      "mean_token_accuracy": 0.16399911940097808,
      "num_tokens": 46866991.0,
      "step": 23125
    },
    {
      "entropy": 6.098387813568115,
      "epoch": 2.0213230796119896,
      "grad_norm": 2.0625,
      "learning_rate": 0.000459040315018566,
      "loss": 5.7894,
      "mean_token_accuracy": 0.16141496747732162,
      "num_tokens": 46876253.0,
      "step": 23130
    },
    {
      "entropy": 6.064151000976563,
      "epoch": 2.0217600279646946,
      "grad_norm": 2.125,
      "learning_rate": 0.00045902238691503675,
      "loss": 5.7104,
      "mean_token_accuracy": 0.1612723633646965,
      "num_tokens": 46887309.0,
      "step": 23135
    },
    {
      "entropy": 6.076644706726074,
      "epoch": 2.0221969763173995,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00045900445528186163,
      "loss": 5.7672,
      "mean_token_accuracy": 0.16368767470121384,
      "num_tokens": 46896722.0,
      "step": 23140
    },
    {
      "entropy": 6.045959186553955,
      "epoch": 2.022633924670104,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045898652011938456,
      "loss": 5.7622,
      "mean_token_accuracy": 0.16018479466438293,
      "num_tokens": 46907519.0,
      "step": 23145
    },
    {
      "entropy": 6.023602628707886,
      "epoch": 2.023070873022809,
      "grad_norm": 2.0,
      "learning_rate": 0.00045896858142794976,
      "loss": 5.6965,
      "mean_token_accuracy": 0.16883748918771743,
      "num_tokens": 46917387.0,
      "step": 23150
    },
    {
      "entropy": 6.092604398727417,
      "epoch": 2.0235078213755133,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004589506392079011,
      "loss": 5.6934,
      "mean_token_accuracy": 0.17138741165399551,
      "num_tokens": 46926477.0,
      "step": 23155
    },
    {
      "entropy": 6.110986328125,
      "epoch": 2.0239447697282182,
      "grad_norm": 2.046875,
      "learning_rate": 0.00045893269345958284,
      "loss": 5.7081,
      "mean_token_accuracy": 0.17339395135641097,
      "num_tokens": 46935807.0,
      "step": 23160
    },
    {
      "entropy": 6.013941240310669,
      "epoch": 2.0243817180809227,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004589147441833391,
      "loss": 5.7305,
      "mean_token_accuracy": 0.1628633677959442,
      "num_tokens": 46946843.0,
      "step": 23165
    },
    {
      "entropy": 6.036965465545654,
      "epoch": 2.0248186664336276,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004588967913795142,
      "loss": 5.7073,
      "mean_token_accuracy": 0.16671242415905,
      "num_tokens": 46956289.0,
      "step": 23170
    },
    {
      "entropy": 6.118314743041992,
      "epoch": 2.025255614786332,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00045887883504845244,
      "loss": 5.7168,
      "mean_token_accuracy": 0.16742930859327315,
      "num_tokens": 46965606.0,
      "step": 23175
    },
    {
      "entropy": 6.045379018783569,
      "epoch": 2.025692563139037,
      "grad_norm": 2.375,
      "learning_rate": 0.00045886087519049837,
      "loss": 5.6367,
      "mean_token_accuracy": 0.17374204993247985,
      "num_tokens": 46974769.0,
      "step": 23180
    },
    {
      "entropy": 6.06144700050354,
      "epoch": 2.0261295114917415,
      "grad_norm": 1.578125,
      "learning_rate": 0.00045884291180599626,
      "loss": 5.7457,
      "mean_token_accuracy": 0.15897135436534882,
      "num_tokens": 46984753.0,
      "step": 23185
    },
    {
      "entropy": 6.00959529876709,
      "epoch": 2.0265664598444464,
      "grad_norm": 1.859375,
      "learning_rate": 0.00045882494489529085,
      "loss": 5.6829,
      "mean_token_accuracy": 0.16159337013959885,
      "num_tokens": 46995403.0,
      "step": 23190
    },
    {
      "entropy": 6.097915124893189,
      "epoch": 2.0270034081971513,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004588069744587267,
      "loss": 5.7066,
      "mean_token_accuracy": 0.16750568151474,
      "num_tokens": 47005054.0,
      "step": 23195
    },
    {
      "entropy": 6.092878770828247,
      "epoch": 2.0274403565498558,
      "grad_norm": 5.28125,
      "learning_rate": 0.0004587890004966484,
      "loss": 5.7528,
      "mean_token_accuracy": 0.16034985184669495,
      "num_tokens": 47016315.0,
      "step": 23200
    },
    {
      "entropy": 6.150987195968628,
      "epoch": 2.0278773049025607,
      "grad_norm": 1.8125,
      "learning_rate": 0.00045877102300940077,
      "loss": 5.6532,
      "mean_token_accuracy": 0.15898548513650895,
      "num_tokens": 47025981.0,
      "step": 23205
    },
    {
      "entropy": 6.054524087905884,
      "epoch": 2.028314253255265,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00045875304199732863,
      "loss": 5.6642,
      "mean_token_accuracy": 0.1674490123987198,
      "num_tokens": 47035906.0,
      "step": 23210
    },
    {
      "entropy": 6.014098358154297,
      "epoch": 2.02875120160797,
      "grad_norm": 2.046875,
      "learning_rate": 0.00045873505746077685,
      "loss": 5.706,
      "mean_token_accuracy": 0.16008613780140876,
      "num_tokens": 47045884.0,
      "step": 23215
    },
    {
      "entropy": 6.0632835865020756,
      "epoch": 2.0291881499606745,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004587170694000904,
      "loss": 5.7276,
      "mean_token_accuracy": 0.1704191744327545,
      "num_tokens": 47055575.0,
      "step": 23220
    },
    {
      "entropy": 6.0386322975158695,
      "epoch": 2.0296250983133794,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004586990778156144,
      "loss": 5.7341,
      "mean_token_accuracy": 0.1617642343044281,
      "num_tokens": 47065484.0,
      "step": 23225
    },
    {
      "entropy": 6.137055587768555,
      "epoch": 2.030062046666084,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00045868108270769373,
      "loss": 5.6827,
      "mean_token_accuracy": 0.16793508678674698,
      "num_tokens": 47076604.0,
      "step": 23230
    },
    {
      "entropy": 6.064190006256103,
      "epoch": 2.030498995018789,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004586630840766736,
      "loss": 5.6928,
      "mean_token_accuracy": 0.16996303498744963,
      "num_tokens": 47086452.0,
      "step": 23235
    },
    {
      "entropy": 5.979254627227784,
      "epoch": 2.0309359433714933,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045864508192289933,
      "loss": 5.7584,
      "mean_token_accuracy": 0.1603222444653511,
      "num_tokens": 47096425.0,
      "step": 23240
    },
    {
      "entropy": 6.045675945281983,
      "epoch": 2.031372891724198,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00045862707624671616,
      "loss": 5.7366,
      "mean_token_accuracy": 0.1642124742269516,
      "num_tokens": 47107214.0,
      "step": 23245
    },
    {
      "entropy": 6.102190065383911,
      "epoch": 2.031809840076903,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045860906704846944,
      "loss": 5.6968,
      "mean_token_accuracy": 0.16576555073261262,
      "num_tokens": 47117498.0,
      "step": 23250
    },
    {
      "entropy": 5.999255990982055,
      "epoch": 2.0322467884296076,
      "grad_norm": 1.6875,
      "learning_rate": 0.00045859105432850464,
      "loss": 5.6481,
      "mean_token_accuracy": 0.1703611895442009,
      "num_tokens": 47128031.0,
      "step": 23255
    },
    {
      "entropy": 6.116655874252319,
      "epoch": 2.0326837367823125,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004585730380871672,
      "loss": 5.7309,
      "mean_token_accuracy": 0.16410210430622102,
      "num_tokens": 47138347.0,
      "step": 23260
    },
    {
      "entropy": 5.9679628849029545,
      "epoch": 2.033120685135017,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004585550183248027,
      "loss": 5.6329,
      "mean_token_accuracy": 0.1677520379424095,
      "num_tokens": 47147642.0,
      "step": 23265
    },
    {
      "entropy": 6.051870059967041,
      "epoch": 2.033557633487722,
      "grad_norm": 1.796875,
      "learning_rate": 0.00045853699504175673,
      "loss": 5.7368,
      "mean_token_accuracy": 0.16770342290401458,
      "num_tokens": 47156855.0,
      "step": 23270
    },
    {
      "entropy": 6.104798173904419,
      "epoch": 2.0339945818404264,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00045851896823837505,
      "loss": 5.6856,
      "mean_token_accuracy": 0.16494086980819703,
      "num_tokens": 47167030.0,
      "step": 23275
    },
    {
      "entropy": 6.087035703659057,
      "epoch": 2.0344315301931313,
      "grad_norm": 2.046875,
      "learning_rate": 0.00045850093791500334,
      "loss": 5.7294,
      "mean_token_accuracy": 0.16177235692739486,
      "num_tokens": 47177471.0,
      "step": 23280
    },
    {
      "entropy": 6.091072654724121,
      "epoch": 2.0348684785458357,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045848290407198754,
      "loss": 5.7843,
      "mean_token_accuracy": 0.15769875645637513,
      "num_tokens": 47187911.0,
      "step": 23285
    },
    {
      "entropy": 6.0601794719696045,
      "epoch": 2.0353054268985407,
      "grad_norm": 2.0,
      "learning_rate": 0.0004584648667096736,
      "loss": 5.6643,
      "mean_token_accuracy": 0.1677509218454361,
      "num_tokens": 47198622.0,
      "step": 23290
    },
    {
      "entropy": 6.055613470077515,
      "epoch": 2.035742375251245,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004584468258284072,
      "loss": 5.7586,
      "mean_token_accuracy": 0.16488331109285354,
      "num_tokens": 47207593.0,
      "step": 23295
    },
    {
      "entropy": 6.098476552963257,
      "epoch": 2.03617932360395,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004584287814285347,
      "loss": 5.7373,
      "mean_token_accuracy": 0.15975527316331864,
      "num_tokens": 47218734.0,
      "step": 23300
    },
    {
      "entropy": 6.096802663803101,
      "epoch": 2.0366162719566545,
      "grad_norm": 2.25,
      "learning_rate": 0.00045841073351040196,
      "loss": 5.762,
      "mean_token_accuracy": 0.1625097304582596,
      "num_tokens": 47228662.0,
      "step": 23305
    },
    {
      "entropy": 6.165839672088623,
      "epoch": 2.0370532203093594,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004583926820743553,
      "loss": 5.8463,
      "mean_token_accuracy": 0.15942087024450302,
      "num_tokens": 47238460.0,
      "step": 23310
    },
    {
      "entropy": 6.126078414916992,
      "epoch": 2.0374901686620643,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004583746271207409,
      "loss": 5.7181,
      "mean_token_accuracy": 0.15858784765005113,
      "num_tokens": 47247789.0,
      "step": 23315
    },
    {
      "entropy": 6.056094169616699,
      "epoch": 2.037927117014769,
      "grad_norm": 2.0,
      "learning_rate": 0.00045835656864990504,
      "loss": 5.7274,
      "mean_token_accuracy": 0.16748207956552505,
      "num_tokens": 47257901.0,
      "step": 23320
    },
    {
      "entropy": 6.083256196975708,
      "epoch": 2.0383640653674737,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004583385066621941,
      "loss": 5.72,
      "mean_token_accuracy": 0.16234707236289977,
      "num_tokens": 47268362.0,
      "step": 23325
    },
    {
      "entropy": 6.068030786514282,
      "epoch": 2.038801013720178,
      "grad_norm": 1.484375,
      "learning_rate": 0.00045832044115795464,
      "loss": 5.7234,
      "mean_token_accuracy": 0.16711512506008147,
      "num_tokens": 47278297.0,
      "step": 23330
    },
    {
      "entropy": 6.080295276641846,
      "epoch": 2.039237962072883,
      "grad_norm": 1.9375,
      "learning_rate": 0.00045830237213753287,
      "loss": 5.7304,
      "mean_token_accuracy": 0.15982200652360917,
      "num_tokens": 47288795.0,
      "step": 23335
    },
    {
      "entropy": 6.062101602554321,
      "epoch": 2.0396749104255876,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004582842996012757,
      "loss": 5.6735,
      "mean_token_accuracy": 0.166242578625679,
      "num_tokens": 47298373.0,
      "step": 23340
    },
    {
      "entropy": 6.057362079620361,
      "epoch": 2.0401118587782925,
      "grad_norm": 3.25,
      "learning_rate": 0.00045826622354952956,
      "loss": 5.7632,
      "mean_token_accuracy": 0.1604417771100998,
      "num_tokens": 47308160.0,
      "step": 23345
    },
    {
      "entropy": 6.0754077434539795,
      "epoch": 2.040548807130997,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045824814398264126,
      "loss": 5.7416,
      "mean_token_accuracy": 0.16330352574586868,
      "num_tokens": 47317626.0,
      "step": 23350
    },
    {
      "entropy": 6.113779163360595,
      "epoch": 2.040985755483702,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00045823006090095757,
      "loss": 5.7754,
      "mean_token_accuracy": 0.16294741183519362,
      "num_tokens": 47327221.0,
      "step": 23355
    },
    {
      "entropy": 6.09765944480896,
      "epoch": 2.0414227038364063,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004582119743048253,
      "loss": 5.7334,
      "mean_token_accuracy": 0.1690336987376213,
      "num_tokens": 47337385.0,
      "step": 23360
    },
    {
      "entropy": 6.016291379928589,
      "epoch": 2.0418596521891113,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004581938841945912,
      "loss": 5.7029,
      "mean_token_accuracy": 0.16142885386943817,
      "num_tokens": 47346894.0,
      "step": 23365
    },
    {
      "entropy": 6.1072886943817135,
      "epoch": 2.042296600541816,
      "grad_norm": 2.375,
      "learning_rate": 0.00045817579057060254,
      "loss": 5.7153,
      "mean_token_accuracy": 0.16720839589834213,
      "num_tokens": 47356192.0,
      "step": 23370
    },
    {
      "entropy": 6.075734567642212,
      "epoch": 2.0427335488945206,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004581576934332063,
      "loss": 5.6649,
      "mean_token_accuracy": 0.16383433789014817,
      "num_tokens": 47366704.0,
      "step": 23375
    },
    {
      "entropy": 6.037346124649048,
      "epoch": 2.0431704972472255,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004581395927827494,
      "loss": 5.6758,
      "mean_token_accuracy": 0.1695689707994461,
      "num_tokens": 47377068.0,
      "step": 23380
    },
    {
      "entropy": 6.032965660095215,
      "epoch": 2.04360744559993,
      "grad_norm": 1.75,
      "learning_rate": 0.00045812148861957923,
      "loss": 5.7069,
      "mean_token_accuracy": 0.17380177229642868,
      "num_tokens": 47387268.0,
      "step": 23385
    },
    {
      "entropy": 6.125617122650146,
      "epoch": 2.044044393952635,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004581033809440429,
      "loss": 5.6677,
      "mean_token_accuracy": 0.16925155818462373,
      "num_tokens": 47397508.0,
      "step": 23390
    },
    {
      "entropy": 6.030388307571411,
      "epoch": 2.0444813423053394,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004580852697564878,
      "loss": 5.7246,
      "mean_token_accuracy": 0.16023533195257186,
      "num_tokens": 47407871.0,
      "step": 23395
    },
    {
      "entropy": 6.046463251113892,
      "epoch": 2.0449182906580443,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00045806715505726135,
      "loss": 5.783,
      "mean_token_accuracy": 0.15346186906099318,
      "num_tokens": 47417089.0,
      "step": 23400
    },
    {
      "entropy": 6.138525772094726,
      "epoch": 2.045355239010749,
      "grad_norm": 2.90625,
      "learning_rate": 0.00045804903684671083,
      "loss": 5.6675,
      "mean_token_accuracy": 0.16763502955436707,
      "num_tokens": 47426746.0,
      "step": 23405
    },
    {
      "entropy": 6.101312923431396,
      "epoch": 2.0457921873634537,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00045803091512518397,
      "loss": 5.7355,
      "mean_token_accuracy": 0.16189888268709182,
      "num_tokens": 47437873.0,
      "step": 23410
    },
    {
      "entropy": 5.9859551906585695,
      "epoch": 2.046229135716158,
      "grad_norm": 2.265625,
      "learning_rate": 0.00045801278989302815,
      "loss": 5.6982,
      "mean_token_accuracy": 0.1693367600440979,
      "num_tokens": 47448520.0,
      "step": 23415
    },
    {
      "entropy": 6.1047272205352785,
      "epoch": 2.046666084068863,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004579946611505912,
      "loss": 5.7401,
      "mean_token_accuracy": 0.16150085777044296,
      "num_tokens": 47458323.0,
      "step": 23420
    },
    {
      "entropy": 6.124477434158325,
      "epoch": 2.0471030324215675,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004579765288982207,
      "loss": 5.815,
      "mean_token_accuracy": 0.1604028970003128,
      "num_tokens": 47468061.0,
      "step": 23425
    },
    {
      "entropy": 6.112526750564575,
      "epoch": 2.0475399807742725,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045795839313626455,
      "loss": 5.7412,
      "mean_token_accuracy": 0.16544900834560394,
      "num_tokens": 47477541.0,
      "step": 23430
    },
    {
      "entropy": 5.983056497573853,
      "epoch": 2.0479769291269774,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004579402538650705,
      "loss": 5.6951,
      "mean_token_accuracy": 0.16548738926649093,
      "num_tokens": 47488570.0,
      "step": 23435
    },
    {
      "entropy": 6.00838713645935,
      "epoch": 2.048413877479682,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004579221110849866,
      "loss": 5.7024,
      "mean_token_accuracy": 0.16170470863580705,
      "num_tokens": 47498386.0,
      "step": 23440
    },
    {
      "entropy": 6.082211351394653,
      "epoch": 2.0488508258323868,
      "grad_norm": 2.5,
      "learning_rate": 0.0004579039647963607,
      "loss": 5.6254,
      "mean_token_accuracy": 0.16956533789634703,
      "num_tokens": 47507292.0,
      "step": 23445
    },
    {
      "entropy": 6.104154443740844,
      "epoch": 2.0492877741850912,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00045788581499954094,
      "loss": 5.7017,
      "mean_token_accuracy": 0.16733806133270263,
      "num_tokens": 47516971.0,
      "step": 23450
    },
    {
      "entropy": 5.977691888809204,
      "epoch": 2.049724722537796,
      "grad_norm": 2.5,
      "learning_rate": 0.00045786766169487536,
      "loss": 5.6051,
      "mean_token_accuracy": 0.1789633020758629,
      "num_tokens": 47527233.0,
      "step": 23455
    },
    {
      "entropy": 6.041594934463501,
      "epoch": 2.0501616708905006,
      "grad_norm": 1.703125,
      "learning_rate": 0.00045784950488271227,
      "loss": 5.6735,
      "mean_token_accuracy": 0.1631466418504715,
      "num_tokens": 47537542.0,
      "step": 23460
    },
    {
      "entropy": 6.072893190383911,
      "epoch": 2.0505986192432055,
      "grad_norm": 1.828125,
      "learning_rate": 0.00045783134456339985,
      "loss": 5.6575,
      "mean_token_accuracy": 0.16731716841459274,
      "num_tokens": 47546613.0,
      "step": 23465
    },
    {
      "entropy": 6.0173101902008055,
      "epoch": 2.05103556759591,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00045781318073728647,
      "loss": 5.6747,
      "mean_token_accuracy": 0.17329651713371277,
      "num_tokens": 47555462.0,
      "step": 23470
    },
    {
      "entropy": 6.084464120864868,
      "epoch": 2.051472515948615,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004577950134047204,
      "loss": 5.6764,
      "mean_token_accuracy": 0.15978261679410935,
      "num_tokens": 47565429.0,
      "step": 23475
    },
    {
      "entropy": 6.122149562835693,
      "epoch": 2.0519094643013194,
      "grad_norm": 1.625,
      "learning_rate": 0.0004577768425660503,
      "loss": 5.8442,
      "mean_token_accuracy": 0.15314441919326782,
      "num_tokens": 47576021.0,
      "step": 23480
    },
    {
      "entropy": 6.108613204956055,
      "epoch": 2.0523464126540243,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004577586682216245,
      "loss": 5.7276,
      "mean_token_accuracy": 0.16554156690835953,
      "num_tokens": 47586016.0,
      "step": 23485
    },
    {
      "entropy": 6.084350061416626,
      "epoch": 2.052783361006729,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004577404903717917,
      "loss": 5.7096,
      "mean_token_accuracy": 0.16335168480873108,
      "num_tokens": 47595673.0,
      "step": 23490
    },
    {
      "entropy": 6.004195833206177,
      "epoch": 2.0532203093594337,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004577223090169006,
      "loss": 5.647,
      "mean_token_accuracy": 0.1706448957324028,
      "num_tokens": 47605252.0,
      "step": 23495
    },
    {
      "entropy": 6.0177093029022215,
      "epoch": 2.0536572577121386,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004577041241572998,
      "loss": 5.6911,
      "mean_token_accuracy": 0.16729393601417542,
      "num_tokens": 47614726.0,
      "step": 23500
    },
    {
      "entropy": 6.048759794235229,
      "epoch": 2.054094206064843,
      "grad_norm": 2.515625,
      "learning_rate": 0.00045768593579333816,
      "loss": 5.7369,
      "mean_token_accuracy": 0.17103956937789916,
      "num_tokens": 47625618.0,
      "step": 23505
    },
    {
      "entropy": 6.150296878814697,
      "epoch": 2.054531154417548,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004576677439253646,
      "loss": 5.7588,
      "mean_token_accuracy": 0.16223463267087937,
      "num_tokens": 47635702.0,
      "step": 23510
    },
    {
      "entropy": 6.009245252609253,
      "epoch": 2.0549681027702524,
      "grad_norm": 2.84375,
      "learning_rate": 0.00045764954855372796,
      "loss": 5.7707,
      "mean_token_accuracy": 0.16296250969171525,
      "num_tokens": 47647301.0,
      "step": 23515
    },
    {
      "entropy": 6.068942070007324,
      "epoch": 2.0554050511229573,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004576313496787772,
      "loss": 5.8111,
      "mean_token_accuracy": 0.15733586996793747,
      "num_tokens": 47656844.0,
      "step": 23520
    },
    {
      "entropy": 6.148377466201782,
      "epoch": 2.055841999475662,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004576131473008616,
      "loss": 5.6751,
      "mean_token_accuracy": 0.17040763050317764,
      "num_tokens": 47666196.0,
      "step": 23525
    },
    {
      "entropy": 6.018743991851807,
      "epoch": 2.0562789478283667,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004575949414203301,
      "loss": 5.6691,
      "mean_token_accuracy": 0.17179162353277205,
      "num_tokens": 47675329.0,
      "step": 23530
    },
    {
      "entropy": 6.005858707427978,
      "epoch": 2.056715896181071,
      "grad_norm": 1.6875,
      "learning_rate": 0.00045757673203753195,
      "loss": 5.687,
      "mean_token_accuracy": 0.163885460793972,
      "num_tokens": 47686806.0,
      "step": 23535
    },
    {
      "entropy": 6.071279573440552,
      "epoch": 2.057152844533776,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004575585191528164,
      "loss": 5.6727,
      "mean_token_accuracy": 0.16698219627141953,
      "num_tokens": 47696664.0,
      "step": 23540
    },
    {
      "entropy": 6.06217131614685,
      "epoch": 2.057589792886481,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00045754030276653275,
      "loss": 5.7563,
      "mean_token_accuracy": 0.16183976083993912,
      "num_tokens": 47707257.0,
      "step": 23545
    },
    {
      "entropy": 6.080485773086548,
      "epoch": 2.0580267412391855,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004575220828790305,
      "loss": 5.6437,
      "mean_token_accuracy": 0.1717106282711029,
      "num_tokens": 47716345.0,
      "step": 23550
    },
    {
      "entropy": 5.994209814071655,
      "epoch": 2.0584636895918904,
      "grad_norm": 1.984375,
      "learning_rate": 0.00045750385949065904,
      "loss": 5.6757,
      "mean_token_accuracy": 0.16945041865110397,
      "num_tokens": 47727779.0,
      "step": 23555
    },
    {
      "entropy": 6.060932254791259,
      "epoch": 2.058900637944595,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004574856326017679,
      "loss": 5.7433,
      "mean_token_accuracy": 0.1647959128022194,
      "num_tokens": 47737591.0,
      "step": 23560
    },
    {
      "entropy": 6.1271679401397705,
      "epoch": 2.0593375862973,
      "grad_norm": 2.0,
      "learning_rate": 0.00045746740221270664,
      "loss": 5.6758,
      "mean_token_accuracy": 0.16665468066930772,
      "num_tokens": 47746871.0,
      "step": 23565
    },
    {
      "entropy": 6.088408422470093,
      "epoch": 2.0597745346500043,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004574491683238251,
      "loss": 5.7259,
      "mean_token_accuracy": 0.16684252843260766,
      "num_tokens": 47757089.0,
      "step": 23570
    },
    {
      "entropy": 6.074892282485962,
      "epoch": 2.060211483002709,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004574309309354729,
      "loss": 5.7256,
      "mean_token_accuracy": 0.16311140805482865,
      "num_tokens": 47766853.0,
      "step": 23575
    },
    {
      "entropy": 6.068169498443604,
      "epoch": 2.0606484313554136,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00045741269004799984,
      "loss": 5.7118,
      "mean_token_accuracy": 0.16319112628698348,
      "num_tokens": 47776549.0,
      "step": 23580
    },
    {
      "entropy": 6.073141574859619,
      "epoch": 2.0610853797081186,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00045739444566175584,
      "loss": 5.7228,
      "mean_token_accuracy": 0.16541013419628142,
      "num_tokens": 47786641.0,
      "step": 23585
    },
    {
      "entropy": 6.109742784500122,
      "epoch": 2.061522328060823,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00045737619777709077,
      "loss": 5.751,
      "mean_token_accuracy": 0.1588936746120453,
      "num_tokens": 47796678.0,
      "step": 23590
    },
    {
      "entropy": 6.084485054016113,
      "epoch": 2.061959276413528,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004573579463943546,
      "loss": 5.6894,
      "mean_token_accuracy": 0.16632806211709977,
      "num_tokens": 47806678.0,
      "step": 23595
    },
    {
      "entropy": 6.064703798294067,
      "epoch": 2.062396224766233,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00045733969151389754,
      "loss": 5.7188,
      "mean_token_accuracy": 0.16451812237501146,
      "num_tokens": 47816619.0,
      "step": 23600
    },
    {
      "entropy": 6.015813159942627,
      "epoch": 2.0628331731189373,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004573214331360696,
      "loss": 5.7043,
      "mean_token_accuracy": 0.1667874738574028,
      "num_tokens": 47825437.0,
      "step": 23605
    },
    {
      "entropy": 6.018704986572265,
      "epoch": 2.0632701214716422,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004573031712612211,
      "loss": 5.6726,
      "mean_token_accuracy": 0.16585295647382736,
      "num_tokens": 47835604.0,
      "step": 23610
    },
    {
      "entropy": 5.983883857727051,
      "epoch": 2.0637070698243467,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004572849058897022,
      "loss": 5.6507,
      "mean_token_accuracy": 0.1762485161423683,
      "num_tokens": 47845299.0,
      "step": 23615
    },
    {
      "entropy": 5.964647531509399,
      "epoch": 2.0641440181770516,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004572666370218633,
      "loss": 5.6922,
      "mean_token_accuracy": 0.1707733988761902,
      "num_tokens": 47855293.0,
      "step": 23620
    },
    {
      "entropy": 6.090656661987305,
      "epoch": 2.064580966529756,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00045724836465805483,
      "loss": 5.7454,
      "mean_token_accuracy": 0.16722723394632338,
      "num_tokens": 47864809.0,
      "step": 23625
    },
    {
      "entropy": 6.061043167114258,
      "epoch": 2.065017914882461,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004572300887986272,
      "loss": 5.7354,
      "mean_token_accuracy": 0.1590841069817543,
      "num_tokens": 47874032.0,
      "step": 23630
    },
    {
      "entropy": 6.105671739578247,
      "epoch": 2.0654548632351655,
      "grad_norm": 3.53125,
      "learning_rate": 0.00045721180944393095,
      "loss": 5.6764,
      "mean_token_accuracy": 0.17344384491443635,
      "num_tokens": 47884465.0,
      "step": 23635
    },
    {
      "entropy": 6.01517071723938,
      "epoch": 2.0658918115878704,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004571935265943168,
      "loss": 5.6552,
      "mean_token_accuracy": 0.176028972864151,
      "num_tokens": 47894030.0,
      "step": 23640
    },
    {
      "entropy": 6.040750360488891,
      "epoch": 2.066328759940575,
      "grad_norm": 1.765625,
      "learning_rate": 0.00045717524025013526,
      "loss": 5.7054,
      "mean_token_accuracy": 0.15915032774209975,
      "num_tokens": 47904338.0,
      "step": 23645
    },
    {
      "entropy": 6.094125032424927,
      "epoch": 2.0667657082932798,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004571569504117372,
      "loss": 5.758,
      "mean_token_accuracy": 0.1622062787413597,
      "num_tokens": 47914384.0,
      "step": 23650
    },
    {
      "entropy": 6.096762228012085,
      "epoch": 2.0672026566459842,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004571386570794734,
      "loss": 5.6419,
      "mean_token_accuracy": 0.1658019930124283,
      "num_tokens": 47924386.0,
      "step": 23655
    },
    {
      "entropy": 6.081996631622315,
      "epoch": 2.067639604998689,
      "grad_norm": 1.828125,
      "learning_rate": 0.00045712036025369465,
      "loss": 5.735,
      "mean_token_accuracy": 0.16273192465305328,
      "num_tokens": 47935379.0,
      "step": 23660
    },
    {
      "entropy": 6.071133184432983,
      "epoch": 2.068076553351394,
      "grad_norm": 1.78125,
      "learning_rate": 0.000457102059934752,
      "loss": 5.7004,
      "mean_token_accuracy": 0.16829513311386107,
      "num_tokens": 47945657.0,
      "step": 23665
    },
    {
      "entropy": 6.073768854141235,
      "epoch": 2.0685135017040985,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00045708375612299646,
      "loss": 5.6909,
      "mean_token_accuracy": 0.16470894813537598,
      "num_tokens": 47956294.0,
      "step": 23670
    },
    {
      "entropy": 6.052688360214233,
      "epoch": 2.0689504500568034,
      "grad_norm": 1.71875,
      "learning_rate": 0.00045706544881877895,
      "loss": 5.8019,
      "mean_token_accuracy": 0.1565958246588707,
      "num_tokens": 47967362.0,
      "step": 23675
    },
    {
      "entropy": 6.126548719406128,
      "epoch": 2.069387398409508,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004570471380224508,
      "loss": 5.7254,
      "mean_token_accuracy": 0.16783382147550582,
      "num_tokens": 47977521.0,
      "step": 23680
    },
    {
      "entropy": 6.0476298332214355,
      "epoch": 2.069824346762213,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00045702882373436317,
      "loss": 5.7253,
      "mean_token_accuracy": 0.16484904140233994,
      "num_tokens": 47988485.0,
      "step": 23685
    },
    {
      "entropy": 6.040408277511597,
      "epoch": 2.0702612951149173,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004570105059548673,
      "loss": 5.745,
      "mean_token_accuracy": 0.15633460134267807,
      "num_tokens": 47998696.0,
      "step": 23690
    },
    {
      "entropy": 6.101782512664795,
      "epoch": 2.070698243467622,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004569921846843146,
      "loss": 5.5543,
      "mean_token_accuracy": 0.17884900271892548,
      "num_tokens": 48007843.0,
      "step": 23695
    },
    {
      "entropy": 6.043536567687989,
      "epoch": 2.0711351918203267,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004569738599230563,
      "loss": 5.676,
      "mean_token_accuracy": 0.16538106352090837,
      "num_tokens": 48017579.0,
      "step": 23700
    },
    {
      "entropy": 5.983508920669555,
      "epoch": 2.0715721401730316,
      "grad_norm": 2.0625,
      "learning_rate": 0.00045695553167144417,
      "loss": 5.6763,
      "mean_token_accuracy": 0.1708482876420021,
      "num_tokens": 48027267.0,
      "step": 23705
    },
    {
      "entropy": 6.084194660186768,
      "epoch": 2.072009088525736,
      "grad_norm": 3.515625,
      "learning_rate": 0.00045693719992982956,
      "loss": 5.7186,
      "mean_token_accuracy": 0.1645863637328148,
      "num_tokens": 48038064.0,
      "step": 23710
    },
    {
      "entropy": 6.138115644454956,
      "epoch": 2.072446036878441,
      "grad_norm": 2.0,
      "learning_rate": 0.00045691886469856406,
      "loss": 5.7954,
      "mean_token_accuracy": 0.15869563072919846,
      "num_tokens": 48048316.0,
      "step": 23715
    },
    {
      "entropy": 6.062406206130982,
      "epoch": 2.072882985231146,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004569005259779995,
      "loss": 5.7353,
      "mean_token_accuracy": 0.1625209003686905,
      "num_tokens": 48059083.0,
      "step": 23720
    },
    {
      "entropy": 6.0320930004119875,
      "epoch": 2.0733199335838504,
      "grad_norm": 2.3125,
      "learning_rate": 0.00045688218376848744,
      "loss": 5.679,
      "mean_token_accuracy": 0.16576071083545685,
      "num_tokens": 48069184.0,
      "step": 23725
    },
    {
      "entropy": 6.043241167068482,
      "epoch": 2.0737568819365553,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004568638380703799,
      "loss": 5.7192,
      "mean_token_accuracy": 0.1621640518307686,
      "num_tokens": 48078737.0,
      "step": 23730
    },
    {
      "entropy": 6.071523284912109,
      "epoch": 2.0741938302892597,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004568454888840285,
      "loss": 5.7238,
      "mean_token_accuracy": 0.16410600244998932,
      "num_tokens": 48088581.0,
      "step": 23735
    },
    {
      "entropy": 5.951975870132446,
      "epoch": 2.0746307786419647,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004568271362097855,
      "loss": 5.5829,
      "mean_token_accuracy": 0.19055956304073335,
      "num_tokens": 48098941.0,
      "step": 23740
    },
    {
      "entropy": 6.047421884536743,
      "epoch": 2.075067726994669,
      "grad_norm": 3.71875,
      "learning_rate": 0.00045680878004800265,
      "loss": 5.7133,
      "mean_token_accuracy": 0.16370175778865814,
      "num_tokens": 48108773.0,
      "step": 23745
    },
    {
      "entropy": 6.045295000076294,
      "epoch": 2.075504675347374,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004567904203990322,
      "loss": 5.7246,
      "mean_token_accuracy": 0.1714803546667099,
      "num_tokens": 48117699.0,
      "step": 23750
    },
    {
      "entropy": 6.0507810592651365,
      "epoch": 2.0759416237000785,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004567720572632261,
      "loss": 5.6626,
      "mean_token_accuracy": 0.17248574197292327,
      "num_tokens": 48127693.0,
      "step": 23755
    },
    {
      "entropy": 6.080027151107788,
      "epoch": 2.0763785720527834,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00045675369064093686,
      "loss": 5.7677,
      "mean_token_accuracy": 0.15649732500314711,
      "num_tokens": 48138073.0,
      "step": 23760
    },
    {
      "entropy": 6.005163764953613,
      "epoch": 2.076815520405488,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004567353205325165,
      "loss": 5.6067,
      "mean_token_accuracy": 0.173769573867321,
      "num_tokens": 48147504.0,
      "step": 23765
    },
    {
      "entropy": 6.063541984558105,
      "epoch": 2.077252468758193,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004567169469383175,
      "loss": 5.6777,
      "mean_token_accuracy": 0.16206910014152526,
      "num_tokens": 48156739.0,
      "step": 23770
    },
    {
      "entropy": 6.04261736869812,
      "epoch": 2.0776894171108973,
      "grad_norm": 3.109375,
      "learning_rate": 0.00045669856985869225,
      "loss": 5.6735,
      "mean_token_accuracy": 0.16540381759405137,
      "num_tokens": 48166883.0,
      "step": 23775
    },
    {
      "entropy": 6.029233455657959,
      "epoch": 2.078126365463602,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004566801892939932,
      "loss": 5.655,
      "mean_token_accuracy": 0.16548389345407485,
      "num_tokens": 48177738.0,
      "step": 23780
    },
    {
      "entropy": 6.044057607650757,
      "epoch": 2.078563313816307,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004566618052445729,
      "loss": 5.6316,
      "mean_token_accuracy": 0.16897036582231523,
      "num_tokens": 48187979.0,
      "step": 23785
    },
    {
      "entropy": 6.041925096511841,
      "epoch": 2.0790002621690116,
      "grad_norm": 1.75,
      "learning_rate": 0.000456643417710784,
      "loss": 5.7272,
      "mean_token_accuracy": 0.16338546872138976,
      "num_tokens": 48198579.0,
      "step": 23790
    },
    {
      "entropy": 6.078736448287964,
      "epoch": 2.0794372105217165,
      "grad_norm": 1.671875,
      "learning_rate": 0.00045662502669297915,
      "loss": 5.8109,
      "mean_token_accuracy": 0.16017695814371108,
      "num_tokens": 48208860.0,
      "step": 23795
    },
    {
      "entropy": 6.13704481124878,
      "epoch": 2.079874158874421,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004566066321915112,
      "loss": 5.7245,
      "mean_token_accuracy": 0.16858733147382737,
      "num_tokens": 48218842.0,
      "step": 23800
    },
    {
      "entropy": 6.077386474609375,
      "epoch": 2.080311107227126,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004565882342067328,
      "loss": 5.6642,
      "mean_token_accuracy": 0.16506125181913375,
      "num_tokens": 48229082.0,
      "step": 23805
    },
    {
      "entropy": 6.036945343017578,
      "epoch": 2.0807480555798303,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00045656983273899694,
      "loss": 5.7608,
      "mean_token_accuracy": 0.1585672065615654,
      "num_tokens": 48239158.0,
      "step": 23810
    },
    {
      "entropy": 6.040354490280151,
      "epoch": 2.0811850039325352,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004565514277886565,
      "loss": 5.7192,
      "mean_token_accuracy": 0.16000073552131652,
      "num_tokens": 48249663.0,
      "step": 23815
    },
    {
      "entropy": 6.11806287765503,
      "epoch": 2.0816219522852397,
      "grad_norm": 2.453125,
      "learning_rate": 0.00045653301935606456,
      "loss": 5.7548,
      "mean_token_accuracy": 0.15847268104553222,
      "num_tokens": 48260319.0,
      "step": 23820
    },
    {
      "entropy": 6.096063184738159,
      "epoch": 2.0820589006379446,
      "grad_norm": 1.828125,
      "learning_rate": 0.00045651460744157424,
      "loss": 5.7501,
      "mean_token_accuracy": 0.15552466362714767,
      "num_tokens": 48270107.0,
      "step": 23825
    },
    {
      "entropy": 6.00379786491394,
      "epoch": 2.082495848990649,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004564961920455387,
      "loss": 5.7438,
      "mean_token_accuracy": 0.16540255397558212,
      "num_tokens": 48280529.0,
      "step": 23830
    },
    {
      "entropy": 6.110647964477539,
      "epoch": 2.082932797343354,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004564777731683109,
      "loss": 5.7708,
      "mean_token_accuracy": 0.15864435136318206,
      "num_tokens": 48290672.0,
      "step": 23835
    },
    {
      "entropy": 6.169767904281616,
      "epoch": 2.083369745696059,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004564593508102445,
      "loss": 5.7751,
      "mean_token_accuracy": 0.16212031096220017,
      "num_tokens": 48301327.0,
      "step": 23840
    },
    {
      "entropy": 6.046170997619629,
      "epoch": 2.0838066940487634,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004564409249716925,
      "loss": 5.7254,
      "mean_token_accuracy": 0.16195729821920396,
      "num_tokens": 48310861.0,
      "step": 23845
    },
    {
      "entropy": 6.057533025741577,
      "epoch": 2.0842436424014683,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045642249565300866,
      "loss": 5.6968,
      "mean_token_accuracy": 0.16225571930408478,
      "num_tokens": 48320381.0,
      "step": 23850
    },
    {
      "entropy": 6.082344961166382,
      "epoch": 2.084680590754173,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00045640406285454616,
      "loss": 5.6613,
      "mean_token_accuracy": 0.16917819976806642,
      "num_tokens": 48330566.0,
      "step": 23855
    },
    {
      "entropy": 6.055637645721435,
      "epoch": 2.0851175391068777,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045638562657665885,
      "loss": 5.7243,
      "mean_token_accuracy": 0.16401671767234802,
      "num_tokens": 48340773.0,
      "step": 23860
    },
    {
      "entropy": 6.047181844711304,
      "epoch": 2.085554487459582,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00045636718681969996,
      "loss": 5.6913,
      "mean_token_accuracy": 0.16419028341770173,
      "num_tokens": 48350972.0,
      "step": 23865
    },
    {
      "entropy": 6.113470220565796,
      "epoch": 2.085991435812287,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004563487435840235,
      "loss": 5.711,
      "mean_token_accuracy": 0.16359276622533797,
      "num_tokens": 48360606.0,
      "step": 23870
    },
    {
      "entropy": 6.079755115509033,
      "epoch": 2.0864283841649915,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00045633029686998304,
      "loss": 5.7711,
      "mean_token_accuracy": 0.16660118997097015,
      "num_tokens": 48371970.0,
      "step": 23875
    },
    {
      "entropy": 6.025723552703857,
      "epoch": 2.0868653325176965,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004563118466779326,
      "loss": 5.6669,
      "mean_token_accuracy": 0.16819481551647186,
      "num_tokens": 48382917.0,
      "step": 23880
    },
    {
      "entropy": 6.058334684371948,
      "epoch": 2.087302280870401,
      "grad_norm": 1.6875,
      "learning_rate": 0.00045629339300822593,
      "loss": 5.7307,
      "mean_token_accuracy": 0.17084163874387742,
      "num_tokens": 48393428.0,
      "step": 23885
    },
    {
      "entropy": 6.135769653320312,
      "epoch": 2.087739229223106,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004562749358612168,
      "loss": 5.7941,
      "mean_token_accuracy": 0.15243995785713196,
      "num_tokens": 48404616.0,
      "step": 23890
    },
    {
      "entropy": 6.059495687484741,
      "epoch": 2.0881761775758108,
      "grad_norm": 1.75,
      "learning_rate": 0.0004562564752372596,
      "loss": 5.6324,
      "mean_token_accuracy": 0.1743351176381111,
      "num_tokens": 48414753.0,
      "step": 23895
    },
    {
      "entropy": 6.0356873035430905,
      "epoch": 2.0886131259285152,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004562380111367082,
      "loss": 5.7469,
      "mean_token_accuracy": 0.15469051599502565,
      "num_tokens": 48424759.0,
      "step": 23900
    },
    {
      "entropy": 6.049475860595703,
      "epoch": 2.08905007428122,
      "grad_norm": 1.859375,
      "learning_rate": 0.00045621954355991675,
      "loss": 5.7101,
      "mean_token_accuracy": 0.16619591265916825,
      "num_tokens": 48435158.0,
      "step": 23905
    },
    {
      "entropy": 6.016748380661011,
      "epoch": 2.0894870226339246,
      "grad_norm": 1.90625,
      "learning_rate": 0.00045620107250723946,
      "loss": 5.6811,
      "mean_token_accuracy": 0.16768094897270203,
      "num_tokens": 48445234.0,
      "step": 23910
    },
    {
      "entropy": 6.096313190460205,
      "epoch": 2.0899239709866295,
      "grad_norm": 1.859375,
      "learning_rate": 0.00045618259797903073,
      "loss": 5.7161,
      "mean_token_accuracy": 0.16511559635400772,
      "num_tokens": 48456053.0,
      "step": 23915
    },
    {
      "entropy": 6.097686100006103,
      "epoch": 2.090360919339334,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004561641199756447,
      "loss": 5.6937,
      "mean_token_accuracy": 0.16641195863485336,
      "num_tokens": 48466293.0,
      "step": 23920
    },
    {
      "entropy": 5.996440410614014,
      "epoch": 2.090797867692039,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00045614563849743605,
      "loss": 5.6787,
      "mean_token_accuracy": 0.17259181290864944,
      "num_tokens": 48476170.0,
      "step": 23925
    },
    {
      "entropy": 6.101129150390625,
      "epoch": 2.0912348160447434,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045612715354475907,
      "loss": 5.6869,
      "mean_token_accuracy": 0.1588258311152458,
      "num_tokens": 48485728.0,
      "step": 23930
    },
    {
      "entropy": 6.060774135589599,
      "epoch": 2.0916717643974483,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045610866511796845,
      "loss": 5.7098,
      "mean_token_accuracy": 0.16363725066184998,
      "num_tokens": 48496531.0,
      "step": 23935
    },
    {
      "entropy": 6.008128929138183,
      "epoch": 2.0921087127501528,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045609017321741863,
      "loss": 5.6508,
      "mean_token_accuracy": 0.16842157393693924,
      "num_tokens": 48506006.0,
      "step": 23940
    },
    {
      "entropy": 6.096497821807861,
      "epoch": 2.0925456611028577,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004560716778434644,
      "loss": 5.7357,
      "mean_token_accuracy": 0.16811443567276002,
      "num_tokens": 48515902.0,
      "step": 23945
    },
    {
      "entropy": 6.110761547088623,
      "epoch": 2.0929826094555626,
      "grad_norm": 2.546875,
      "learning_rate": 0.00045605317899646066,
      "loss": 5.7469,
      "mean_token_accuracy": 0.16430339515209197,
      "num_tokens": 48526263.0,
      "step": 23950
    },
    {
      "entropy": 6.062701463699341,
      "epoch": 2.093419557808267,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00045603467667676195,
      "loss": 5.7233,
      "mean_token_accuracy": 0.16501651257276534,
      "num_tokens": 48537415.0,
      "step": 23955
    },
    {
      "entropy": 6.1258796691894535,
      "epoch": 2.093856506160972,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004560161708847233,
      "loss": 5.7785,
      "mean_token_accuracy": 0.16411732584238053,
      "num_tokens": 48548911.0,
      "step": 23960
    },
    {
      "entropy": 6.116952133178711,
      "epoch": 2.0942934545136764,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004559976616206996,
      "loss": 5.7175,
      "mean_token_accuracy": 0.1643388256430626,
      "num_tokens": 48560266.0,
      "step": 23965
    },
    {
      "entropy": 6.123257446289062,
      "epoch": 2.0947304028663813,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00045597914888504597,
      "loss": 5.7856,
      "mean_token_accuracy": 0.155583056807518,
      "num_tokens": 48570696.0,
      "step": 23970
    },
    {
      "entropy": 6.0772990703582765,
      "epoch": 2.095167351219086,
      "grad_norm": 1.765625,
      "learning_rate": 0.00045596063267811746,
      "loss": 5.6805,
      "mean_token_accuracy": 0.1710346445441246,
      "num_tokens": 48580725.0,
      "step": 23975
    },
    {
      "entropy": 6.013621759414673,
      "epoch": 2.0956042995717907,
      "grad_norm": 2.25,
      "learning_rate": 0.0004559421130002691,
      "loss": 5.7706,
      "mean_token_accuracy": 0.16844784766435622,
      "num_tokens": 48591968.0,
      "step": 23980
    },
    {
      "entropy": 6.044961404800415,
      "epoch": 2.096041247924495,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004559235898518563,
      "loss": 5.6788,
      "mean_token_accuracy": 0.17243699133396148,
      "num_tokens": 48602074.0,
      "step": 23985
    },
    {
      "entropy": 6.066086483001709,
      "epoch": 2.0964781962772,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004559050632332342,
      "loss": 5.7248,
      "mean_token_accuracy": 0.16666474044322968,
      "num_tokens": 48613323.0,
      "step": 23990
    },
    {
      "entropy": 6.105467462539673,
      "epoch": 2.0969151446299046,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004558865331447581,
      "loss": 5.749,
      "mean_token_accuracy": 0.16105140149593353,
      "num_tokens": 48623386.0,
      "step": 23995
    },
    {
      "entropy": 6.014468860626221,
      "epoch": 2.0973520929826095,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004558679995867837,
      "loss": 5.6557,
      "mean_token_accuracy": 0.17858687341213225,
      "num_tokens": 48633379.0,
      "step": 24000
    },
    {
      "epoch": 2.0973520929826095,
      "eval_entropy": 5.853188803676707,
      "eval_loss": 5.781116485595703,
      "eval_mean_token_accuracy": 0.16994275756040195,
      "eval_num_tokens": 48633379.0,
      "eval_runtime": 28.4299,
      "eval_samples_per_second": 1301.938,
      "eval_steps_per_second": 162.751,
      "step": 24000
    },
    {
      "entropy": 6.071912717819214,
      "epoch": 2.097789041335314,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004558494625596662,
      "loss": 5.6834,
      "mean_token_accuracy": 0.16656850278377533,
      "num_tokens": 48643918.0,
      "step": 24005
    },
    {
      "entropy": 6.036042356491089,
      "epoch": 2.098225989688019,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004558309220637612,
      "loss": 5.7929,
      "mean_token_accuracy": 0.16240979880094528,
      "num_tokens": 48654534.0,
      "step": 24010
    },
    {
      "entropy": 6.040375089645385,
      "epoch": 2.098662938040724,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004558123780994244,
      "loss": 5.6979,
      "mean_token_accuracy": 0.1700185641646385,
      "num_tokens": 48664928.0,
      "step": 24015
    },
    {
      "entropy": 6.05636682510376,
      "epoch": 2.0990998863934283,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004557938306670115,
      "loss": 5.6896,
      "mean_token_accuracy": 0.16652902215719223,
      "num_tokens": 48675441.0,
      "step": 24020
    },
    {
      "entropy": 6.004779243469239,
      "epoch": 2.099536834746133,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00045577527976687813,
      "loss": 5.6172,
      "mean_token_accuracy": 0.17237697839736937,
      "num_tokens": 48685173.0,
      "step": 24025
    },
    {
      "entropy": 5.96911849975586,
      "epoch": 2.0999737830988376,
      "grad_norm": 1.484375,
      "learning_rate": 0.00045575672539938014,
      "loss": 5.6746,
      "mean_token_accuracy": 0.1710652858018875,
      "num_tokens": 48695539.0,
      "step": 24030
    },
    {
      "entropy": 6.014195585250855,
      "epoch": 2.1004107314515426,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00045573816756487345,
      "loss": 5.7331,
      "mean_token_accuracy": 0.16641592532396315,
      "num_tokens": 48705977.0,
      "step": 24035
    },
    {
      "entropy": 6.022192907333374,
      "epoch": 2.100847679804247,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004557196062637141,
      "loss": 5.6521,
      "mean_token_accuracy": 0.16729170382022857,
      "num_tokens": 48715924.0,
      "step": 24040
    },
    {
      "entropy": 6.0987104892730715,
      "epoch": 2.101284628156952,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004557010414962579,
      "loss": 5.7396,
      "mean_token_accuracy": 0.16033118963241577,
      "num_tokens": 48727071.0,
      "step": 24045
    },
    {
      "entropy": 6.098828363418579,
      "epoch": 2.1017215765096564,
      "grad_norm": 2.90625,
      "learning_rate": 0.00045568247326286103,
      "loss": 5.7158,
      "mean_token_accuracy": 0.16445517241954805,
      "num_tokens": 48736959.0,
      "step": 24050
    },
    {
      "entropy": 5.995248699188233,
      "epoch": 2.1021585248623613,
      "grad_norm": 1.765625,
      "learning_rate": 0.00045566390156387974,
      "loss": 5.7153,
      "mean_token_accuracy": 0.1647403880953789,
      "num_tokens": 48748419.0,
      "step": 24055
    },
    {
      "entropy": 6.020018529891968,
      "epoch": 2.102595473215066,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004556453263996701,
      "loss": 5.6811,
      "mean_token_accuracy": 0.1706630617380142,
      "num_tokens": 48758480.0,
      "step": 24060
    },
    {
      "entropy": 6.094992351531983,
      "epoch": 2.1030324215677707,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00045562674777058844,
      "loss": 5.8119,
      "mean_token_accuracy": 0.16410609930753708,
      "num_tokens": 48768182.0,
      "step": 24065
    },
    {
      "entropy": 6.080344390869141,
      "epoch": 2.103469369920475,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004556081656769911,
      "loss": 5.6859,
      "mean_token_accuracy": 0.16671450287103654,
      "num_tokens": 48777729.0,
      "step": 24070
    },
    {
      "entropy": 6.088178968429565,
      "epoch": 2.10390631827318,
      "grad_norm": 1.6875,
      "learning_rate": 0.00045558958011923456,
      "loss": 5.6654,
      "mean_token_accuracy": 0.17570373862981797,
      "num_tokens": 48787109.0,
      "step": 24075
    },
    {
      "entropy": 6.099465322494507,
      "epoch": 2.104343266625885,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004555709910976752,
      "loss": 5.7112,
      "mean_token_accuracy": 0.16232388317584992,
      "num_tokens": 48797152.0,
      "step": 24080
    },
    {
      "entropy": 6.037410926818848,
      "epoch": 2.1047802149785895,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00045555239861266964,
      "loss": 5.6711,
      "mean_token_accuracy": 0.17254701852798462,
      "num_tokens": 48806288.0,
      "step": 24085
    },
    {
      "entropy": 5.93019733428955,
      "epoch": 2.1052171633312944,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004555338026645744,
      "loss": 5.6206,
      "mean_token_accuracy": 0.1796501621603966,
      "num_tokens": 48816025.0,
      "step": 24090
    },
    {
      "entropy": 6.071573495864868,
      "epoch": 2.105654111683999,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004555152032537463,
      "loss": 5.7443,
      "mean_token_accuracy": 0.16335502564907073,
      "num_tokens": 48827150.0,
      "step": 24095
    },
    {
      "entropy": 6.021552228927613,
      "epoch": 2.1060910600367038,
      "grad_norm": 1.859375,
      "learning_rate": 0.00045549660038054205,
      "loss": 5.6729,
      "mean_token_accuracy": 0.1602245956659317,
      "num_tokens": 48837190.0,
      "step": 24100
    },
    {
      "entropy": 6.073708772659302,
      "epoch": 2.1065280083894082,
      "grad_norm": 1.984375,
      "learning_rate": 0.00045547799404531826,
      "loss": 5.7378,
      "mean_token_accuracy": 0.16557217091321946,
      "num_tokens": 48847573.0,
      "step": 24105
    },
    {
      "entropy": 6.0290977478027346,
      "epoch": 2.106964956742113,
      "grad_norm": 1.5,
      "learning_rate": 0.00045545938424843214,
      "loss": 5.6165,
      "mean_token_accuracy": 0.1664557009935379,
      "num_tokens": 48858966.0,
      "step": 24110
    },
    {
      "entropy": 6.026561975479126,
      "epoch": 2.1074019050948176,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004554407709902404,
      "loss": 5.7253,
      "mean_token_accuracy": 0.1598760649561882,
      "num_tokens": 48869027.0,
      "step": 24115
    },
    {
      "entropy": 6.014429903030395,
      "epoch": 2.1078388534475225,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004554221542711001,
      "loss": 5.7175,
      "mean_token_accuracy": 0.17282819598913193,
      "num_tokens": 48879325.0,
      "step": 24120
    },
    {
      "entropy": 6.004347467422486,
      "epoch": 2.108275801800227,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045540353409136836,
      "loss": 5.6179,
      "mean_token_accuracy": 0.17450709342956544,
      "num_tokens": 48888939.0,
      "step": 24125
    },
    {
      "entropy": 6.00196418762207,
      "epoch": 2.108712750152932,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045538491045140234,
      "loss": 5.6897,
      "mean_token_accuracy": 0.16573662757873536,
      "num_tokens": 48898997.0,
      "step": 24130
    },
    {
      "entropy": 6.0946276664733885,
      "epoch": 2.109149698505637,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00045536628335155914,
      "loss": 5.6803,
      "mean_token_accuracy": 0.1664903074502945,
      "num_tokens": 48909063.0,
      "step": 24135
    },
    {
      "entropy": 5.976613712310791,
      "epoch": 2.1095866468583413,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00045534765279219626,
      "loss": 5.5753,
      "mean_token_accuracy": 0.18109840005636216,
      "num_tokens": 48918694.0,
      "step": 24140
    },
    {
      "entropy": 6.037923431396484,
      "epoch": 2.110023595211046,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004553290187736708,
      "loss": 5.781,
      "mean_token_accuracy": 0.16262913048267363,
      "num_tokens": 48928223.0,
      "step": 24145
    },
    {
      "entropy": 6.063584709167481,
      "epoch": 2.1104605435637507,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004553103812963402,
      "loss": 5.7036,
      "mean_token_accuracy": 0.1669417917728424,
      "num_tokens": 48938193.0,
      "step": 24150
    },
    {
      "entropy": 6.083756399154663,
      "epoch": 2.1108974919164556,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00045529174036056217,
      "loss": 5.7878,
      "mean_token_accuracy": 0.15493564158678055,
      "num_tokens": 48949693.0,
      "step": 24155
    },
    {
      "entropy": 6.014966154098511,
      "epoch": 2.11133444026916,
      "grad_norm": 1.71875,
      "learning_rate": 0.00045527309596669397,
      "loss": 5.6706,
      "mean_token_accuracy": 0.171674944460392,
      "num_tokens": 48960199.0,
      "step": 24160
    },
    {
      "entropy": 6.070574426651001,
      "epoch": 2.111771388621865,
      "grad_norm": 1.515625,
      "learning_rate": 0.00045525444811509336,
      "loss": 5.7332,
      "mean_token_accuracy": 0.16958037614822388,
      "num_tokens": 48969380.0,
      "step": 24165
    },
    {
      "entropy": 6.0151214599609375,
      "epoch": 2.1122083369745694,
      "grad_norm": 1.71875,
      "learning_rate": 0.000455235796806118,
      "loss": 5.6398,
      "mean_token_accuracy": 0.16839511692523956,
      "num_tokens": 48978963.0,
      "step": 24170
    },
    {
      "entropy": 6.061747312545776,
      "epoch": 2.1126452853272744,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004552171420401256,
      "loss": 5.7662,
      "mean_token_accuracy": 0.16853755712509155,
      "num_tokens": 48989568.0,
      "step": 24175
    },
    {
      "entropy": 6.132940721511841,
      "epoch": 2.113082233679979,
      "grad_norm": 1.5,
      "learning_rate": 0.00045519848381747396,
      "loss": 5.7491,
      "mean_token_accuracy": 0.16357953399419783,
      "num_tokens": 49000684.0,
      "step": 24180
    },
    {
      "entropy": 6.087660646438598,
      "epoch": 2.1135191820326837,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000455179822138521,
      "loss": 5.7921,
      "mean_token_accuracy": 0.15347075909376146,
      "num_tokens": 49010194.0,
      "step": 24185
    },
    {
      "entropy": 6.0459792137146,
      "epoch": 2.1139561303853887,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00045516115700362466,
      "loss": 5.6872,
      "mean_token_accuracy": 0.168986576795578,
      "num_tokens": 49020180.0,
      "step": 24190
    },
    {
      "entropy": 6.128573751449585,
      "epoch": 2.114393078738093,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045514248841314295,
      "loss": 5.7272,
      "mean_token_accuracy": 0.16738078743219376,
      "num_tokens": 49030665.0,
      "step": 24195
    },
    {
      "entropy": 6.137915992736817,
      "epoch": 2.114830027090798,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004551238163674338,
      "loss": 5.7077,
      "mean_token_accuracy": 0.16761550456285476,
      "num_tokens": 49040197.0,
      "step": 24200
    },
    {
      "entropy": 6.006913423538208,
      "epoch": 2.1152669754435025,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00045510514086685553,
      "loss": 5.6869,
      "mean_token_accuracy": 0.16637028902769088,
      "num_tokens": 49049996.0,
      "step": 24205
    },
    {
      "entropy": 6.127159643173218,
      "epoch": 2.1157039237962074,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004550864619117664,
      "loss": 5.7901,
      "mean_token_accuracy": 0.15883069187402726,
      "num_tokens": 49061691.0,
      "step": 24210
    },
    {
      "entropy": 6.128850889205933,
      "epoch": 2.116140872148912,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004550677795025244,
      "loss": 5.8066,
      "mean_token_accuracy": 0.15586798936128615,
      "num_tokens": 49071605.0,
      "step": 24215
    },
    {
      "entropy": 6.049497127532959,
      "epoch": 2.116577820501617,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004550490936394881,
      "loss": 5.6453,
      "mean_token_accuracy": 0.17594247311353683,
      "num_tokens": 49081317.0,
      "step": 24220
    },
    {
      "entropy": 6.001521968841553,
      "epoch": 2.1170147688543213,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00045503040432301586,
      "loss": 5.6503,
      "mean_token_accuracy": 0.16471074372529984,
      "num_tokens": 49090662.0,
      "step": 24225
    },
    {
      "entropy": 6.1659773826599125,
      "epoch": 2.117451717207026,
      "grad_norm": 2.0,
      "learning_rate": 0.0004550117115534661,
      "loss": 5.8322,
      "mean_token_accuracy": 0.15231750756502152,
      "num_tokens": 49102046.0,
      "step": 24230
    },
    {
      "entropy": 6.104298067092896,
      "epoch": 2.1178886655597307,
      "grad_norm": 1.375,
      "learning_rate": 0.00045499301533119736,
      "loss": 5.6477,
      "mean_token_accuracy": 0.1701686143875122,
      "num_tokens": 49111047.0,
      "step": 24235
    },
    {
      "entropy": 6.037862396240234,
      "epoch": 2.1183256139124356,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004549743156565684,
      "loss": 5.6725,
      "mean_token_accuracy": 0.1613580569624901,
      "num_tokens": 49120606.0,
      "step": 24240
    },
    {
      "entropy": 6.0718451023101805,
      "epoch": 2.1187625622651405,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004549556125299377,
      "loss": 5.7421,
      "mean_token_accuracy": 0.16447239518165588,
      "num_tokens": 49130981.0,
      "step": 24245
    },
    {
      "entropy": 6.095796012878418,
      "epoch": 2.119199510617845,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004549369059516639,
      "loss": 5.7648,
      "mean_token_accuracy": 0.16095532700419427,
      "num_tokens": 49140843.0,
      "step": 24250
    },
    {
      "entropy": 6.091570615768433,
      "epoch": 2.11963645897055,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004549181959221061,
      "loss": 5.7525,
      "mean_token_accuracy": 0.15382864475250244,
      "num_tokens": 49150492.0,
      "step": 24255
    },
    {
      "entropy": 6.052959299087524,
      "epoch": 2.1200734073232543,
      "grad_norm": 1.65625,
      "learning_rate": 0.00045489948244162304,
      "loss": 5.6903,
      "mean_token_accuracy": 0.16430056244134902,
      "num_tokens": 49159973.0,
      "step": 24260
    },
    {
      "entropy": 6.1352156639099125,
      "epoch": 2.1205103556759592,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004548807655105736,
      "loss": 5.7877,
      "mean_token_accuracy": 0.15910400152206422,
      "num_tokens": 49170711.0,
      "step": 24265
    },
    {
      "entropy": 6.1181141376495365,
      "epoch": 2.1209473040286637,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00045486204512931684,
      "loss": 5.7473,
      "mean_token_accuracy": 0.1613219454884529,
      "num_tokens": 49181121.0,
      "step": 24270
    },
    {
      "entropy": 5.885158967971802,
      "epoch": 2.1213842523813686,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004548433212982118,
      "loss": 5.5923,
      "mean_token_accuracy": 0.18503270000219346,
      "num_tokens": 49191528.0,
      "step": 24275
    },
    {
      "entropy": 6.040424251556397,
      "epoch": 2.121821200734073,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004548245940176176,
      "loss": 5.7453,
      "mean_token_accuracy": 0.15896914005279542,
      "num_tokens": 49202244.0,
      "step": 24280
    },
    {
      "entropy": 6.044428586959839,
      "epoch": 2.122258149086778,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004548058632878935,
      "loss": 5.7058,
      "mean_token_accuracy": 0.16917437314987183,
      "num_tokens": 49212964.0,
      "step": 24285
    },
    {
      "entropy": 6.106925010681152,
      "epoch": 2.1226950974394825,
      "grad_norm": 1.8125,
      "learning_rate": 0.00045478712910939874,
      "loss": 5.7246,
      "mean_token_accuracy": 0.16333457678556443,
      "num_tokens": 49223478.0,
      "step": 24290
    },
    {
      "entropy": 6.1096936702728275,
      "epoch": 2.1231320457921874,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00045476839148249256,
      "loss": 5.7779,
      "mean_token_accuracy": 0.15842169374227524,
      "num_tokens": 49233933.0,
      "step": 24295
    },
    {
      "entropy": 6.046383476257324,
      "epoch": 2.123568994144892,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045474965040753453,
      "loss": 5.6327,
      "mean_token_accuracy": 0.1719887763261795,
      "num_tokens": 49243147.0,
      "step": 24300
    },
    {
      "entropy": 5.99568247795105,
      "epoch": 2.1240059424975968,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004547309058848839,
      "loss": 5.6434,
      "mean_token_accuracy": 0.16917731761932372,
      "num_tokens": 49252524.0,
      "step": 24305
    },
    {
      "entropy": 6.031407451629638,
      "epoch": 2.1244428908503017,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045471215791490047,
      "loss": 5.7174,
      "mean_token_accuracy": 0.16743783056735992,
      "num_tokens": 49261845.0,
      "step": 24310
    },
    {
      "entropy": 6.059521627426148,
      "epoch": 2.124879839203006,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00045469340649794353,
      "loss": 5.6624,
      "mean_token_accuracy": 0.17068691551685333,
      "num_tokens": 49272437.0,
      "step": 24315
    },
    {
      "entropy": 6.059312772750855,
      "epoch": 2.125316787555711,
      "grad_norm": 1.65625,
      "learning_rate": 0.000454674651634373,
      "loss": 5.6359,
      "mean_token_accuracy": 0.17694657295942307,
      "num_tokens": 49281687.0,
      "step": 24320
    },
    {
      "entropy": 5.991838788986206,
      "epoch": 2.1257537359084155,
      "grad_norm": 1.625,
      "learning_rate": 0.00045465589332454844,
      "loss": 5.5939,
      "mean_token_accuracy": 0.16968926936388015,
      "num_tokens": 49291524.0,
      "step": 24325
    },
    {
      "entropy": 6.045383596420288,
      "epoch": 2.1261906842611205,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00045463713156882976,
      "loss": 5.733,
      "mean_token_accuracy": 0.163093264400959,
      "num_tokens": 49303124.0,
      "step": 24330
    },
    {
      "entropy": 6.108282995223999,
      "epoch": 2.126627632613825,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00045461836636757674,
      "loss": 5.6534,
      "mean_token_accuracy": 0.1679699495434761,
      "num_tokens": 49313050.0,
      "step": 24335
    },
    {
      "entropy": 5.966231632232666,
      "epoch": 2.12706458096653,
      "grad_norm": 1.953125,
      "learning_rate": 0.00045459959772114935,
      "loss": 5.6253,
      "mean_token_accuracy": 0.17580663561820983,
      "num_tokens": 49321652.0,
      "step": 24340
    },
    {
      "entropy": 5.979233741760254,
      "epoch": 2.1275015293192343,
      "grad_norm": 1.703125,
      "learning_rate": 0.00045458082562990754,
      "loss": 5.6576,
      "mean_token_accuracy": 0.17458923459053038,
      "num_tokens": 49331713.0,
      "step": 24345
    },
    {
      "entropy": 6.0218034267425535,
      "epoch": 2.127938477671939,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004545620500942114,
      "loss": 5.6767,
      "mean_token_accuracy": 0.17044712603092194,
      "num_tokens": 49340370.0,
      "step": 24350
    },
    {
      "entropy": 6.0784543514251705,
      "epoch": 2.1283754260246437,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004545432711144211,
      "loss": 5.7227,
      "mean_token_accuracy": 0.16587792485952377,
      "num_tokens": 49349894.0,
      "step": 24355
    },
    {
      "entropy": 6.011568355560303,
      "epoch": 2.1288123743773486,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004545244886908967,
      "loss": 5.6043,
      "mean_token_accuracy": 0.17664652168750763,
      "num_tokens": 49359738.0,
      "step": 24360
    },
    {
      "entropy": 6.025173473358154,
      "epoch": 2.129249322730053,
      "grad_norm": 2.734375,
      "learning_rate": 0.00045450570282399856,
      "loss": 5.6948,
      "mean_token_accuracy": 0.16612299978733064,
      "num_tokens": 49370495.0,
      "step": 24365
    },
    {
      "entropy": 6.115649604797364,
      "epoch": 2.129686271082758,
      "grad_norm": 2.890625,
      "learning_rate": 0.000454486913514087,
      "loss": 5.7255,
      "mean_token_accuracy": 0.16978394538164138,
      "num_tokens": 49382191.0,
      "step": 24370
    },
    {
      "entropy": 6.047225427627564,
      "epoch": 2.130123219435463,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004544681207615224,
      "loss": 5.7278,
      "mean_token_accuracy": 0.16337917000055313,
      "num_tokens": 49391724.0,
      "step": 24375
    },
    {
      "entropy": 6.0808844566345215,
      "epoch": 2.1305601677881674,
      "grad_norm": 2.25,
      "learning_rate": 0.0004544493245666652,
      "loss": 5.6955,
      "mean_token_accuracy": 0.16830963939428328,
      "num_tokens": 49401117.0,
      "step": 24380
    },
    {
      "entropy": 5.983380746841431,
      "epoch": 2.1309971161408723,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004544305249298759,
      "loss": 5.6592,
      "mean_token_accuracy": 0.17382308691740037,
      "num_tokens": 49410731.0,
      "step": 24385
    },
    {
      "entropy": 6.02885479927063,
      "epoch": 2.1314340644935768,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004544117218515152,
      "loss": 5.7114,
      "mean_token_accuracy": 0.16590906530618668,
      "num_tokens": 49420373.0,
      "step": 24390
    },
    {
      "entropy": 6.051775598526001,
      "epoch": 2.1318710128462817,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00045439291533194356,
      "loss": 5.6401,
      "mean_token_accuracy": 0.16879219710826873,
      "num_tokens": 49430021.0,
      "step": 24395
    },
    {
      "entropy": 5.990060949325562,
      "epoch": 2.132307961198986,
      "grad_norm": 1.625,
      "learning_rate": 0.0004543741053715218,
      "loss": 5.612,
      "mean_token_accuracy": 0.1695458322763443,
      "num_tokens": 49439687.0,
      "step": 24400
    },
    {
      "entropy": 6.0082347869873045,
      "epoch": 2.132744909551691,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004543552919706107,
      "loss": 5.8138,
      "mean_token_accuracy": 0.15501240640878677,
      "num_tokens": 49450320.0,
      "step": 24405
    },
    {
      "entropy": 6.057753610610962,
      "epoch": 2.1331818579043955,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004543364751295712,
      "loss": 5.6222,
      "mean_token_accuracy": 0.16710594445466995,
      "num_tokens": 49461272.0,
      "step": 24410
    },
    {
      "entropy": 6.107842922210693,
      "epoch": 2.1336188062571004,
      "grad_norm": 3.015625,
      "learning_rate": 0.00045431765484876406,
      "loss": 5.7574,
      "mean_token_accuracy": 0.16442478746175765,
      "num_tokens": 49473087.0,
      "step": 24415
    },
    {
      "entropy": 5.996852731704712,
      "epoch": 2.134055754609805,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004542988311285505,
      "loss": 5.6662,
      "mean_token_accuracy": 0.17314796149730682,
      "num_tokens": 49483174.0,
      "step": 24420
    },
    {
      "entropy": 6.069650602340698,
      "epoch": 2.13449270296251,
      "grad_norm": 1.375,
      "learning_rate": 0.00045428000396929125,
      "loss": 5.6746,
      "mean_token_accuracy": 0.1704002171754837,
      "num_tokens": 49492267.0,
      "step": 24425
    },
    {
      "entropy": 6.0107231616973875,
      "epoch": 2.1349296513152147,
      "grad_norm": 1.875,
      "learning_rate": 0.00045426117337134757,
      "loss": 5.6437,
      "mean_token_accuracy": 0.16571990996599198,
      "num_tokens": 49502338.0,
      "step": 24430
    },
    {
      "entropy": 6.0528114318847654,
      "epoch": 2.135366599667919,
      "grad_norm": 1.484375,
      "learning_rate": 0.00045424233933508075,
      "loss": 5.7139,
      "mean_token_accuracy": 0.1700506940484047,
      "num_tokens": 49512991.0,
      "step": 24435
    },
    {
      "entropy": 6.031635475158692,
      "epoch": 2.135803548020624,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004542235018608519,
      "loss": 5.6003,
      "mean_token_accuracy": 0.1722400039434433,
      "num_tokens": 49522224.0,
      "step": 24440
    },
    {
      "entropy": 5.998216342926026,
      "epoch": 2.1362404963733286,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00045420466094902234,
      "loss": 5.709,
      "mean_token_accuracy": 0.16919582933187485,
      "num_tokens": 49531982.0,
      "step": 24445
    },
    {
      "entropy": 6.1102429866790775,
      "epoch": 2.1366774447260335,
      "grad_norm": 1.625,
      "learning_rate": 0.0004541858165999535,
      "loss": 5.6406,
      "mean_token_accuracy": 0.16931478977203368,
      "num_tokens": 49541487.0,
      "step": 24450
    },
    {
      "entropy": 6.052166843414307,
      "epoch": 2.137114393078738,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004541669688140068,
      "loss": 5.7631,
      "mean_token_accuracy": 0.15442387536168098,
      "num_tokens": 49552052.0,
      "step": 24455
    },
    {
      "entropy": 6.0252269268035885,
      "epoch": 2.137551341431443,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004541481175915438,
      "loss": 5.7105,
      "mean_token_accuracy": 0.163950052857399,
      "num_tokens": 49561807.0,
      "step": 24460
    },
    {
      "entropy": 6.095153284072876,
      "epoch": 2.1379882897841473,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000454129262932926,
      "loss": 5.7105,
      "mean_token_accuracy": 0.16682215929031372,
      "num_tokens": 49570687.0,
      "step": 24465
    },
    {
      "entropy": 6.06023325920105,
      "epoch": 2.1384252381368523,
      "grad_norm": 3.578125,
      "learning_rate": 0.00045411040483851506,
      "loss": 5.7195,
      "mean_token_accuracy": 0.16873498260974884,
      "num_tokens": 49581000.0,
      "step": 24470
    },
    {
      "entropy": 6.043283176422119,
      "epoch": 2.1388621864895567,
      "grad_norm": 2.375,
      "learning_rate": 0.00045409154330867274,
      "loss": 5.7619,
      "mean_token_accuracy": 0.16618835180997849,
      "num_tokens": 49590442.0,
      "step": 24475
    },
    {
      "entropy": 6.110496234893799,
      "epoch": 2.1392991348422616,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045407267834376075,
      "loss": 5.7011,
      "mean_token_accuracy": 0.1646003395318985,
      "num_tokens": 49600068.0,
      "step": 24480
    },
    {
      "entropy": 6.072856807708741,
      "epoch": 2.1397360831949666,
      "grad_norm": 1.890625,
      "learning_rate": 0.000454053809944141,
      "loss": 5.743,
      "mean_token_accuracy": 0.1574442505836487,
      "num_tokens": 49610696.0,
      "step": 24485
    },
    {
      "entropy": 6.089563322067261,
      "epoch": 2.140173031547671,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004540349381101753,
      "loss": 5.6996,
      "mean_token_accuracy": 0.16955415159463882,
      "num_tokens": 49619489.0,
      "step": 24490
    },
    {
      "entropy": 6.135804128646851,
      "epoch": 2.140609979900376,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00045401606284222563,
      "loss": 5.7055,
      "mean_token_accuracy": 0.15914686173200607,
      "num_tokens": 49629572.0,
      "step": 24495
    },
    {
      "entropy": 6.094232082366943,
      "epoch": 2.1410469282530804,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00045399718414065415,
      "loss": 5.6626,
      "mean_token_accuracy": 0.16561827361583709,
      "num_tokens": 49640035.0,
      "step": 24500
    },
    {
      "entropy": 5.987979459762573,
      "epoch": 2.1414838766057853,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004539783020058229,
      "loss": 5.6356,
      "mean_token_accuracy": 0.1787312552332878,
      "num_tokens": 49649671.0,
      "step": 24505
    },
    {
      "entropy": 6.015727758407593,
      "epoch": 2.14192082495849,
      "grad_norm": 1.390625,
      "learning_rate": 0.000453959416438094,
      "loss": 5.7117,
      "mean_token_accuracy": 0.1693689927458763,
      "num_tokens": 49659122.0,
      "step": 24510
    },
    {
      "entropy": 6.050340938568115,
      "epoch": 2.1423577733111947,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004539405274378297,
      "loss": 5.6405,
      "mean_token_accuracy": 0.1652588203549385,
      "num_tokens": 49669938.0,
      "step": 24515
    },
    {
      "entropy": 6.071232080459595,
      "epoch": 2.142794721663899,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004539216350053924,
      "loss": 5.6979,
      "mean_token_accuracy": 0.16494329273700714,
      "num_tokens": 49680783.0,
      "step": 24520
    },
    {
      "entropy": 6.030286884307861,
      "epoch": 2.143231670016604,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004539027391411443,
      "loss": 5.7492,
      "mean_token_accuracy": 0.16389731764793397,
      "num_tokens": 49691271.0,
      "step": 24525
    },
    {
      "entropy": 6.024486494064331,
      "epoch": 2.1436686183693086,
      "grad_norm": 2.453125,
      "learning_rate": 0.00045388383984544794,
      "loss": 5.7177,
      "mean_token_accuracy": 0.16004132777452468,
      "num_tokens": 49702496.0,
      "step": 24530
    },
    {
      "entropy": 6.049809360504151,
      "epoch": 2.1441055667220135,
      "grad_norm": 2.484375,
      "learning_rate": 0.00045386493711866577,
      "loss": 5.7257,
      "mean_token_accuracy": 0.16668348759412766,
      "num_tokens": 49712145.0,
      "step": 24535
    },
    {
      "entropy": 6.109587049484253,
      "epoch": 2.1445425150747184,
      "grad_norm": 1.71875,
      "learning_rate": 0.00045384603096116044,
      "loss": 5.6847,
      "mean_token_accuracy": 0.16602075845003128,
      "num_tokens": 49722243.0,
      "step": 24540
    },
    {
      "entropy": 6.1421887397766115,
      "epoch": 2.144979463427423,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004538271213732945,
      "loss": 5.7484,
      "mean_token_accuracy": 0.16142755150794982,
      "num_tokens": 49731804.0,
      "step": 24545
    },
    {
      "entropy": 6.0712299823760985,
      "epoch": 2.1454164117801278,
      "grad_norm": 2.046875,
      "learning_rate": 0.00045380820835543066,
      "loss": 5.6453,
      "mean_token_accuracy": 0.16475223004817963,
      "num_tokens": 49741275.0,
      "step": 24550
    },
    {
      "entropy": 5.981617069244384,
      "epoch": 2.1458533601328322,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004537892919079317,
      "loss": 5.6303,
      "mean_token_accuracy": 0.16829307079315187,
      "num_tokens": 49751140.0,
      "step": 24555
    },
    {
      "entropy": 6.03525013923645,
      "epoch": 2.146290308485537,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045377037203116053,
      "loss": 5.7281,
      "mean_token_accuracy": 0.16449174880981446,
      "num_tokens": 49760747.0,
      "step": 24560
    },
    {
      "entropy": 6.157594776153564,
      "epoch": 2.1467272568382416,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045375144872547975,
      "loss": 5.7473,
      "mean_token_accuracy": 0.15927583128213882,
      "num_tokens": 49771491.0,
      "step": 24565
    },
    {
      "entropy": 6.117906761169434,
      "epoch": 2.1471642051909465,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004537325219912527,
      "loss": 5.6688,
      "mean_token_accuracy": 0.1700497806072235,
      "num_tokens": 49781523.0,
      "step": 24570
    },
    {
      "entropy": 6.116732263565064,
      "epoch": 2.147601153543651,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004537135918288421,
      "loss": 5.7759,
      "mean_token_accuracy": 0.15980155467987062,
      "num_tokens": 49791560.0,
      "step": 24575
    },
    {
      "entropy": 6.022258996963501,
      "epoch": 2.148038101896356,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004536946582386112,
      "loss": 5.6262,
      "mean_token_accuracy": 0.1670142024755478,
      "num_tokens": 49800588.0,
      "step": 24580
    },
    {
      "entropy": 5.979426527023316,
      "epoch": 2.1484750502490604,
      "grad_norm": 2.71875,
      "learning_rate": 0.00045367572122092316,
      "loss": 5.6966,
      "mean_token_accuracy": 0.16892575323581696,
      "num_tokens": 49810381.0,
      "step": 24585
    },
    {
      "entropy": 6.089911413192749,
      "epoch": 2.1489119986017653,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004536567807761411,
      "loss": 5.7346,
      "mean_token_accuracy": 0.16223241537809371,
      "num_tokens": 49820931.0,
      "step": 24590
    },
    {
      "entropy": 6.074801778793335,
      "epoch": 2.14934894695447,
      "grad_norm": 2.609375,
      "learning_rate": 0.00045363783690462845,
      "loss": 5.7381,
      "mean_token_accuracy": 0.16493975818157197,
      "num_tokens": 49831264.0,
      "step": 24595
    },
    {
      "entropy": 6.042364168167114,
      "epoch": 2.1497858953071747,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004536188896067483,
      "loss": 5.7385,
      "mean_token_accuracy": 0.16553574353456496,
      "num_tokens": 49841656.0,
      "step": 24600
    },
    {
      "entropy": 6.063490581512451,
      "epoch": 2.1502228436598796,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004535999388828645,
      "loss": 5.6018,
      "mean_token_accuracy": 0.17415306866168975,
      "num_tokens": 49851127.0,
      "step": 24605
    },
    {
      "entropy": 6.078747844696045,
      "epoch": 2.150659792012584,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004535809847333402,
      "loss": 5.6775,
      "mean_token_accuracy": 0.1681322678923607,
      "num_tokens": 49860563.0,
      "step": 24610
    },
    {
      "entropy": 6.012295913696289,
      "epoch": 2.151096740365289,
      "grad_norm": 1.78125,
      "learning_rate": 0.000453562027158539,
      "loss": 5.6004,
      "mean_token_accuracy": 0.17001181840896606,
      "num_tokens": 49869741.0,
      "step": 24615
    },
    {
      "entropy": 5.994249582290649,
      "epoch": 2.1515336887179934,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004535430661588245,
      "loss": 5.723,
      "mean_token_accuracy": 0.16378276646137238,
      "num_tokens": 49879664.0,
      "step": 24620
    },
    {
      "entropy": 6.009509897232055,
      "epoch": 2.1519706370706984,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00045352410173456057,
      "loss": 5.6925,
      "mean_token_accuracy": 0.1671627551317215,
      "num_tokens": 49889322.0,
      "step": 24625
    },
    {
      "entropy": 6.105820178985596,
      "epoch": 2.152407585423403,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00045350513388611074,
      "loss": 5.7445,
      "mean_token_accuracy": 0.15982823967933654,
      "num_tokens": 49899172.0,
      "step": 24630
    },
    {
      "entropy": 6.126983976364135,
      "epoch": 2.1528445337761077,
      "grad_norm": 5.28125,
      "learning_rate": 0.00045348616261383897,
      "loss": 5.6913,
      "mean_token_accuracy": 0.1674824371933937,
      "num_tokens": 49909211.0,
      "step": 24635
    },
    {
      "entropy": 6.049197626113892,
      "epoch": 2.153281482128812,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00045346718791810896,
      "loss": 5.7168,
      "mean_token_accuracy": 0.169246706366539,
      "num_tokens": 49918884.0,
      "step": 24640
    },
    {
      "entropy": 6.021184682846069,
      "epoch": 2.153718430481517,
      "grad_norm": 2.140625,
      "learning_rate": 0.00045344820979928484,
      "loss": 5.6829,
      "mean_token_accuracy": 0.17149195224046707,
      "num_tokens": 49929558.0,
      "step": 24645
    },
    {
      "entropy": 6.029374313354492,
      "epoch": 2.1541553788342216,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045342922825773057,
      "loss": 5.7826,
      "mean_token_accuracy": 0.16390273496508598,
      "num_tokens": 49939831.0,
      "step": 24650
    },
    {
      "entropy": 6.0643235206604,
      "epoch": 2.1545923271869265,
      "grad_norm": 2.96875,
      "learning_rate": 0.00045341024329381024,
      "loss": 5.6645,
      "mean_token_accuracy": 0.16730208098888397,
      "num_tokens": 49949690.0,
      "step": 24655
    },
    {
      "entropy": 6.058569240570068,
      "epoch": 2.1550292755396314,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004533912549078879,
      "loss": 5.7065,
      "mean_token_accuracy": 0.15987886041402816,
      "num_tokens": 49959154.0,
      "step": 24660
    },
    {
      "entropy": 6.097177886962891,
      "epoch": 2.155466223892336,
      "grad_norm": 2.046875,
      "learning_rate": 0.00045337226310032783,
      "loss": 5.7268,
      "mean_token_accuracy": 0.1660898059606552,
      "num_tokens": 49970185.0,
      "step": 24665
    },
    {
      "entropy": 6.136545276641845,
      "epoch": 2.155903172245041,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004533532678714943,
      "loss": 5.7765,
      "mean_token_accuracy": 0.16707293093204498,
      "num_tokens": 49980489.0,
      "step": 24670
    },
    {
      "entropy": 6.0235679149627686,
      "epoch": 2.1563401205977453,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00045333426922175156,
      "loss": 5.7187,
      "mean_token_accuracy": 0.16931333243846894,
      "num_tokens": 49990038.0,
      "step": 24675
    },
    {
      "entropy": 6.0411842346191404,
      "epoch": 2.15677706895045,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004533152671514641,
      "loss": 5.6761,
      "mean_token_accuracy": 0.1709304690361023,
      "num_tokens": 49998888.0,
      "step": 24680
    },
    {
      "entropy": 6.022076177597046,
      "epoch": 2.1572140173031547,
      "grad_norm": 1.90625,
      "learning_rate": 0.00045329626166099653,
      "loss": 5.6497,
      "mean_token_accuracy": 0.17432095110416412,
      "num_tokens": 50009349.0,
      "step": 24685
    },
    {
      "entropy": 6.086068820953369,
      "epoch": 2.1576509656558596,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045327725275071313,
      "loss": 5.7814,
      "mean_token_accuracy": 0.15820435285568238,
      "num_tokens": 50019626.0,
      "step": 24690
    },
    {
      "entropy": 6.071339511871338,
      "epoch": 2.158087914008564,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004532582404209786,
      "loss": 5.6372,
      "mean_token_accuracy": 0.16643576174974442,
      "num_tokens": 50029281.0,
      "step": 24695
    },
    {
      "entropy": 6.083745813369751,
      "epoch": 2.158524862361269,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004532392246721576,
      "loss": 5.6897,
      "mean_token_accuracy": 0.16860136091709138,
      "num_tokens": 50038700.0,
      "step": 24700
    },
    {
      "entropy": 6.038189697265625,
      "epoch": 2.1589618107139734,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004532202055046149,
      "loss": 5.6631,
      "mean_token_accuracy": 0.1669368863105774,
      "num_tokens": 50048904.0,
      "step": 24705
    },
    {
      "entropy": 6.001475763320923,
      "epoch": 2.1593987590666783,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004532011829187153,
      "loss": 5.673,
      "mean_token_accuracy": 0.16599988639354707,
      "num_tokens": 50059188.0,
      "step": 24710
    },
    {
      "entropy": 6.00831708908081,
      "epoch": 2.159835707419383,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004531821569148236,
      "loss": 5.6452,
      "mean_token_accuracy": 0.1725715532898903,
      "num_tokens": 50069069.0,
      "step": 24715
    },
    {
      "entropy": 6.014378547668457,
      "epoch": 2.1602726557720877,
      "grad_norm": 1.609375,
      "learning_rate": 0.00045316312749330473,
      "loss": 5.6736,
      "mean_token_accuracy": 0.16877803206443787,
      "num_tokens": 50078904.0,
      "step": 24720
    },
    {
      "entropy": 5.990948629379273,
      "epoch": 2.1607096041247926,
      "grad_norm": 1.734375,
      "learning_rate": 0.00045314409465452385,
      "loss": 5.6563,
      "mean_token_accuracy": 0.17039738297462464,
      "num_tokens": 50089897.0,
      "step": 24725
    },
    {
      "entropy": 6.1007743835449215,
      "epoch": 2.161146552477497,
      "grad_norm": 2.1875,
      "learning_rate": 0.00045312505839884573,
      "loss": 5.7264,
      "mean_token_accuracy": 0.16441825330257415,
      "num_tokens": 50100030.0,
      "step": 24730
    },
    {
      "entropy": 6.096139049530029,
      "epoch": 2.161583500830202,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00045310601872663574,
      "loss": 5.6839,
      "mean_token_accuracy": 0.16433712095022202,
      "num_tokens": 50110043.0,
      "step": 24735
    },
    {
      "entropy": 6.023439979553222,
      "epoch": 2.1620204491829065,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00045308697563825894,
      "loss": 5.6957,
      "mean_token_accuracy": 0.16935904920101166,
      "num_tokens": 50120081.0,
      "step": 24740
    },
    {
      "entropy": 6.010700702667236,
      "epoch": 2.1624573975356114,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004530679291340807,
      "loss": 5.7023,
      "mean_token_accuracy": 0.16988381892442703,
      "num_tokens": 50129834.0,
      "step": 24745
    },
    {
      "entropy": 6.03944616317749,
      "epoch": 2.162894345888316,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004530488792144662,
      "loss": 5.7533,
      "mean_token_accuracy": 0.15560752749443055,
      "num_tokens": 50140853.0,
      "step": 24750
    },
    {
      "entropy": 6.088108205795288,
      "epoch": 2.1633312942410208,
      "grad_norm": 1.921875,
      "learning_rate": 0.000453029825879781,
      "loss": 5.775,
      "mean_token_accuracy": 0.16550963670015334,
      "num_tokens": 50151341.0,
      "step": 24755
    },
    {
      "entropy": 6.064794206619263,
      "epoch": 2.1637682425937252,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00045301076913039035,
      "loss": 5.648,
      "mean_token_accuracy": 0.17375319451093674,
      "num_tokens": 50160826.0,
      "step": 24760
    },
    {
      "entropy": 6.0124249935150145,
      "epoch": 2.16420519094643,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004529917089666599,
      "loss": 5.6445,
      "mean_token_accuracy": 0.17038801461458206,
      "num_tokens": 50170720.0,
      "step": 24765
    },
    {
      "entropy": 6.069689798355102,
      "epoch": 2.1646421392991346,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004529726453889552,
      "loss": 5.6906,
      "mean_token_accuracy": 0.1668471500277519,
      "num_tokens": 50180707.0,
      "step": 24770
    },
    {
      "entropy": 5.942304229736328,
      "epoch": 2.1650790876518395,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00045295357839764193,
      "loss": 5.6453,
      "mean_token_accuracy": 0.17577226161956788,
      "num_tokens": 50190714.0,
      "step": 24775
    },
    {
      "entropy": 6.079454040527343,
      "epoch": 2.1655160360045445,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004529345079930858,
      "loss": 5.7524,
      "mean_token_accuracy": 0.16267912536859513,
      "num_tokens": 50201023.0,
      "step": 24780
    },
    {
      "entropy": 6.104010534286499,
      "epoch": 2.165952984357249,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004529154341756525,
      "loss": 5.6883,
      "mean_token_accuracy": 0.1621909812092781,
      "num_tokens": 50211064.0,
      "step": 24785
    },
    {
      "entropy": 6.100869798660279,
      "epoch": 2.166389932709954,
      "grad_norm": 1.59375,
      "learning_rate": 0.00045289635694570796,
      "loss": 5.7408,
      "mean_token_accuracy": 0.15934788435697556,
      "num_tokens": 50220696.0,
      "step": 24790
    },
    {
      "entropy": 5.98139533996582,
      "epoch": 2.1668268810626583,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00045287727630361814,
      "loss": 5.6654,
      "mean_token_accuracy": 0.17451476603746413,
      "num_tokens": 50230629.0,
      "step": 24795
    },
    {
      "entropy": 6.005504465103149,
      "epoch": 2.167263829415363,
      "grad_norm": 1.703125,
      "learning_rate": 0.00045285819224974894,
      "loss": 5.6703,
      "mean_token_accuracy": 0.17429026812314988,
      "num_tokens": 50241042.0,
      "step": 24800
    },
    {
      "entropy": 5.986588716506958,
      "epoch": 2.1677007777680677,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004528391047844664,
      "loss": 5.6227,
      "mean_token_accuracy": 0.16892089545726777,
      "num_tokens": 50251731.0,
      "step": 24805
    },
    {
      "entropy": 6.076538372039795,
      "epoch": 2.1681377261207726,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004528200139081366,
      "loss": 5.7288,
      "mean_token_accuracy": 0.16460297107696534,
      "num_tokens": 50262893.0,
      "step": 24810
    },
    {
      "entropy": 6.081942081451416,
      "epoch": 2.168574674473477,
      "grad_norm": 1.5,
      "learning_rate": 0.0004528009196211258,
      "loss": 5.8189,
      "mean_token_accuracy": 0.1623825877904892,
      "num_tokens": 50273666.0,
      "step": 24815
    },
    {
      "entropy": 6.039019250869751,
      "epoch": 2.169011622826182,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004527818219238002,
      "loss": 5.6589,
      "mean_token_accuracy": 0.1681652769446373,
      "num_tokens": 50283165.0,
      "step": 24820
    },
    {
      "entropy": 6.0388147830963135,
      "epoch": 2.1694485711788865,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004527627208165261,
      "loss": 5.603,
      "mean_token_accuracy": 0.1672138825058937,
      "num_tokens": 50291958.0,
      "step": 24825
    },
    {
      "entropy": 6.09111442565918,
      "epoch": 2.1698855195315914,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00045274361629966985,
      "loss": 5.7598,
      "mean_token_accuracy": 0.1640464186668396,
      "num_tokens": 50302944.0,
      "step": 24830
    },
    {
      "entropy": 6.049911642074585,
      "epoch": 2.1703224678842963,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045272450837359783,
      "loss": 5.6904,
      "mean_token_accuracy": 0.1686779424548149,
      "num_tokens": 50313088.0,
      "step": 24835
    },
    {
      "entropy": 6.061626338958741,
      "epoch": 2.1707594162370007,
      "grad_norm": 1.46875,
      "learning_rate": 0.00045270539703867675,
      "loss": 5.7053,
      "mean_token_accuracy": 0.16333846747875214,
      "num_tokens": 50323121.0,
      "step": 24840
    },
    {
      "entropy": 6.055344343185425,
      "epoch": 2.1711963645897057,
      "grad_norm": 2.3125,
      "learning_rate": 0.000452686282295273,
      "loss": 5.7088,
      "mean_token_accuracy": 0.1691074401140213,
      "num_tokens": 50333091.0,
      "step": 24845
    },
    {
      "entropy": 6.0176136016845705,
      "epoch": 2.17163331294241,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00045266716414375324,
      "loss": 5.5996,
      "mean_token_accuracy": 0.17277619242668152,
      "num_tokens": 50342752.0,
      "step": 24850
    },
    {
      "entropy": 5.975629663467407,
      "epoch": 2.172070261295115,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045264804258448416,
      "loss": 5.6685,
      "mean_token_accuracy": 0.17036280930042266,
      "num_tokens": 50352213.0,
      "step": 24855
    },
    {
      "entropy": 6.009254693984985,
      "epoch": 2.1725072096478195,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004526289176178325,
      "loss": 5.7136,
      "mean_token_accuracy": 0.1650250568985939,
      "num_tokens": 50362586.0,
      "step": 24860
    },
    {
      "entropy": 6.03377685546875,
      "epoch": 2.1729441580005244,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004526097892441652,
      "loss": 5.5944,
      "mean_token_accuracy": 0.17195843756198884,
      "num_tokens": 50372742.0,
      "step": 24865
    },
    {
      "entropy": 5.900769281387329,
      "epoch": 2.173381106353229,
      "grad_norm": 1.5,
      "learning_rate": 0.00045259065746384904,
      "loss": 5.5859,
      "mean_token_accuracy": 0.18077024966478347,
      "num_tokens": 50382857.0,
      "step": 24870
    },
    {
      "entropy": 6.06974573135376,
      "epoch": 2.173818054705934,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000452571522277251,
      "loss": 5.7031,
      "mean_token_accuracy": 0.16867413222789765,
      "num_tokens": 50393638.0,
      "step": 24875
    },
    {
      "entropy": 6.064177227020264,
      "epoch": 2.1742550030586383,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004525523836847382,
      "loss": 5.7357,
      "mean_token_accuracy": 0.1641719900071621,
      "num_tokens": 50404928.0,
      "step": 24880
    },
    {
      "entropy": 5.969749021530151,
      "epoch": 2.174691951411343,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004525332416866776,
      "loss": 5.5924,
      "mean_token_accuracy": 0.17361184507608413,
      "num_tokens": 50414382.0,
      "step": 24885
    },
    {
      "entropy": 5.997975063323975,
      "epoch": 2.175128899764048,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004525140962834363,
      "loss": 5.6902,
      "mean_token_accuracy": 0.16739565432071685,
      "num_tokens": 50425203.0,
      "step": 24890
    },
    {
      "entropy": 6.1185791969299315,
      "epoch": 2.1755658481167526,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004524949474753817,
      "loss": 5.8512,
      "mean_token_accuracy": 0.15231299102306367,
      "num_tokens": 50436395.0,
      "step": 24895
    },
    {
      "entropy": 6.0219831466674805,
      "epoch": 2.1760027964694575,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004524757952628811,
      "loss": 5.6565,
      "mean_token_accuracy": 0.17479688674211502,
      "num_tokens": 50446990.0,
      "step": 24900
    },
    {
      "entropy": 6.035008573532105,
      "epoch": 2.176439744822162,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00045245663964630164,
      "loss": 5.7421,
      "mean_token_accuracy": 0.15826011598110198,
      "num_tokens": 50457964.0,
      "step": 24905
    },
    {
      "entropy": 6.131070804595947,
      "epoch": 2.176876693174867,
      "grad_norm": 4.0,
      "learning_rate": 0.0004524374806260108,
      "loss": 5.6563,
      "mean_token_accuracy": 0.16774121522903443,
      "num_tokens": 50467329.0,
      "step": 24910
    },
    {
      "entropy": 6.041166877746582,
      "epoch": 2.1773136415275713,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004524183182023762,
      "loss": 5.6713,
      "mean_token_accuracy": 0.16834862381219864,
      "num_tokens": 50476852.0,
      "step": 24915
    },
    {
      "entropy": 5.997316932678222,
      "epoch": 2.1777505898802763,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00045239915237576525,
      "loss": 5.668,
      "mean_token_accuracy": 0.1678691878914833,
      "num_tokens": 50486207.0,
      "step": 24920
    },
    {
      "entropy": 6.047819471359253,
      "epoch": 2.1781875382329807,
      "grad_norm": 1.640625,
      "learning_rate": 0.00045237998314654555,
      "loss": 5.7332,
      "mean_token_accuracy": 0.1712133601307869,
      "num_tokens": 50496457.0,
      "step": 24925
    },
    {
      "entropy": 6.0634383201599125,
      "epoch": 2.1786244865856856,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004523608105150849,
      "loss": 5.7301,
      "mean_token_accuracy": 0.15895978957414628,
      "num_tokens": 50506430.0,
      "step": 24930
    },
    {
      "entropy": 6.054630517959595,
      "epoch": 2.17906143493839,
      "grad_norm": 1.671875,
      "learning_rate": 0.000452341634481751,
      "loss": 5.6403,
      "mean_token_accuracy": 0.16312606632709503,
      "num_tokens": 50516204.0,
      "step": 24935
    },
    {
      "entropy": 6.008943510055542,
      "epoch": 2.179498383291095,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004523224550469115,
      "loss": 5.6729,
      "mean_token_accuracy": 0.16982772052288056,
      "num_tokens": 50525978.0,
      "step": 24940
    },
    {
      "entropy": 5.99343900680542,
      "epoch": 2.1799353316438,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004523032722109345,
      "loss": 5.6482,
      "mean_token_accuracy": 0.16492017656564711,
      "num_tokens": 50535579.0,
      "step": 24945
    },
    {
      "entropy": 6.030957317352295,
      "epoch": 2.1803722799965044,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004522840859741878,
      "loss": 5.655,
      "mean_token_accuracy": 0.16850954443216323,
      "num_tokens": 50545765.0,
      "step": 24950
    },
    {
      "entropy": 6.101188993453979,
      "epoch": 2.1808092283492093,
      "grad_norm": 1.609375,
      "learning_rate": 0.00045226489633703944,
      "loss": 5.8134,
      "mean_token_accuracy": 0.16496120840311052,
      "num_tokens": 50556237.0,
      "step": 24955
    },
    {
      "entropy": 6.146008682250977,
      "epoch": 2.181246176701914,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045224570329985747,
      "loss": 5.7761,
      "mean_token_accuracy": 0.16541430801153184,
      "num_tokens": 50567720.0,
      "step": 24960
    },
    {
      "entropy": 6.033121633529663,
      "epoch": 2.1816831250546187,
      "grad_norm": 1.9375,
      "learning_rate": 0.00045222650686301,
      "loss": 5.6505,
      "mean_token_accuracy": 0.16609373390674592,
      "num_tokens": 50577234.0,
      "step": 24965
    },
    {
      "entropy": 5.996250343322754,
      "epoch": 2.182120073407323,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004522073070268654,
      "loss": 5.734,
      "mean_token_accuracy": 0.16708829402923583,
      "num_tokens": 50587416.0,
      "step": 24970
    },
    {
      "entropy": 6.040242528915405,
      "epoch": 2.182557021760028,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004521881037917917,
      "loss": 5.6208,
      "mean_token_accuracy": 0.1801169127225876,
      "num_tokens": 50598317.0,
      "step": 24975
    },
    {
      "entropy": 6.091832065582276,
      "epoch": 2.1829939701127326,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004521688971581573,
      "loss": 5.7336,
      "mean_token_accuracy": 0.16388595178723336,
      "num_tokens": 50608507.0,
      "step": 24980
    },
    {
      "entropy": 6.005315685272217,
      "epoch": 2.1834309184654375,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00045214968712633066,
      "loss": 5.6627,
      "mean_token_accuracy": 0.1748122066259384,
      "num_tokens": 50618625.0,
      "step": 24985
    },
    {
      "entropy": 6.074332666397095,
      "epoch": 2.183867866818142,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004521304736966802,
      "loss": 5.6874,
      "mean_token_accuracy": 0.16913948357105255,
      "num_tokens": 50628696.0,
      "step": 24990
    },
    {
      "entropy": 6.123511552810669,
      "epoch": 2.184304815170847,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045211125686957443,
      "loss": 5.7567,
      "mean_token_accuracy": 0.1606249064207077,
      "num_tokens": 50639313.0,
      "step": 24995
    },
    {
      "entropy": 6.0085214138031,
      "epoch": 2.1847417635235513,
      "grad_norm": 2.046875,
      "learning_rate": 0.00045209203664538195,
      "loss": 5.6749,
      "mean_token_accuracy": 0.16148023307323456,
      "num_tokens": 50650715.0,
      "step": 25000
    },
    {
      "entropy": 6.04421911239624,
      "epoch": 2.1851787118762562,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004520728130244714,
      "loss": 5.669,
      "mean_token_accuracy": 0.17004555612802505,
      "num_tokens": 50660224.0,
      "step": 25005
    },
    {
      "entropy": 6.072675371170044,
      "epoch": 2.185615660228961,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004520535860072115,
      "loss": 5.6754,
      "mean_token_accuracy": 0.16907211095094682,
      "num_tokens": 50670346.0,
      "step": 25010
    },
    {
      "entropy": 6.003724670410156,
      "epoch": 2.1860526085816656,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00045203435559397113,
      "loss": 5.6753,
      "mean_token_accuracy": 0.16614235043525696,
      "num_tokens": 50680274.0,
      "step": 25015
    },
    {
      "entropy": 5.925422096252442,
      "epoch": 2.1864895569343705,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004520151217851189,
      "loss": 5.6727,
      "mean_token_accuracy": 0.1669465720653534,
      "num_tokens": 50690360.0,
      "step": 25020
    },
    {
      "entropy": 6.05927529335022,
      "epoch": 2.186926505287075,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000451995884581024,
      "loss": 5.7127,
      "mean_token_accuracy": 0.16910812258720398,
      "num_tokens": 50700798.0,
      "step": 25025
    },
    {
      "entropy": 6.078246450424194,
      "epoch": 2.18736345363978,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004519766439820552,
      "loss": 5.7284,
      "mean_token_accuracy": 0.1644744098186493,
      "num_tokens": 50710120.0,
      "step": 25030
    },
    {
      "entropy": 6.055625963211059,
      "epoch": 2.1878004019924844,
      "grad_norm": 1.5,
      "learning_rate": 0.0004519573999885817,
      "loss": 5.7951,
      "mean_token_accuracy": 0.1624461218714714,
      "num_tokens": 50719903.0,
      "step": 25035
    },
    {
      "entropy": 5.997176599502564,
      "epoch": 2.1882373503451893,
      "grad_norm": 1.6875,
      "learning_rate": 0.00045193815260097253,
      "loss": 5.7258,
      "mean_token_accuracy": 0.16596255898475648,
      "num_tokens": 50730632.0,
      "step": 25040
    },
    {
      "entropy": 6.138902997970581,
      "epoch": 2.1886742986978938,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045191890181959696,
      "loss": 5.7493,
      "mean_token_accuracy": 0.16155174374580383,
      "num_tokens": 50741263.0,
      "step": 25045
    },
    {
      "entropy": 5.997435569763184,
      "epoch": 2.1891112470505987,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045189964764482407,
      "loss": 5.6961,
      "mean_token_accuracy": 0.16533025354146957,
      "num_tokens": 50751394.0,
      "step": 25050
    },
    {
      "entropy": 6.007758569717407,
      "epoch": 2.189548195403303,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004518803900770232,
      "loss": 5.6926,
      "mean_token_accuracy": 0.16579405814409257,
      "num_tokens": 50762038.0,
      "step": 25055
    },
    {
      "entropy": 6.094683265686035,
      "epoch": 2.189985143756008,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00045186112911656387,
      "loss": 5.7744,
      "mean_token_accuracy": 0.16070689707994462,
      "num_tokens": 50774063.0,
      "step": 25060
    },
    {
      "entropy": 6.169772148132324,
      "epoch": 2.1904220921087125,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004518418647638153,
      "loss": 5.7439,
      "mean_token_accuracy": 0.16108395010232926,
      "num_tokens": 50784606.0,
      "step": 25065
    },
    {
      "entropy": 6.019498348236084,
      "epoch": 2.1908590404614174,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004518225970191471,
      "loss": 5.7151,
      "mean_token_accuracy": 0.1593012437224388,
      "num_tokens": 50794892.0,
      "step": 25070
    },
    {
      "entropy": 5.971560907363892,
      "epoch": 2.1912959888141224,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00045180332588292896,
      "loss": 5.7439,
      "mean_token_accuracy": 0.16608887910842896,
      "num_tokens": 50805903.0,
      "step": 25075
    },
    {
      "entropy": 6.041752767562866,
      "epoch": 2.191732937166827,
      "grad_norm": 5.125,
      "learning_rate": 0.0004517840513555303,
      "loss": 5.5801,
      "mean_token_accuracy": 0.17970509976148605,
      "num_tokens": 50815339.0,
      "step": 25080
    },
    {
      "entropy": 6.071961832046509,
      "epoch": 2.1921698855195317,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004517647734373209,
      "loss": 5.7186,
      "mean_token_accuracy": 0.16485579013824464,
      "num_tokens": 50825575.0,
      "step": 25085
    },
    {
      "entropy": 6.062701845169068,
      "epoch": 2.192606833872236,
      "grad_norm": 1.546875,
      "learning_rate": 0.00045174549212867053,
      "loss": 5.7614,
      "mean_token_accuracy": 0.1604332722723484,
      "num_tokens": 50836106.0,
      "step": 25090
    },
    {
      "entropy": 6.022215366363525,
      "epoch": 2.193043782224941,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045172620742994904,
      "loss": 5.7293,
      "mean_token_accuracy": 0.1617071397602558,
      "num_tokens": 50847350.0,
      "step": 25095
    },
    {
      "entropy": 6.040527105331421,
      "epoch": 2.1934807305776456,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004517069193415263,
      "loss": 5.6053,
      "mean_token_accuracy": 0.16948920637369155,
      "num_tokens": 50857543.0,
      "step": 25100
    },
    {
      "entropy": 6.071678924560547,
      "epoch": 2.1939176789303505,
      "grad_norm": 1.421875,
      "learning_rate": 0.00045168762786377214,
      "loss": 5.7281,
      "mean_token_accuracy": 0.159186252951622,
      "num_tokens": 50868791.0,
      "step": 25105
    },
    {
      "entropy": 6.0542069435119625,
      "epoch": 2.194354627283055,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00045166833299705686,
      "loss": 5.687,
      "mean_token_accuracy": 0.16692813783884047,
      "num_tokens": 50879244.0,
      "step": 25110
    },
    {
      "entropy": 6.0828972339630125,
      "epoch": 2.19479157563576,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00045164903474175026,
      "loss": 5.7059,
      "mean_token_accuracy": 0.16075878143310546,
      "num_tokens": 50888935.0,
      "step": 25115
    },
    {
      "entropy": 6.01882586479187,
      "epoch": 2.1952285239884644,
      "grad_norm": 1.46875,
      "learning_rate": 0.00045162973309822264,
      "loss": 5.6757,
      "mean_token_accuracy": 0.16656736433506011,
      "num_tokens": 50898152.0,
      "step": 25120
    },
    {
      "entropy": 5.995517253875732,
      "epoch": 2.1956654723411693,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004516104280668442,
      "loss": 5.7122,
      "mean_token_accuracy": 0.16272360235452651,
      "num_tokens": 50908555.0,
      "step": 25125
    },
    {
      "entropy": 6.057443761825562,
      "epoch": 2.196102420693874,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00045159111964798514,
      "loss": 5.8031,
      "mean_token_accuracy": 0.1566713497042656,
      "num_tokens": 50919078.0,
      "step": 25130
    },
    {
      "entropy": 6.098584365844727,
      "epoch": 2.1965393690465786,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00045157180784201594,
      "loss": 5.6944,
      "mean_token_accuracy": 0.16650781780481339,
      "num_tokens": 50928879.0,
      "step": 25135
    },
    {
      "entropy": 6.086713218688965,
      "epoch": 2.1969763173992836,
      "grad_norm": 1.28125,
      "learning_rate": 0.000451552492649307,
      "loss": 5.7021,
      "mean_token_accuracy": 0.1671939015388489,
      "num_tokens": 50939456.0,
      "step": 25140
    },
    {
      "entropy": 5.987115907669067,
      "epoch": 2.197413265751988,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00045153317407022866,
      "loss": 5.6036,
      "mean_token_accuracy": 0.17275154143571853,
      "num_tokens": 50948789.0,
      "step": 25145
    },
    {
      "entropy": 5.964096450805664,
      "epoch": 2.197850214104693,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00045151385210515153,
      "loss": 5.6582,
      "mean_token_accuracy": 0.1657981365919113,
      "num_tokens": 50958492.0,
      "step": 25150
    },
    {
      "entropy": 5.981822443008423,
      "epoch": 2.1982871624573974,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00045149452675444625,
      "loss": 5.6928,
      "mean_token_accuracy": 0.1706938177347183,
      "num_tokens": 50968814.0,
      "step": 25155
    },
    {
      "entropy": 6.105430269241333,
      "epoch": 2.1987241108101023,
      "grad_norm": 2.140625,
      "learning_rate": 0.00045147519801848346,
      "loss": 5.6994,
      "mean_token_accuracy": 0.16199028491973877,
      "num_tokens": 50979680.0,
      "step": 25160
    },
    {
      "entropy": 6.068871736526489,
      "epoch": 2.199161059162807,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00045145586589763394,
      "loss": 5.729,
      "mean_token_accuracy": 0.16711133420467378,
      "num_tokens": 50990827.0,
      "step": 25165
    },
    {
      "entropy": 6.012142276763916,
      "epoch": 2.1995980075155117,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004514365303922684,
      "loss": 5.6746,
      "mean_token_accuracy": 0.16934641897678376,
      "num_tokens": 51001065.0,
      "step": 25170
    },
    {
      "entropy": 6.024152612686157,
      "epoch": 2.200034955868216,
      "grad_norm": 1.765625,
      "learning_rate": 0.00045141719150275773,
      "loss": 5.6543,
      "mean_token_accuracy": 0.1716211572289467,
      "num_tokens": 51011348.0,
      "step": 25175
    },
    {
      "entropy": 6.06028642654419,
      "epoch": 2.200471904220921,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00045139784922947296,
      "loss": 5.6596,
      "mean_token_accuracy": 0.16454448252916337,
      "num_tokens": 51022033.0,
      "step": 25180
    },
    {
      "entropy": 6.005195379257202,
      "epoch": 2.200908852573626,
      "grad_norm": 1.53125,
      "learning_rate": 0.000451378503572785,
      "loss": 5.7093,
      "mean_token_accuracy": 0.16434979438781738,
      "num_tokens": 51032001.0,
      "step": 25185
    },
    {
      "entropy": 6.006819486618042,
      "epoch": 2.2013458009263305,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00045135915453306487,
      "loss": 5.7352,
      "mean_token_accuracy": 0.16817126274108887,
      "num_tokens": 51041227.0,
      "step": 25190
    },
    {
      "entropy": 5.935329961776733,
      "epoch": 2.2017827492790354,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00045133980211068374,
      "loss": 5.5384,
      "mean_token_accuracy": 0.17906509190797806,
      "num_tokens": 51050642.0,
      "step": 25195
    },
    {
      "entropy": 6.055674695968628,
      "epoch": 2.20221969763174,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004513204463060129,
      "loss": 5.6321,
      "mean_token_accuracy": 0.17372808903455733,
      "num_tokens": 51060646.0,
      "step": 25200
    },
    {
      "entropy": 6.036360549926758,
      "epoch": 2.2026566459844448,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004513010871194234,
      "loss": 5.7586,
      "mean_token_accuracy": 0.16579004973173142,
      "num_tokens": 51071762.0,
      "step": 25205
    },
    {
      "entropy": 6.048735475540161,
      "epoch": 2.2030935943371492,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00045128172455128674,
      "loss": 5.7168,
      "mean_token_accuracy": 0.1740128993988037,
      "num_tokens": 51080802.0,
      "step": 25210
    },
    {
      "entropy": 6.054001808166504,
      "epoch": 2.203530542689854,
      "grad_norm": 1.5,
      "learning_rate": 0.0004512623586019742,
      "loss": 5.7437,
      "mean_token_accuracy": 0.16517787873744966,
      "num_tokens": 51091802.0,
      "step": 25215
    },
    {
      "entropy": 6.063002204895019,
      "epoch": 2.2039674910425586,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004512429892718573,
      "loss": 5.6453,
      "mean_token_accuracy": 0.17575251460075378,
      "num_tokens": 51100939.0,
      "step": 25220
    },
    {
      "entropy": 5.975306463241577,
      "epoch": 2.2044044393952635,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004512236165613074,
      "loss": 5.6661,
      "mean_token_accuracy": 0.1694047674536705,
      "num_tokens": 51110928.0,
      "step": 25225
    },
    {
      "entropy": 6.0360359191894535,
      "epoch": 2.204841387747968,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00045120424047069633,
      "loss": 5.7213,
      "mean_token_accuracy": 0.1631992220878601,
      "num_tokens": 51121126.0,
      "step": 25230
    },
    {
      "entropy": 6.112569761276245,
      "epoch": 2.205278336100673,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045118486100039553,
      "loss": 5.6011,
      "mean_token_accuracy": 0.17504008561372758,
      "num_tokens": 51130425.0,
      "step": 25235
    },
    {
      "entropy": 5.98485951423645,
      "epoch": 2.205715284453378,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004511654781507768,
      "loss": 5.6625,
      "mean_token_accuracy": 0.1666882500052452,
      "num_tokens": 51139994.0,
      "step": 25240
    },
    {
      "entropy": 5.939914655685425,
      "epoch": 2.2061522328060823,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045114609192221196,
      "loss": 5.6978,
      "mean_token_accuracy": 0.1602497950196266,
      "num_tokens": 51149762.0,
      "step": 25245
    },
    {
      "entropy": 5.944825220108032,
      "epoch": 2.206589181158787,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004511267023150727,
      "loss": 5.6775,
      "mean_token_accuracy": 0.17436384558677673,
      "num_tokens": 51161091.0,
      "step": 25250
    },
    {
      "entropy": 6.072478342056274,
      "epoch": 2.2070261295114917,
      "grad_norm": 2.5625,
      "learning_rate": 0.000451107309329731,
      "loss": 5.6614,
      "mean_token_accuracy": 0.16610408276319505,
      "num_tokens": 51170749.0,
      "step": 25255
    },
    {
      "entropy": 6.090936613082886,
      "epoch": 2.2074630778641966,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00045108791296655877,
      "loss": 5.6542,
      "mean_token_accuracy": 0.17079697847366332,
      "num_tokens": 51179515.0,
      "step": 25260
    },
    {
      "entropy": 5.991197204589843,
      "epoch": 2.207900026216901,
      "grad_norm": 1.828125,
      "learning_rate": 0.00045106851322592817,
      "loss": 5.715,
      "mean_token_accuracy": 0.17608828842639923,
      "num_tokens": 51189036.0,
      "step": 25265
    },
    {
      "entropy": 6.059875106811523,
      "epoch": 2.208336974569606,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004510491101082112,
      "loss": 5.7451,
      "mean_token_accuracy": 0.1646776333451271,
      "num_tokens": 51199967.0,
      "step": 25270
    },
    {
      "entropy": 5.980136108398438,
      "epoch": 2.2087739229223105,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004510297036137801,
      "loss": 5.6951,
      "mean_token_accuracy": 0.16966909617185594,
      "num_tokens": 51210281.0,
      "step": 25275
    },
    {
      "entropy": 6.061878633499146,
      "epoch": 2.2092108712750154,
      "grad_norm": 1.578125,
      "learning_rate": 0.00045101029374300703,
      "loss": 5.7594,
      "mean_token_accuracy": 0.16543068885803222,
      "num_tokens": 51220209.0,
      "step": 25280
    },
    {
      "entropy": 6.1077718257904055,
      "epoch": 2.20964781962772,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004509908804962643,
      "loss": 5.704,
      "mean_token_accuracy": 0.16943866312503814,
      "num_tokens": 51231108.0,
      "step": 25285
    },
    {
      "entropy": 5.982352161407471,
      "epoch": 2.2100847679804247,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00045097146387392423,
      "loss": 5.6544,
      "mean_token_accuracy": 0.16786369532346726,
      "num_tokens": 51241316.0,
      "step": 25290
    },
    {
      "entropy": 5.991304779052735,
      "epoch": 2.210521716333129,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004509520438763593,
      "loss": 5.6848,
      "mean_token_accuracy": 0.17275773584842682,
      "num_tokens": 51250854.0,
      "step": 25295
    },
    {
      "entropy": 6.002609539031982,
      "epoch": 2.210958664685834,
      "grad_norm": 1.5,
      "learning_rate": 0.00045093262050394193,
      "loss": 5.6957,
      "mean_token_accuracy": 0.163202004134655,
      "num_tokens": 51261001.0,
      "step": 25300
    },
    {
      "entropy": 6.080059766769409,
      "epoch": 2.211395613038539,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045091319375704475,
      "loss": 5.736,
      "mean_token_accuracy": 0.16778630316257476,
      "num_tokens": 51270815.0,
      "step": 25305
    },
    {
      "entropy": 6.082044267654419,
      "epoch": 2.2118325613912435,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045089376363604037,
      "loss": 5.6808,
      "mean_token_accuracy": 0.1631609931588173,
      "num_tokens": 51281370.0,
      "step": 25310
    },
    {
      "entropy": 6.060933828353882,
      "epoch": 2.2122695097439484,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00045087433014130143,
      "loss": 5.7059,
      "mean_token_accuracy": 0.1764939844608307,
      "num_tokens": 51290147.0,
      "step": 25315
    },
    {
      "entropy": 6.061863279342651,
      "epoch": 2.212706458096653,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004508548932732006,
      "loss": 5.7407,
      "mean_token_accuracy": 0.16514404416084288,
      "num_tokens": 51299487.0,
      "step": 25320
    },
    {
      "entropy": 6.071087980270386,
      "epoch": 2.213143406449358,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004508354530321108,
      "loss": 5.6655,
      "mean_token_accuracy": 0.16912917643785477,
      "num_tokens": 51309425.0,
      "step": 25325
    },
    {
      "entropy": 6.047150278091431,
      "epoch": 2.2135803548020623,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004508160094184049,
      "loss": 5.6897,
      "mean_token_accuracy": 0.16356602013111116,
      "num_tokens": 51319962.0,
      "step": 25330
    },
    {
      "entropy": 6.007628154754639,
      "epoch": 2.214017303154767,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004507965624324558,
      "loss": 5.6922,
      "mean_token_accuracy": 0.16947639286518096,
      "num_tokens": 51330892.0,
      "step": 25335
    },
    {
      "entropy": 6.063543319702148,
      "epoch": 2.2144542515074717,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045077711207463654,
      "loss": 5.7045,
      "mean_token_accuracy": 0.16407223492860795,
      "num_tokens": 51342783.0,
      "step": 25340
    },
    {
      "entropy": 6.148137903213501,
      "epoch": 2.2148911998601766,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00045075765834532024,
      "loss": 5.7265,
      "mean_token_accuracy": 0.1624673843383789,
      "num_tokens": 51352890.0,
      "step": 25345
    },
    {
      "entropy": 6.064050579071045,
      "epoch": 2.215328148212881,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004507382012448798,
      "loss": 5.7516,
      "mean_token_accuracy": 0.1609877735376358,
      "num_tokens": 51363479.0,
      "step": 25350
    },
    {
      "entropy": 5.9974525451660154,
      "epoch": 2.215765096565586,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004507187407736886,
      "loss": 5.6948,
      "mean_token_accuracy": 0.1692547082901001,
      "num_tokens": 51374304.0,
      "step": 25355
    },
    {
      "entropy": 6.052069044113159,
      "epoch": 2.2162020449182904,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004506992769321199,
      "loss": 5.723,
      "mean_token_accuracy": 0.16658305525779724,
      "num_tokens": 51384649.0,
      "step": 25360
    },
    {
      "entropy": 6.035133647918701,
      "epoch": 2.2166389932709953,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00045067980972054707,
      "loss": 5.6621,
      "mean_token_accuracy": 0.16916657835245133,
      "num_tokens": 51394582.0,
      "step": 25365
    },
    {
      "entropy": 6.043423461914062,
      "epoch": 2.2170759416237003,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004506603391393434,
      "loss": 5.6639,
      "mean_token_accuracy": 0.17140417397022248,
      "num_tokens": 51404618.0,
      "step": 25370
    },
    {
      "entropy": 6.031374883651734,
      "epoch": 2.2175128899764047,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004506408651888824,
      "loss": 5.7086,
      "mean_token_accuracy": 0.16704034060239792,
      "num_tokens": 51414810.0,
      "step": 25375
    },
    {
      "entropy": 6.060408687591552,
      "epoch": 2.2179498383291096,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004506213878695375,
      "loss": 5.7669,
      "mean_token_accuracy": 0.167933152616024,
      "num_tokens": 51426118.0,
      "step": 25380
    },
    {
      "entropy": 6.083729887008667,
      "epoch": 2.218386786681814,
      "grad_norm": 1.5625,
      "learning_rate": 0.00045060190718168244,
      "loss": 5.6746,
      "mean_token_accuracy": 0.17228417247533798,
      "num_tokens": 51435260.0,
      "step": 25385
    },
    {
      "entropy": 6.015804195404053,
      "epoch": 2.218823735034519,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004505824231256907,
      "loss": 5.6478,
      "mean_token_accuracy": 0.16879157423973085,
      "num_tokens": 51444778.0,
      "step": 25390
    },
    {
      "entropy": 5.995780038833618,
      "epoch": 2.2192606833872235,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004505629357019361,
      "loss": 5.6997,
      "mean_token_accuracy": 0.1669447422027588,
      "num_tokens": 51456030.0,
      "step": 25395
    },
    {
      "entropy": 6.099411582946777,
      "epoch": 2.2196976317399284,
      "grad_norm": 1.375,
      "learning_rate": 0.0004505434449107925,
      "loss": 5.6794,
      "mean_token_accuracy": 0.16798832565546035,
      "num_tokens": 51465836.0,
      "step": 25400
    },
    {
      "entropy": 5.987078142166138,
      "epoch": 2.220134580092633,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004505239507526335,
      "loss": 5.6175,
      "mean_token_accuracy": 0.17335645109415054,
      "num_tokens": 51475330.0,
      "step": 25405
    },
    {
      "entropy": 5.975121974945068,
      "epoch": 2.220571528445338,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045050445322783326,
      "loss": 5.6332,
      "mean_token_accuracy": 0.16911076307296752,
      "num_tokens": 51486646.0,
      "step": 25410
    },
    {
      "entropy": 6.051708745956421,
      "epoch": 2.2210084767980423,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004504849523367656,
      "loss": 5.655,
      "mean_token_accuracy": 0.17178364992141723,
      "num_tokens": 51496878.0,
      "step": 25415
    },
    {
      "entropy": 6.101105880737305,
      "epoch": 2.221445425150747,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004504654480798046,
      "loss": 5.7598,
      "mean_token_accuracy": 0.1595654755830765,
      "num_tokens": 51507092.0,
      "step": 25420
    },
    {
      "entropy": 5.981357526779175,
      "epoch": 2.221882373503452,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004504459404573243,
      "loss": 5.6878,
      "mean_token_accuracy": 0.16827785819768906,
      "num_tokens": 51517521.0,
      "step": 25425
    },
    {
      "entropy": 6.032043886184693,
      "epoch": 2.2223193218561565,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004504264294696991,
      "loss": 5.7334,
      "mean_token_accuracy": 0.16768995076417922,
      "num_tokens": 51527902.0,
      "step": 25430
    },
    {
      "entropy": 6.072186088562011,
      "epoch": 2.2227562702088615,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004504069151173029,
      "loss": 5.7102,
      "mean_token_accuracy": 0.16436870098114015,
      "num_tokens": 51539062.0,
      "step": 25435
    },
    {
      "entropy": 6.055197763442993,
      "epoch": 2.223193218561566,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004503873974005102,
      "loss": 5.6126,
      "mean_token_accuracy": 0.1815616250038147,
      "num_tokens": 51548938.0,
      "step": 25440
    },
    {
      "entropy": 6.016711568832397,
      "epoch": 2.223630166914271,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004503678763196953,
      "loss": 5.6883,
      "mean_token_accuracy": 0.17212090641260147,
      "num_tokens": 51559330.0,
      "step": 25445
    },
    {
      "entropy": 6.0346979141235355,
      "epoch": 2.2240671152669753,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004503483518752327,
      "loss": 5.7058,
      "mean_token_accuracy": 0.16882974207401275,
      "num_tokens": 51569617.0,
      "step": 25450
    },
    {
      "entropy": 6.026480484008789,
      "epoch": 2.2245040636196802,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004503288240674968,
      "loss": 5.6776,
      "mean_token_accuracy": 0.16675887554883956,
      "num_tokens": 51579545.0,
      "step": 25455
    },
    {
      "entropy": 6.042633438110352,
      "epoch": 2.2249410119723847,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004503092928968622,
      "loss": 5.6616,
      "mean_token_accuracy": 0.16882938295602798,
      "num_tokens": 51589375.0,
      "step": 25460
    },
    {
      "entropy": 6.05024471282959,
      "epoch": 2.2253779603250896,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004502897583637035,
      "loss": 5.6688,
      "mean_token_accuracy": 0.16908148527145386,
      "num_tokens": 51600586.0,
      "step": 25465
    },
    {
      "entropy": 6.026455354690552,
      "epoch": 2.225814908677794,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045027022046839534,
      "loss": 5.6188,
      "mean_token_accuracy": 0.17003423124551773,
      "num_tokens": 51609568.0,
      "step": 25470
    },
    {
      "entropy": 6.018143844604492,
      "epoch": 2.226251857030499,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004502506792113126,
      "loss": 5.6367,
      "mean_token_accuracy": 0.171963869035244,
      "num_tokens": 51619450.0,
      "step": 25475
    },
    {
      "entropy": 6.057275199890137,
      "epoch": 2.226688805383204,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004502311345928299,
      "loss": 5.6516,
      "mean_token_accuracy": 0.165044105052948,
      "num_tokens": 51630528.0,
      "step": 25480
    },
    {
      "entropy": 6.01384859085083,
      "epoch": 2.2271257537359084,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045021158661332227,
      "loss": 5.659,
      "mean_token_accuracy": 0.1739836484193802,
      "num_tokens": 51640204.0,
      "step": 25485
    },
    {
      "entropy": 6.0723662853240965,
      "epoch": 2.2275627020886133,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004501920352731646,
      "loss": 5.8085,
      "mean_token_accuracy": 0.16522168964147568,
      "num_tokens": 51649531.0,
      "step": 25490
    },
    {
      "entropy": 6.132616710662842,
      "epoch": 2.2279996504413178,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004501724805727319,
      "loss": 5.7792,
      "mean_token_accuracy": 0.16206145733594896,
      "num_tokens": 51660943.0,
      "step": 25495
    },
    {
      "entropy": 6.060165739059448,
      "epoch": 2.2284365987940227,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004501529225123992,
      "loss": 5.5912,
      "mean_token_accuracy": 0.17514251470565795,
      "num_tokens": 51670606.0,
      "step": 25500
    },
    {
      "entropy": 6.0010215759277346,
      "epoch": 2.228873547146727,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045013336109254174,
      "loss": 5.6584,
      "mean_token_accuracy": 0.16999367177486419,
      "num_tokens": 51681634.0,
      "step": 25505
    },
    {
      "entropy": 6.054671335220337,
      "epoch": 2.229310495499432,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00045011379631353457,
      "loss": 5.7077,
      "mean_token_accuracy": 0.16343710869550704,
      "num_tokens": 51691579.0,
      "step": 25510
    },
    {
      "entropy": 5.991079711914063,
      "epoch": 2.2297474438521365,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045009422817575304,
      "loss": 5.59,
      "mean_token_accuracy": 0.182220458984375,
      "num_tokens": 51700739.0,
      "step": 25515
    },
    {
      "entropy": 6.010309886932373,
      "epoch": 2.2301843922048414,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004500746566795725,
      "loss": 5.6917,
      "mean_token_accuracy": 0.1671616852283478,
      "num_tokens": 51711905.0,
      "step": 25520
    },
    {
      "entropy": 6.009018993377685,
      "epoch": 2.230621340557546,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045005508182536824,
      "loss": 5.6067,
      "mean_token_accuracy": 0.16917201578617097,
      "num_tokens": 51722020.0,
      "step": 25525
    },
    {
      "entropy": 6.065915060043335,
      "epoch": 2.231058288910251,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004500355036135158,
      "loss": 5.7109,
      "mean_token_accuracy": 0.1691088244318962,
      "num_tokens": 51731763.0,
      "step": 25530
    },
    {
      "entropy": 6.056370401382447,
      "epoch": 2.2314952372629557,
      "grad_norm": 1.921875,
      "learning_rate": 0.00045001592204439077,
      "loss": 5.7395,
      "mean_token_accuracy": 0.15928076431155205,
      "num_tokens": 51741425.0,
      "step": 25535
    },
    {
      "entropy": 6.05377197265625,
      "epoch": 2.23193218561566,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004499963371183686,
      "loss": 5.7232,
      "mean_token_accuracy": 0.16222632378339769,
      "num_tokens": 51751788.0,
      "step": 25540
    },
    {
      "entropy": 6.062930154800415,
      "epoch": 2.232369133968365,
      "grad_norm": 1.59375,
      "learning_rate": 0.00044997674883582494,
      "loss": 5.7429,
      "mean_token_accuracy": 0.16294606477022172,
      "num_tokens": 51763205.0,
      "step": 25545
    },
    {
      "entropy": 6.098311281204223,
      "epoch": 2.2328060823210696,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004499571571971356,
      "loss": 5.6959,
      "mean_token_accuracy": 0.16706916987895964,
      "num_tokens": 51772817.0,
      "step": 25550
    },
    {
      "entropy": 6.0660233974456785,
      "epoch": 2.2332430306737745,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004499375622026763,
      "loss": 5.6876,
      "mean_token_accuracy": 0.16866272538900376,
      "num_tokens": 51781596.0,
      "step": 25555
    },
    {
      "entropy": 6.008811807632446,
      "epoch": 2.233679979026479,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004499179638528229,
      "loss": 5.6569,
      "mean_token_accuracy": 0.16868989765644074,
      "num_tokens": 51792452.0,
      "step": 25560
    },
    {
      "entropy": 6.055217552185058,
      "epoch": 2.234116927379184,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004498983621479513,
      "loss": 5.7207,
      "mean_token_accuracy": 0.16439780294895173,
      "num_tokens": 51802554.0,
      "step": 25565
    },
    {
      "entropy": 6.1026939868927,
      "epoch": 2.2345538757318884,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004498787570884374,
      "loss": 5.7256,
      "mean_token_accuracy": 0.15944695025682448,
      "num_tokens": 51812855.0,
      "step": 25570
    },
    {
      "entropy": 6.064946985244751,
      "epoch": 2.2349908240845933,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004498591486746573,
      "loss": 5.7011,
      "mean_token_accuracy": 0.1693050280213356,
      "num_tokens": 51822306.0,
      "step": 25575
    },
    {
      "entropy": 6.086545372009278,
      "epoch": 2.2354277724372977,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00044983953690698724,
      "loss": 5.7252,
      "mean_token_accuracy": 0.16457351446151733,
      "num_tokens": 51832209.0,
      "step": 25580
    },
    {
      "entropy": 6.077377843856811,
      "epoch": 2.2358647207900026,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004498199217858032,
      "loss": 5.7317,
      "mean_token_accuracy": 0.16200240701436996,
      "num_tokens": 51841794.0,
      "step": 25585
    },
    {
      "entropy": 6.074223232269287,
      "epoch": 2.2363016691427076,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00044980030331148144,
      "loss": 5.7251,
      "mean_token_accuracy": 0.1633213371038437,
      "num_tokens": 51851973.0,
      "step": 25590
    },
    {
      "entropy": 6.112968254089355,
      "epoch": 2.236738617495412,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00044978068148439826,
      "loss": 5.7315,
      "mean_token_accuracy": 0.1617191955447197,
      "num_tokens": 51862173.0,
      "step": 25595
    },
    {
      "entropy": 6.085508537292481,
      "epoch": 2.237175565848117,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00044976105630493,
      "loss": 5.7175,
      "mean_token_accuracy": 0.16126049011945726,
      "num_tokens": 51872960.0,
      "step": 25600
    },
    {
      "entropy": 6.0040946960449215,
      "epoch": 2.2376125142008214,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004497414277734532,
      "loss": 5.6591,
      "mean_token_accuracy": 0.16853218525648117,
      "num_tokens": 51883911.0,
      "step": 25605
    },
    {
      "entropy": 6.036505031585693,
      "epoch": 2.2380494625535263,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00044972179589034407,
      "loss": 5.6713,
      "mean_token_accuracy": 0.16704053580760955,
      "num_tokens": 51893982.0,
      "step": 25610
    },
    {
      "entropy": 6.001794528961182,
      "epoch": 2.238486410906231,
      "grad_norm": 2.078125,
      "learning_rate": 0.00044970216065597947,
      "loss": 5.609,
      "mean_token_accuracy": 0.17350169122219086,
      "num_tokens": 51903808.0,
      "step": 25615
    },
    {
      "entropy": 5.945404005050659,
      "epoch": 2.2389233592589357,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044968252207073584,
      "loss": 5.5988,
      "mean_token_accuracy": 0.17495093643665313,
      "num_tokens": 51912918.0,
      "step": 25620
    },
    {
      "entropy": 6.071046447753906,
      "epoch": 2.23936030761164,
      "grad_norm": 1.890625,
      "learning_rate": 0.00044966288013499,
      "loss": 5.7247,
      "mean_token_accuracy": 0.16881265640258789,
      "num_tokens": 51923894.0,
      "step": 25625
    },
    {
      "entropy": 6.0982647895812985,
      "epoch": 2.239797255964345,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004496432348491185,
      "loss": 5.7155,
      "mean_token_accuracy": 0.1657194897532463,
      "num_tokens": 51934569.0,
      "step": 25630
    },
    {
      "entropy": 6.0672577857971195,
      "epoch": 2.2402342043170496,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004496235862134982,
      "loss": 5.7099,
      "mean_token_accuracy": 0.16214026510715485,
      "num_tokens": 51944678.0,
      "step": 25635
    },
    {
      "entropy": 6.143296146392823,
      "epoch": 2.2406711526697545,
      "grad_norm": 1.546875,
      "learning_rate": 0.00044960393422850616,
      "loss": 5.763,
      "mean_token_accuracy": 0.1610124886035919,
      "num_tokens": 51954785.0,
      "step": 25640
    },
    {
      "entropy": 6.043195295333862,
      "epoch": 2.241108101022459,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004495842788945191,
      "loss": 5.6229,
      "mean_token_accuracy": 0.17109410911798478,
      "num_tokens": 51964427.0,
      "step": 25645
    },
    {
      "entropy": 6.011103963851928,
      "epoch": 2.241545049375164,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00044956462021191406,
      "loss": 5.7971,
      "mean_token_accuracy": 0.15956073850393296,
      "num_tokens": 51974367.0,
      "step": 25650
    },
    {
      "entropy": 6.083911037445068,
      "epoch": 2.2419819977278688,
      "grad_norm": 1.5,
      "learning_rate": 0.00044954495818106814,
      "loss": 5.6996,
      "mean_token_accuracy": 0.16365032345056535,
      "num_tokens": 51984935.0,
      "step": 25655
    },
    {
      "entropy": 6.050069379806518,
      "epoch": 2.2424189460805732,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004495252928023585,
      "loss": 5.6956,
      "mean_token_accuracy": 0.16939800083637238,
      "num_tokens": 51995132.0,
      "step": 25660
    },
    {
      "entropy": 5.9823469638824465,
      "epoch": 2.242855894433278,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004495056240761623,
      "loss": 5.7481,
      "mean_token_accuracy": 0.16732405871152878,
      "num_tokens": 52005828.0,
      "step": 25665
    },
    {
      "entropy": 6.003464794158935,
      "epoch": 2.2432928427859826,
      "grad_norm": 2.1875,
      "learning_rate": 0.00044948595200285677,
      "loss": 5.5491,
      "mean_token_accuracy": 0.1773311734199524,
      "num_tokens": 52015381.0,
      "step": 25670
    },
    {
      "entropy": 6.020023822784424,
      "epoch": 2.2437297911386875,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00044946627658281924,
      "loss": 5.7168,
      "mean_token_accuracy": 0.16394668370485305,
      "num_tokens": 52025699.0,
      "step": 25675
    },
    {
      "entropy": 6.04924054145813,
      "epoch": 2.244166739491392,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004494465978164271,
      "loss": 5.64,
      "mean_token_accuracy": 0.17221324443817138,
      "num_tokens": 52035903.0,
      "step": 25680
    },
    {
      "entropy": 6.043978023529053,
      "epoch": 2.244603687844097,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004494269157040579,
      "loss": 5.6333,
      "mean_token_accuracy": 0.17248565405607225,
      "num_tokens": 52046039.0,
      "step": 25685
    },
    {
      "entropy": 6.08071231842041,
      "epoch": 2.2450406361968014,
      "grad_norm": 1.578125,
      "learning_rate": 0.00044940723024608893,
      "loss": 5.7335,
      "mean_token_accuracy": 0.16030873954296113,
      "num_tokens": 52057146.0,
      "step": 25690
    },
    {
      "entropy": 6.036278915405274,
      "epoch": 2.2454775845495063,
      "grad_norm": 2.265625,
      "learning_rate": 0.00044938754144289793,
      "loss": 5.6748,
      "mean_token_accuracy": 0.1688736841082573,
      "num_tokens": 52067355.0,
      "step": 25695
    },
    {
      "entropy": 6.052249765396118,
      "epoch": 2.2459145329022108,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00044936784929486257,
      "loss": 5.6647,
      "mean_token_accuracy": 0.17127774208784102,
      "num_tokens": 52077062.0,
      "step": 25700
    },
    {
      "entropy": 6.068854331970215,
      "epoch": 2.2463514812549157,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004493481538023604,
      "loss": 5.644,
      "mean_token_accuracy": 0.17344126105308533,
      "num_tokens": 52087918.0,
      "step": 25705
    },
    {
      "entropy": 5.98949236869812,
      "epoch": 2.24678842960762,
      "grad_norm": 1.421875,
      "learning_rate": 0.00044932845496576936,
      "loss": 5.6794,
      "mean_token_accuracy": 0.16916493922472,
      "num_tokens": 52097283.0,
      "step": 25710
    },
    {
      "entropy": 6.032104444503784,
      "epoch": 2.247225377960325,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004493087527854672,
      "loss": 5.7364,
      "mean_token_accuracy": 0.1658173069357872,
      "num_tokens": 52109037.0,
      "step": 25715
    },
    {
      "entropy": 6.098556423187256,
      "epoch": 2.24766232631303,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004492890472618318,
      "loss": 5.7012,
      "mean_token_accuracy": 0.16927562952041625,
      "num_tokens": 52119809.0,
      "step": 25720
    },
    {
      "entropy": 6.041884088516236,
      "epoch": 2.2480992746657344,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044926933839524116,
      "loss": 5.6047,
      "mean_token_accuracy": 0.17049603760242463,
      "num_tokens": 52129641.0,
      "step": 25725
    },
    {
      "entropy": 5.9658904552459715,
      "epoch": 2.2485362230184394,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00044924962618607324,
      "loss": 5.6055,
      "mean_token_accuracy": 0.17901480048894883,
      "num_tokens": 52139701.0,
      "step": 25730
    },
    {
      "entropy": 5.985057592391968,
      "epoch": 2.248973171371144,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004492299106347062,
      "loss": 5.6165,
      "mean_token_accuracy": 0.1779615193605423,
      "num_tokens": 52148402.0,
      "step": 25735
    },
    {
      "entropy": 5.986203002929687,
      "epoch": 2.2494101197238487,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004492101917415183,
      "loss": 5.6759,
      "mean_token_accuracy": 0.1604973241686821,
      "num_tokens": 52159238.0,
      "step": 25740
    },
    {
      "entropy": 6.077200222015381,
      "epoch": 2.249847068076553,
      "grad_norm": 2.46875,
      "learning_rate": 0.00044919046950688744,
      "loss": 5.6346,
      "mean_token_accuracy": 0.17130900770425797,
      "num_tokens": 52169172.0,
      "step": 25745
    },
    {
      "entropy": 6.043716049194336,
      "epoch": 2.250284016429258,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00044917074393119224,
      "loss": 5.6281,
      "mean_token_accuracy": 0.17451539784669876,
      "num_tokens": 52179432.0,
      "step": 25750
    },
    {
      "entropy": 6.032696294784546,
      "epoch": 2.2507209647819626,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00044915101501481086,
      "loss": 5.6995,
      "mean_token_accuracy": 0.1682278633117676,
      "num_tokens": 52190939.0,
      "step": 25755
    },
    {
      "entropy": 5.961972379684449,
      "epoch": 2.2511579131346675,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00044913128275812184,
      "loss": 5.6364,
      "mean_token_accuracy": 0.17143068462610245,
      "num_tokens": 52201644.0,
      "step": 25760
    },
    {
      "entropy": 6.021077632904053,
      "epoch": 2.251594861487372,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004491115471615035,
      "loss": 5.6499,
      "mean_token_accuracy": 0.16912802159786225,
      "num_tokens": 52211353.0,
      "step": 25765
    },
    {
      "entropy": 6.057596683502197,
      "epoch": 2.252031809840077,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004490918082253344,
      "loss": 5.7117,
      "mean_token_accuracy": 0.1622875839471817,
      "num_tokens": 52221717.0,
      "step": 25770
    },
    {
      "entropy": 5.996135234832764,
      "epoch": 2.252468758192782,
      "grad_norm": 1.765625,
      "learning_rate": 0.00044907206594999325,
      "loss": 5.7153,
      "mean_token_accuracy": 0.16495542973279953,
      "num_tokens": 52232419.0,
      "step": 25775
    },
    {
      "entropy": 6.076769065856934,
      "epoch": 2.2529057065454863,
      "grad_norm": 2.109375,
      "learning_rate": 0.00044905232033585867,
      "loss": 5.7091,
      "mean_token_accuracy": 0.16627279222011565,
      "num_tokens": 52242386.0,
      "step": 25780
    },
    {
      "entropy": 6.074491930007935,
      "epoch": 2.253342654898191,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004490325713833094,
      "loss": 5.6947,
      "mean_token_accuracy": 0.16313983798027037,
      "num_tokens": 52252569.0,
      "step": 25785
    },
    {
      "entropy": 5.964844608306885,
      "epoch": 2.2537796032508957,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004490128190927242,
      "loss": 5.5803,
      "mean_token_accuracy": 0.18350290060043334,
      "num_tokens": 52263101.0,
      "step": 25790
    },
    {
      "entropy": 6.012781095504761,
      "epoch": 2.2542165516036006,
      "grad_norm": 1.65625,
      "learning_rate": 0.00044899306346448195,
      "loss": 5.6573,
      "mean_token_accuracy": 0.17014070153236388,
      "num_tokens": 52273250.0,
      "step": 25795
    },
    {
      "entropy": 6.04005446434021,
      "epoch": 2.254653499956305,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004489733044989616,
      "loss": 5.7526,
      "mean_token_accuracy": 0.1623348668217659,
      "num_tokens": 52284351.0,
      "step": 25800
    },
    {
      "entropy": 5.986283254623413,
      "epoch": 2.25509044830901,
      "grad_norm": 2.0625,
      "learning_rate": 0.00044895354219654204,
      "loss": 5.6497,
      "mean_token_accuracy": 0.1738375872373581,
      "num_tokens": 52294642.0,
      "step": 25805
    },
    {
      "entropy": 5.9163459777832035,
      "epoch": 2.2555273966617144,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00044893377655760245,
      "loss": 5.5702,
      "mean_token_accuracy": 0.17584152817726134,
      "num_tokens": 52303793.0,
      "step": 25810
    },
    {
      "entropy": 5.922292613983155,
      "epoch": 2.2559643450144193,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00044891400758252187,
      "loss": 5.6811,
      "mean_token_accuracy": 0.16140032708644866,
      "num_tokens": 52313541.0,
      "step": 25815
    },
    {
      "entropy": 6.088237571716308,
      "epoch": 2.256401293367124,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004488942352716795,
      "loss": 5.7105,
      "mean_token_accuracy": 0.16384453177452088,
      "num_tokens": 52325428.0,
      "step": 25820
    },
    {
      "entropy": 6.0403971672058105,
      "epoch": 2.2568382417198287,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004488744596254545,
      "loss": 5.6983,
      "mean_token_accuracy": 0.17295930087566375,
      "num_tokens": 52335837.0,
      "step": 25825
    },
    {
      "entropy": 5.956763839721679,
      "epoch": 2.2572751900725336,
      "grad_norm": 2.125,
      "learning_rate": 0.00044885468064422636,
      "loss": 5.62,
      "mean_token_accuracy": 0.16725492030382155,
      "num_tokens": 52346925.0,
      "step": 25830
    },
    {
      "entropy": 6.02131757736206,
      "epoch": 2.257712138425238,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004488348983283743,
      "loss": 5.7903,
      "mean_token_accuracy": 0.1608935922384262,
      "num_tokens": 52357244.0,
      "step": 25835
    },
    {
      "entropy": 5.995322799682617,
      "epoch": 2.258149086777943,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004488151126782779,
      "loss": 5.6375,
      "mean_token_accuracy": 0.17141054719686508,
      "num_tokens": 52367657.0,
      "step": 25840
    },
    {
      "entropy": 6.06936616897583,
      "epoch": 2.2585860351306475,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00044879532369431635,
      "loss": 5.6908,
      "mean_token_accuracy": 0.16488672494888307,
      "num_tokens": 52377298.0,
      "step": 25845
    },
    {
      "entropy": 6.071857929229736,
      "epoch": 2.2590229834833524,
      "grad_norm": 1.640625,
      "learning_rate": 0.00044877553137686953,
      "loss": 5.7348,
      "mean_token_accuracy": 0.16874280869960784,
      "num_tokens": 52387829.0,
      "step": 25850
    },
    {
      "entropy": 6.050209283828735,
      "epoch": 2.259459931836057,
      "grad_norm": 2.921875,
      "learning_rate": 0.00044875573572631706,
      "loss": 5.6818,
      "mean_token_accuracy": 0.1699944406747818,
      "num_tokens": 52398214.0,
      "step": 25855
    },
    {
      "entropy": 5.975145292282105,
      "epoch": 2.259896880188762,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004487359367430384,
      "loss": 5.6367,
      "mean_token_accuracy": 0.17238867729902269,
      "num_tokens": 52407619.0,
      "step": 25860
    },
    {
      "entropy": 5.985115432739258,
      "epoch": 2.2603338285414663,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004487161344274135,
      "loss": 5.6841,
      "mean_token_accuracy": 0.16764370501041412,
      "num_tokens": 52417465.0,
      "step": 25865
    },
    {
      "entropy": 6.017056894302368,
      "epoch": 2.260770776894171,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004486963287798221,
      "loss": 5.6447,
      "mean_token_accuracy": 0.17618416100740433,
      "num_tokens": 52427000.0,
      "step": 25870
    },
    {
      "entropy": 6.045793533325195,
      "epoch": 2.2612077252468756,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004486765198006441,
      "loss": 5.7,
      "mean_token_accuracy": 0.16142588406801223,
      "num_tokens": 52437449.0,
      "step": 25875
    },
    {
      "entropy": 6.102050113677978,
      "epoch": 2.2616446735995805,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00044865670749025943,
      "loss": 5.7021,
      "mean_token_accuracy": 0.16541186720132828,
      "num_tokens": 52446946.0,
      "step": 25880
    },
    {
      "entropy": 5.999821281433105,
      "epoch": 2.2620816219522855,
      "grad_norm": 1.5,
      "learning_rate": 0.00044863689184904815,
      "loss": 5.6637,
      "mean_token_accuracy": 0.17147019654512405,
      "num_tokens": 52456993.0,
      "step": 25885
    },
    {
      "entropy": 6.060312986373901,
      "epoch": 2.26251857030499,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004486170728773903,
      "loss": 5.7601,
      "mean_token_accuracy": 0.1547287881374359,
      "num_tokens": 52467999.0,
      "step": 25890
    },
    {
      "entropy": 6.13176851272583,
      "epoch": 2.262955518657695,
      "grad_norm": 2.25,
      "learning_rate": 0.000448597250575666,
      "loss": 5.742,
      "mean_token_accuracy": 0.1604446455836296,
      "num_tokens": 52479556.0,
      "step": 25895
    },
    {
      "entropy": 6.076519441604614,
      "epoch": 2.2633924670103993,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004485774249442555,
      "loss": 5.6873,
      "mean_token_accuracy": 0.16639498770236968,
      "num_tokens": 52489160.0,
      "step": 25900
    },
    {
      "entropy": 5.944971132278442,
      "epoch": 2.2638294153631042,
      "grad_norm": 2.25,
      "learning_rate": 0.000448557595983539,
      "loss": 5.6287,
      "mean_token_accuracy": 0.1663893476128578,
      "num_tokens": 52500102.0,
      "step": 25905
    },
    {
      "entropy": 5.985164642333984,
      "epoch": 2.2642663637158087,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044853776369389685,
      "loss": 5.6504,
      "mean_token_accuracy": 0.1694265842437744,
      "num_tokens": 52509615.0,
      "step": 25910
    },
    {
      "entropy": 5.999120378494263,
      "epoch": 2.2647033120685136,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004485179280757095,
      "loss": 5.6662,
      "mean_token_accuracy": 0.17407902479171752,
      "num_tokens": 52520796.0,
      "step": 25915
    },
    {
      "entropy": 6.018314599990845,
      "epoch": 2.265140260421218,
      "grad_norm": 1.484375,
      "learning_rate": 0.00044849808912935735,
      "loss": 5.6218,
      "mean_token_accuracy": 0.17145248204469682,
      "num_tokens": 52530641.0,
      "step": 25920
    },
    {
      "entropy": 6.02027907371521,
      "epoch": 2.265577208773923,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000448478246855221,
      "loss": 5.6951,
      "mean_token_accuracy": 0.16684998720884323,
      "num_tokens": 52541626.0,
      "step": 25925
    },
    {
      "entropy": 6.017548942565918,
      "epoch": 2.2660141571266275,
      "grad_norm": 1.875,
      "learning_rate": 0.0004484584012536809,
      "loss": 5.7195,
      "mean_token_accuracy": 0.1646649122238159,
      "num_tokens": 52552301.0,
      "step": 25930
    },
    {
      "entropy": 5.948980236053467,
      "epoch": 2.2664511054793324,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004484385523251177,
      "loss": 5.584,
      "mean_token_accuracy": 0.1879386603832245,
      "num_tokens": 52562463.0,
      "step": 25935
    },
    {
      "entropy": 6.029582929611206,
      "epoch": 2.2668880538320373,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004484187000699123,
      "loss": 5.6929,
      "mean_token_accuracy": 0.16731839776039123,
      "num_tokens": 52572782.0,
      "step": 25940
    },
    {
      "entropy": 5.970148515701294,
      "epoch": 2.2673250021847418,
      "grad_norm": 1.359375,
      "learning_rate": 0.00044839884448844534,
      "loss": 5.6167,
      "mean_token_accuracy": 0.17592423409223557,
      "num_tokens": 52582293.0,
      "step": 25945
    },
    {
      "entropy": 6.026807498931885,
      "epoch": 2.2677619505374462,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004483789855810977,
      "loss": 5.6658,
      "mean_token_accuracy": 0.1654139891266823,
      "num_tokens": 52591950.0,
      "step": 25950
    },
    {
      "entropy": 6.071843051910401,
      "epoch": 2.268198898890151,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044835912334825024,
      "loss": 5.7021,
      "mean_token_accuracy": 0.16768461391329764,
      "num_tokens": 52603998.0,
      "step": 25955
    },
    {
      "entropy": 6.090425062179565,
      "epoch": 2.268635847242856,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044833925779028393,
      "loss": 5.704,
      "mean_token_accuracy": 0.16532388776540757,
      "num_tokens": 52614955.0,
      "step": 25960
    },
    {
      "entropy": 6.091980457305908,
      "epoch": 2.2690727955955605,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004483193889075799,
      "loss": 5.7097,
      "mean_token_accuracy": 0.16755841821432113,
      "num_tokens": 52625305.0,
      "step": 25965
    },
    {
      "entropy": 5.9413553237915036,
      "epoch": 2.2695097439482654,
      "grad_norm": 1.375,
      "learning_rate": 0.0004482995167005191,
      "loss": 5.6053,
      "mean_token_accuracy": 0.17248796969652175,
      "num_tokens": 52635310.0,
      "step": 25970
    },
    {
      "entropy": 6.000210285186768,
      "epoch": 2.26994669230097,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044827964116948284,
      "loss": 5.6731,
      "mean_token_accuracy": 0.16575170308351517,
      "num_tokens": 52644946.0,
      "step": 25975
    },
    {
      "entropy": 6.057907390594482,
      "epoch": 2.270383640653675,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004482597623148522,
      "loss": 5.666,
      "mean_token_accuracy": 0.169821298122406,
      "num_tokens": 52655056.0,
      "step": 25980
    },
    {
      "entropy": 5.9914774894714355,
      "epoch": 2.2708205890063793,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004482398801370085,
      "loss": 5.6911,
      "mean_token_accuracy": 0.1664822742342949,
      "num_tokens": 52665423.0,
      "step": 25985
    },
    {
      "entropy": 6.061847066879272,
      "epoch": 2.271257537359084,
      "grad_norm": 1.8125,
      "learning_rate": 0.00044821999463633323,
      "loss": 5.712,
      "mean_token_accuracy": 0.17092252224683763,
      "num_tokens": 52675300.0,
      "step": 25990
    },
    {
      "entropy": 6.089271497726441,
      "epoch": 2.271694485711789,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044820010581320764,
      "loss": 5.6656,
      "mean_token_accuracy": 0.1649615705013275,
      "num_tokens": 52685431.0,
      "step": 25995
    },
    {
      "entropy": 6.027157831192016,
      "epoch": 2.2721314340644936,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00044818021366801325,
      "loss": 5.7606,
      "mean_token_accuracy": 0.16683038473129272,
      "num_tokens": 52695756.0,
      "step": 26000
    },
    {
      "entropy": 6.003421115875244,
      "epoch": 2.272568382417198,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004481603182011317,
      "loss": 5.663,
      "mean_token_accuracy": 0.16740321218967438,
      "num_tokens": 52704889.0,
      "step": 26005
    },
    {
      "entropy": 5.991657066345215,
      "epoch": 2.273005330769903,
      "grad_norm": 1.734375,
      "learning_rate": 0.00044814041941294435,
      "loss": 5.5959,
      "mean_token_accuracy": 0.17367976903915405,
      "num_tokens": 52714227.0,
      "step": 26010
    },
    {
      "entropy": 5.964133310317993,
      "epoch": 2.273442279122608,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00044812051730383316,
      "loss": 5.6026,
      "mean_token_accuracy": 0.1781155914068222,
      "num_tokens": 52725150.0,
      "step": 26015
    },
    {
      "entropy": 6.021527051925659,
      "epoch": 2.2738792274753123,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004481006118741798,
      "loss": 5.6385,
      "mean_token_accuracy": 0.16842408031225203,
      "num_tokens": 52734825.0,
      "step": 26020
    },
    {
      "entropy": 6.008208799362182,
      "epoch": 2.2743161758280173,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004480807031243658,
      "loss": 5.619,
      "mean_token_accuracy": 0.17783764451742173,
      "num_tokens": 52745185.0,
      "step": 26025
    },
    {
      "entropy": 6.005960702896118,
      "epoch": 2.2747531241807217,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00044806079105477335,
      "loss": 5.6913,
      "mean_token_accuracy": 0.16678230315446854,
      "num_tokens": 52755122.0,
      "step": 26030
    },
    {
      "entropy": 6.02956748008728,
      "epoch": 2.2751900725334266,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004480408756657842,
      "loss": 5.6724,
      "mean_token_accuracy": 0.1663777694106102,
      "num_tokens": 52765338.0,
      "step": 26035
    },
    {
      "entropy": 6.147032356262207,
      "epoch": 2.275627020886131,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00044802095695778047,
      "loss": 5.7322,
      "mean_token_accuracy": 0.16271887570619584,
      "num_tokens": 52775045.0,
      "step": 26040
    },
    {
      "entropy": 6.020278644561768,
      "epoch": 2.276063969238836,
      "grad_norm": 1.5,
      "learning_rate": 0.00044800103493114403,
      "loss": 5.6858,
      "mean_token_accuracy": 0.1706191271543503,
      "num_tokens": 52785942.0,
      "step": 26045
    },
    {
      "entropy": 5.938012886047363,
      "epoch": 2.2765009175915405,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004479811095862571,
      "loss": 5.6186,
      "mean_token_accuracy": 0.1803907185792923,
      "num_tokens": 52794905.0,
      "step": 26050
    },
    {
      "entropy": 6.049458789825439,
      "epoch": 2.2769378659442454,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004479611809235019,
      "loss": 5.578,
      "mean_token_accuracy": 0.1756748378276825,
      "num_tokens": 52804091.0,
      "step": 26055
    },
    {
      "entropy": 6.0506744384765625,
      "epoch": 2.27737481429695,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004479412489432605,
      "loss": 5.7112,
      "mean_token_accuracy": 0.16209357380867004,
      "num_tokens": 52814400.0,
      "step": 26060
    },
    {
      "entropy": 6.022920703887939,
      "epoch": 2.277811762649655,
      "grad_norm": 1.453125,
      "learning_rate": 0.00044792131364591544,
      "loss": 5.6966,
      "mean_token_accuracy": 0.16864215433597565,
      "num_tokens": 52824423.0,
      "step": 26065
    },
    {
      "entropy": 5.991725587844849,
      "epoch": 2.2782487110023597,
      "grad_norm": 1.515625,
      "learning_rate": 0.00044790137503184895,
      "loss": 5.6926,
      "mean_token_accuracy": 0.1692475914955139,
      "num_tokens": 52833852.0,
      "step": 26070
    },
    {
      "entropy": 6.031326770782471,
      "epoch": 2.278685659355064,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004478814331014435,
      "loss": 5.6934,
      "mean_token_accuracy": 0.1625077947974205,
      "num_tokens": 52844759.0,
      "step": 26075
    },
    {
      "entropy": 6.03802375793457,
      "epoch": 2.279122607707769,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004478614878550815,
      "loss": 5.64,
      "mean_token_accuracy": 0.1685519829392433,
      "num_tokens": 52854302.0,
      "step": 26080
    },
    {
      "entropy": 6.012339115142822,
      "epoch": 2.2795595560604736,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004478415392931457,
      "loss": 5.6641,
      "mean_token_accuracy": 0.16856910288333893,
      "num_tokens": 52864086.0,
      "step": 26085
    },
    {
      "entropy": 6.0675920963287355,
      "epoch": 2.2799965044131785,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044782158741601854,
      "loss": 5.5901,
      "mean_token_accuracy": 0.16977194100618362,
      "num_tokens": 52874894.0,
      "step": 26090
    },
    {
      "entropy": 5.957976007461548,
      "epoch": 2.280433452765883,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044780163222408277,
      "loss": 5.6563,
      "mean_token_accuracy": 0.16897083967924117,
      "num_tokens": 52885418.0,
      "step": 26095
    },
    {
      "entropy": 6.024492788314819,
      "epoch": 2.280870401118588,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00044778167371772117,
      "loss": 5.7093,
      "mean_token_accuracy": 0.16981846541166307,
      "num_tokens": 52895515.0,
      "step": 26100
    },
    {
      "entropy": 6.018922996520996,
      "epoch": 2.2813073494712923,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004477617118973165,
      "loss": 5.6749,
      "mean_token_accuracy": 0.1798751100897789,
      "num_tokens": 52905599.0,
      "step": 26105
    },
    {
      "entropy": 6.069145441055298,
      "epoch": 2.2817442978239972,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004477417467632517,
      "loss": 5.7727,
      "mean_token_accuracy": 0.1623073026537895,
      "num_tokens": 52916599.0,
      "step": 26110
    },
    {
      "entropy": 6.071946239471435,
      "epoch": 2.2821812461767017,
      "grad_norm": 1.71875,
      "learning_rate": 0.00044772177831590965,
      "loss": 5.6716,
      "mean_token_accuracy": 0.17863062024116516,
      "num_tokens": 52928131.0,
      "step": 26115
    },
    {
      "entropy": 6.049965333938599,
      "epoch": 2.2826181945294066,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00044770180655567343,
      "loss": 5.704,
      "mean_token_accuracy": 0.16770943850278855,
      "num_tokens": 52939305.0,
      "step": 26120
    },
    {
      "entropy": 5.987267112731933,
      "epoch": 2.2830551428821115,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004476818314829261,
      "loss": 5.5652,
      "mean_token_accuracy": 0.1732360303401947,
      "num_tokens": 52948814.0,
      "step": 26125
    },
    {
      "entropy": 6.0140125274658205,
      "epoch": 2.283492091234816,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004476618530980506,
      "loss": 5.7197,
      "mean_token_accuracy": 0.1660978004336357,
      "num_tokens": 52958964.0,
      "step": 26130
    },
    {
      "entropy": 6.0522401332855225,
      "epoch": 2.283929039587521,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004476418714014305,
      "loss": 5.7427,
      "mean_token_accuracy": 0.16099512130022048,
      "num_tokens": 52969837.0,
      "step": 26135
    },
    {
      "entropy": 6.058587265014649,
      "epoch": 2.2843659879402254,
      "grad_norm": 1.90625,
      "learning_rate": 0.00044762188639344875,
      "loss": 5.6135,
      "mean_token_accuracy": 0.16995978802442552,
      "num_tokens": 52979184.0,
      "step": 26140
    },
    {
      "entropy": 6.000274515151977,
      "epoch": 2.2848029362929303,
      "grad_norm": 1.765625,
      "learning_rate": 0.00044760189807448874,
      "loss": 5.7881,
      "mean_token_accuracy": 0.16198060512542725,
      "num_tokens": 52989867.0,
      "step": 26145
    },
    {
      "entropy": 6.062900257110596,
      "epoch": 2.2852398846456348,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004475819064449338,
      "loss": 5.6966,
      "mean_token_accuracy": 0.16796880066394806,
      "num_tokens": 53000177.0,
      "step": 26150
    },
    {
      "entropy": 6.0318684577941895,
      "epoch": 2.2856768329983397,
      "grad_norm": 1.421875,
      "learning_rate": 0.00044756191150516764,
      "loss": 5.6018,
      "mean_token_accuracy": 0.16413575261831284,
      "num_tokens": 53009372.0,
      "step": 26155
    },
    {
      "entropy": 6.023671817779541,
      "epoch": 2.286113781351044,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004475419132555735,
      "loss": 5.6755,
      "mean_token_accuracy": 0.16621800661087036,
      "num_tokens": 53020405.0,
      "step": 26160
    },
    {
      "entropy": 6.056244850158691,
      "epoch": 2.286550729703749,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00044752191169653504,
      "loss": 5.702,
      "mean_token_accuracy": 0.16796981692314147,
      "num_tokens": 53030713.0,
      "step": 26165
    },
    {
      "entropy": 5.97939395904541,
      "epoch": 2.2869876780564535,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00044750190682843596,
      "loss": 5.6648,
      "mean_token_accuracy": 0.17095537781715392,
      "num_tokens": 53040868.0,
      "step": 26170
    },
    {
      "entropy": 5.959879302978516,
      "epoch": 2.2874246264091584,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004474818986516599,
      "loss": 5.5224,
      "mean_token_accuracy": 0.17309182286262512,
      "num_tokens": 53050349.0,
      "step": 26175
    },
    {
      "entropy": 6.004105949401856,
      "epoch": 2.2878615747618634,
      "grad_norm": 2.25,
      "learning_rate": 0.00044746188716659064,
      "loss": 5.6355,
      "mean_token_accuracy": 0.17130592167377473,
      "num_tokens": 53060446.0,
      "step": 26180
    },
    {
      "entropy": 6.034496831893921,
      "epoch": 2.288298523114568,
      "grad_norm": 2.5,
      "learning_rate": 0.00044744187237361207,
      "loss": 5.701,
      "mean_token_accuracy": 0.16754332929849625,
      "num_tokens": 53071099.0,
      "step": 26185
    },
    {
      "entropy": 6.00064606666565,
      "epoch": 2.2887354714672727,
      "grad_norm": 2.078125,
      "learning_rate": 0.00044742185427310797,
      "loss": 5.6028,
      "mean_token_accuracy": 0.17527889013290404,
      "num_tokens": 53080712.0,
      "step": 26190
    },
    {
      "entropy": 6.066651630401611,
      "epoch": 2.289172419819977,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004474018328654623,
      "loss": 5.7396,
      "mean_token_accuracy": 0.1623690813779831,
      "num_tokens": 53091160.0,
      "step": 26195
    },
    {
      "entropy": 5.953252220153809,
      "epoch": 2.289609368172682,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004473818081510592,
      "loss": 5.6875,
      "mean_token_accuracy": 0.16833970844745635,
      "num_tokens": 53100767.0,
      "step": 26200
    },
    {
      "entropy": 6.045944404602051,
      "epoch": 2.2900463165253866,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00044736178013028276,
      "loss": 5.6831,
      "mean_token_accuracy": 0.17366621345281602,
      "num_tokens": 53111641.0,
      "step": 26205
    },
    {
      "entropy": 6.010534811019897,
      "epoch": 2.2904832648780915,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000447341748803517,
      "loss": 5.6426,
      "mean_token_accuracy": 0.17164625227451324,
      "num_tokens": 53121982.0,
      "step": 26210
    },
    {
      "entropy": 6.070220565795898,
      "epoch": 2.290920213230796,
      "grad_norm": 1.90625,
      "learning_rate": 0.00044732171417114614,
      "loss": 5.7109,
      "mean_token_accuracy": 0.16599299609661103,
      "num_tokens": 53132649.0,
      "step": 26215
    },
    {
      "entropy": 6.0048792362213135,
      "epoch": 2.291357161583501,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004473016762335546,
      "loss": 5.717,
      "mean_token_accuracy": 0.1718108519911766,
      "num_tokens": 53142249.0,
      "step": 26220
    },
    {
      "entropy": 6.033442497253418,
      "epoch": 2.2917941099362054,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004472816349911266,
      "loss": 5.6415,
      "mean_token_accuracy": 0.17589010596275328,
      "num_tokens": 53152375.0,
      "step": 26225
    },
    {
      "entropy": 6.0494975566864015,
      "epoch": 2.2922310582889103,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004472615904442465,
      "loss": 5.7176,
      "mean_token_accuracy": 0.16441831439733506,
      "num_tokens": 53162699.0,
      "step": 26230
    },
    {
      "entropy": 6.045289945602417,
      "epoch": 2.292668006641615,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00044724154259329893,
      "loss": 5.6997,
      "mean_token_accuracy": 0.16349086612462999,
      "num_tokens": 53172890.0,
      "step": 26235
    },
    {
      "entropy": 6.077430629730225,
      "epoch": 2.2931049549943197,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004472214914386682,
      "loss": 5.7278,
      "mean_token_accuracy": 0.16510384529829025,
      "num_tokens": 53183735.0,
      "step": 26240
    },
    {
      "entropy": 6.079256200790406,
      "epoch": 2.2935419033470246,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004472014369807391,
      "loss": 5.6762,
      "mean_token_accuracy": 0.16632324159145356,
      "num_tokens": 53193046.0,
      "step": 26245
    },
    {
      "entropy": 5.981498432159424,
      "epoch": 2.293978851699729,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004471813792198962,
      "loss": 5.6352,
      "mean_token_accuracy": 0.17556983530521392,
      "num_tokens": 53203367.0,
      "step": 26250
    },
    {
      "entropy": 6.083117151260376,
      "epoch": 2.294415800052434,
      "grad_norm": 1.46875,
      "learning_rate": 0.00044716131815652425,
      "loss": 5.8032,
      "mean_token_accuracy": 0.1631105661392212,
      "num_tokens": 53214207.0,
      "step": 26255
    },
    {
      "entropy": 6.077807474136352,
      "epoch": 2.2948527484051384,
      "grad_norm": 1.890625,
      "learning_rate": 0.000447141253791008,
      "loss": 5.6828,
      "mean_token_accuracy": 0.1662491664290428,
      "num_tokens": 53225061.0,
      "step": 26260
    },
    {
      "entropy": 6.04677038192749,
      "epoch": 2.2952896967578433,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00044712118612373223,
      "loss": 5.7052,
      "mean_token_accuracy": 0.1648727685213089,
      "num_tokens": 53235463.0,
      "step": 26265
    },
    {
      "entropy": 6.071451568603516,
      "epoch": 2.295726645110548,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000447101115155082,
      "loss": 5.7364,
      "mean_token_accuracy": 0.16262757927179336,
      "num_tokens": 53245923.0,
      "step": 26270
    },
    {
      "entropy": 6.042762660980225,
      "epoch": 2.2961635934632527,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00044708104088544216,
      "loss": 5.6491,
      "mean_token_accuracy": 0.16801637709140776,
      "num_tokens": 53255756.0,
      "step": 26275
    },
    {
      "entropy": 5.986064195632935,
      "epoch": 2.296600541815957,
      "grad_norm": 1.625,
      "learning_rate": 0.0004470609633151978,
      "loss": 5.6746,
      "mean_token_accuracy": 0.16648093611001968,
      "num_tokens": 53266176.0,
      "step": 26280
    },
    {
      "entropy": 6.001582622528076,
      "epoch": 2.297037490168662,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044704088244473404,
      "loss": 5.6592,
      "mean_token_accuracy": 0.17046640068292618,
      "num_tokens": 53276703.0,
      "step": 26285
    },
    {
      "entropy": 6.0965033054351805,
      "epoch": 2.297474438521367,
      "grad_norm": 1.53125,
      "learning_rate": 0.000447020798274436,
      "loss": 5.7666,
      "mean_token_accuracy": 0.16216451823711395,
      "num_tokens": 53286511.0,
      "step": 26290
    },
    {
      "entropy": 6.09011435508728,
      "epoch": 2.2979113868740715,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00044700071080468884,
      "loss": 5.6798,
      "mean_token_accuracy": 0.16895153224468232,
      "num_tokens": 53297261.0,
      "step": 26295
    },
    {
      "entropy": 6.053393030166626,
      "epoch": 2.298348335226776,
      "grad_norm": 1.640625,
      "learning_rate": 0.000446980620035878,
      "loss": 5.6666,
      "mean_token_accuracy": 0.16675179600715637,
      "num_tokens": 53307054.0,
      "step": 26300
    },
    {
      "entropy": 6.076235246658325,
      "epoch": 2.298785283579481,
      "grad_norm": 1.59375,
      "learning_rate": 0.00044696052596838873,
      "loss": 5.6717,
      "mean_token_accuracy": 0.16977445632219315,
      "num_tokens": 53318022.0,
      "step": 26305
    },
    {
      "entropy": 6.0133442878723145,
      "epoch": 2.299222231932186,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004469404286026064,
      "loss": 5.6615,
      "mean_token_accuracy": 0.17448236495256425,
      "num_tokens": 53328462.0,
      "step": 26310
    },
    {
      "entropy": 5.983560800552368,
      "epoch": 2.2996591802848902,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044692032793891665,
      "loss": 5.637,
      "mean_token_accuracy": 0.1774396076798439,
      "num_tokens": 53337837.0,
      "step": 26315
    },
    {
      "entropy": 5.968329906463623,
      "epoch": 2.300096128637595,
      "grad_norm": 1.59375,
      "learning_rate": 0.00044690022397770484,
      "loss": 5.6392,
      "mean_token_accuracy": 0.1710306629538536,
      "num_tokens": 53348026.0,
      "step": 26320
    },
    {
      "entropy": 6.062911128997802,
      "epoch": 2.3005330769902996,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004468801167193567,
      "loss": 5.6853,
      "mean_token_accuracy": 0.16790140569210052,
      "num_tokens": 53358881.0,
      "step": 26325
    },
    {
      "entropy": 6.040018844604492,
      "epoch": 2.3009700253430045,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00044686000616425795,
      "loss": 5.6698,
      "mean_token_accuracy": 0.1711098685860634,
      "num_tokens": 53369732.0,
      "step": 26330
    },
    {
      "entropy": 6.014130544662476,
      "epoch": 2.301406973695709,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044683989231279406,
      "loss": 5.6965,
      "mean_token_accuracy": 0.16562426164746286,
      "num_tokens": 53379607.0,
      "step": 26335
    },
    {
      "entropy": 6.155791568756103,
      "epoch": 2.301843922048414,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00044681977516535106,
      "loss": 5.7288,
      "mean_token_accuracy": 0.1599295988678932,
      "num_tokens": 53390630.0,
      "step": 26340
    },
    {
      "entropy": 5.993897819519043,
      "epoch": 2.3022808704011184,
      "grad_norm": 2.125,
      "learning_rate": 0.0004467996547223148,
      "loss": 5.5776,
      "mean_token_accuracy": 0.18366660922765732,
      "num_tokens": 53400830.0,
      "step": 26345
    },
    {
      "entropy": 5.949699020385742,
      "epoch": 2.3027178187538233,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044677953098407107,
      "loss": 5.654,
      "mean_token_accuracy": 0.16837113052606584,
      "num_tokens": 53411150.0,
      "step": 26350
    },
    {
      "entropy": 5.9908709049224855,
      "epoch": 2.303154767106528,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004467594039510059,
      "loss": 5.6511,
      "mean_token_accuracy": 0.1769055962562561,
      "num_tokens": 53421072.0,
      "step": 26355
    },
    {
      "entropy": 6.060421800613403,
      "epoch": 2.3035917154592327,
      "grad_norm": 1.78125,
      "learning_rate": 0.00044673927362350545,
      "loss": 5.6189,
      "mean_token_accuracy": 0.1698520377278328,
      "num_tokens": 53430659.0,
      "step": 26360
    },
    {
      "entropy": 5.977136278152466,
      "epoch": 2.3040286638119376,
      "grad_norm": 1.671875,
      "learning_rate": 0.00044671914000195566,
      "loss": 5.5533,
      "mean_token_accuracy": 0.1776006042957306,
      "num_tokens": 53440883.0,
      "step": 26365
    },
    {
      "entropy": 5.927409076690674,
      "epoch": 2.304465612164642,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00044669900308674283,
      "loss": 5.6771,
      "mean_token_accuracy": 0.16733695268630983,
      "num_tokens": 53450491.0,
      "step": 26370
    },
    {
      "entropy": 5.995192956924439,
      "epoch": 2.304902560517347,
      "grad_norm": 2.359375,
      "learning_rate": 0.00044667886287825316,
      "loss": 5.674,
      "mean_token_accuracy": 0.17079152762889863,
      "num_tokens": 53460864.0,
      "step": 26375
    },
    {
      "entropy": 6.04335355758667,
      "epoch": 2.3053395088700515,
      "grad_norm": 1.5,
      "learning_rate": 0.00044665871937687296,
      "loss": 5.6458,
      "mean_token_accuracy": 0.17359767258167266,
      "num_tokens": 53470965.0,
      "step": 26380
    },
    {
      "entropy": 5.990081024169922,
      "epoch": 2.3057764572227564,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004466385725829885,
      "loss": 5.5741,
      "mean_token_accuracy": 0.17871508449316026,
      "num_tokens": 53480272.0,
      "step": 26385
    },
    {
      "entropy": 5.967279005050659,
      "epoch": 2.306213405575461,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004466184224969863,
      "loss": 5.6911,
      "mean_token_accuracy": 0.17376211285591125,
      "num_tokens": 53490285.0,
      "step": 26390
    },
    {
      "entropy": 5.980613803863525,
      "epoch": 2.3066503539281658,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004465982691192529,
      "loss": 5.5714,
      "mean_token_accuracy": 0.174943807721138,
      "num_tokens": 53499712.0,
      "step": 26395
    },
    {
      "entropy": 5.997421216964722,
      "epoch": 2.3070873022808702,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004465781124501747,
      "loss": 5.6619,
      "mean_token_accuracy": 0.17374014407396315,
      "num_tokens": 53510144.0,
      "step": 26400
    },
    {
      "entropy": 6.029710197448731,
      "epoch": 2.307524250633575,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044655795249013845,
      "loss": 5.6581,
      "mean_token_accuracy": 0.17278134524822236,
      "num_tokens": 53519936.0,
      "step": 26405
    },
    {
      "entropy": 6.008672618865967,
      "epoch": 2.3079611989862796,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004465377892395308,
      "loss": 5.6794,
      "mean_token_accuracy": 0.16849332451820373,
      "num_tokens": 53530447.0,
      "step": 26410
    },
    {
      "entropy": 6.081845712661743,
      "epoch": 2.3083981473389845,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004465176226987384,
      "loss": 5.7277,
      "mean_token_accuracy": 0.16858163774013518,
      "num_tokens": 53540929.0,
      "step": 26415
    },
    {
      "entropy": 6.060597705841064,
      "epoch": 2.3088350956916894,
      "grad_norm": 3.25,
      "learning_rate": 0.0004464974528681481,
      "loss": 5.6266,
      "mean_token_accuracy": 0.17795035243034363,
      "num_tokens": 53551575.0,
      "step": 26420
    },
    {
      "entropy": 6.036684036254883,
      "epoch": 2.309272044044394,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004464772797481469,
      "loss": 5.7574,
      "mean_token_accuracy": 0.16378295719623565,
      "num_tokens": 53563081.0,
      "step": 26425
    },
    {
      "entropy": 6.0402875423431395,
      "epoch": 2.309708992397099,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00044645710333912153,
      "loss": 5.6902,
      "mean_token_accuracy": 0.1727952778339386,
      "num_tokens": 53572303.0,
      "step": 26430
    },
    {
      "entropy": 6.041515874862671,
      "epoch": 2.3101459407498033,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044643692364145914,
      "loss": 5.6321,
      "mean_token_accuracy": 0.17001091986894606,
      "num_tokens": 53582033.0,
      "step": 26435
    },
    {
      "entropy": 5.871869993209839,
      "epoch": 2.310582889102508,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044641674065554665,
      "loss": 5.4556,
      "mean_token_accuracy": 0.19873325675725936,
      "num_tokens": 53592140.0,
      "step": 26440
    },
    {
      "entropy": 5.99180269241333,
      "epoch": 2.3110198374552127,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00044639655438177134,
      "loss": 5.6561,
      "mean_token_accuracy": 0.1653912767767906,
      "num_tokens": 53602601.0,
      "step": 26445
    },
    {
      "entropy": 5.991765737533569,
      "epoch": 2.3114567858079176,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044637636482052027,
      "loss": 5.6433,
      "mean_token_accuracy": 0.16931255012750626,
      "num_tokens": 53613292.0,
      "step": 26450
    },
    {
      "entropy": 6.073553800582886,
      "epoch": 2.311893734160622,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004463561719721807,
      "loss": 5.6544,
      "mean_token_accuracy": 0.1687464475631714,
      "num_tokens": 53624493.0,
      "step": 26455
    },
    {
      "entropy": 5.999756860733032,
      "epoch": 2.312330682513327,
      "grad_norm": 1.59375,
      "learning_rate": 0.00044633597583713995,
      "loss": 5.659,
      "mean_token_accuracy": 0.17221900820732117,
      "num_tokens": 53633739.0,
      "step": 26460
    },
    {
      "entropy": 6.025134897232055,
      "epoch": 2.3127676308660314,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004463157764157855,
      "loss": 5.6328,
      "mean_token_accuracy": 0.17254052311182022,
      "num_tokens": 53643025.0,
      "step": 26465
    },
    {
      "entropy": 6.009926128387451,
      "epoch": 2.3132045792187363,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004462955737085046,
      "loss": 5.6032,
      "mean_token_accuracy": 0.17224823236465453,
      "num_tokens": 53652905.0,
      "step": 26470
    },
    {
      "entropy": 5.98455662727356,
      "epoch": 2.3136415275714413,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004462753677156849,
      "loss": 5.6002,
      "mean_token_accuracy": 0.18067677617073058,
      "num_tokens": 53663189.0,
      "step": 26475
    },
    {
      "entropy": 6.016342401504517,
      "epoch": 2.3140784759241457,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004462551584377138,
      "loss": 5.6754,
      "mean_token_accuracy": 0.17318083494901657,
      "num_tokens": 53673068.0,
      "step": 26480
    },
    {
      "entropy": 6.055838537216187,
      "epoch": 2.3145154242768506,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004462349458749791,
      "loss": 5.7187,
      "mean_token_accuracy": 0.1628237135708332,
      "num_tokens": 53683929.0,
      "step": 26485
    },
    {
      "entropy": 6.0687463760375975,
      "epoch": 2.314952372629555,
      "grad_norm": 1.953125,
      "learning_rate": 0.00044621473002786846,
      "loss": 5.7306,
      "mean_token_accuracy": 0.1670241743326187,
      "num_tokens": 53694170.0,
      "step": 26490
    },
    {
      "entropy": 6.0451148509979244,
      "epoch": 2.31538932098226,
      "grad_norm": 2.375,
      "learning_rate": 0.00044619451089676946,
      "loss": 5.6514,
      "mean_token_accuracy": 0.17146702855825424,
      "num_tokens": 53704491.0,
      "step": 26495
    },
    {
      "entropy": 6.081975650787354,
      "epoch": 2.3158262693349645,
      "grad_norm": 1.625,
      "learning_rate": 0.0004461742884820701,
      "loss": 5.6907,
      "mean_token_accuracy": 0.1680471271276474,
      "num_tokens": 53714971.0,
      "step": 26500
    },
    {
      "entropy": 5.752343463897705,
      "epoch": 2.3162632176876694,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004461540627841582,
      "loss": 5.4088,
      "mean_token_accuracy": 0.20896890610456467,
      "num_tokens": 53723965.0,
      "step": 26505
    },
    {
      "entropy": 6.06356406211853,
      "epoch": 2.316700166040374,
      "grad_norm": 1.375,
      "learning_rate": 0.00044613383380342163,
      "loss": 5.6702,
      "mean_token_accuracy": 0.16653164476156235,
      "num_tokens": 53734561.0,
      "step": 26510
    },
    {
      "entropy": 6.040018844604492,
      "epoch": 2.317137114393079,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004461136015402485,
      "loss": 5.6314,
      "mean_token_accuracy": 0.16823340803384781,
      "num_tokens": 53745115.0,
      "step": 26515
    },
    {
      "entropy": 5.947022247314453,
      "epoch": 2.3175740627457833,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004460933659950269,
      "loss": 5.5753,
      "mean_token_accuracy": 0.17450499534606934,
      "num_tokens": 53754884.0,
      "step": 26520
    },
    {
      "entropy": 5.976708650588989,
      "epoch": 2.318011011098488,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004460731271681448,
      "loss": 5.6697,
      "mean_token_accuracy": 0.16533956676721573,
      "num_tokens": 53764772.0,
      "step": 26525
    },
    {
      "entropy": 6.057402420043945,
      "epoch": 2.318447959451193,
      "grad_norm": 1.859375,
      "learning_rate": 0.00044605288505999047,
      "loss": 5.7337,
      "mean_token_accuracy": 0.16090954542160035,
      "num_tokens": 53775391.0,
      "step": 26530
    },
    {
      "entropy": 6.061769723892212,
      "epoch": 2.3188849078038976,
      "grad_norm": 2.25,
      "learning_rate": 0.00044603263967095224,
      "loss": 5.6786,
      "mean_token_accuracy": 0.167123506963253,
      "num_tokens": 53785698.0,
      "step": 26535
    },
    {
      "entropy": 6.039416408538818,
      "epoch": 2.3193218561566025,
      "grad_norm": 2.078125,
      "learning_rate": 0.00044601239100141833,
      "loss": 5.6434,
      "mean_token_accuracy": 0.1694673016667366,
      "num_tokens": 53795432.0,
      "step": 26540
    },
    {
      "entropy": 5.989963006973267,
      "epoch": 2.319758804509307,
      "grad_norm": 2.203125,
      "learning_rate": 0.00044599213905177724,
      "loss": 5.7042,
      "mean_token_accuracy": 0.16635838598012925,
      "num_tokens": 53805295.0,
      "step": 26545
    },
    {
      "entropy": 6.01372184753418,
      "epoch": 2.320195752862012,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004459718838224171,
      "loss": 5.6681,
      "mean_token_accuracy": 0.17402344197034836,
      "num_tokens": 53814765.0,
      "step": 26550
    },
    {
      "entropy": 6.037171840667725,
      "epoch": 2.3206327012147163,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044595162531372684,
      "loss": 5.6265,
      "mean_token_accuracy": 0.17292014956474305,
      "num_tokens": 53824154.0,
      "step": 26555
    },
    {
      "entropy": 6.026534223556519,
      "epoch": 2.3210696495674212,
      "grad_norm": 1.65625,
      "learning_rate": 0.00044593136352609475,
      "loss": 5.6983,
      "mean_token_accuracy": 0.1623430833220482,
      "num_tokens": 53834867.0,
      "step": 26560
    },
    {
      "entropy": 6.061479091644287,
      "epoch": 2.3215065979201257,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004459110984599096,
      "loss": 5.7339,
      "mean_token_accuracy": 0.16179954558610915,
      "num_tokens": 53844901.0,
      "step": 26565
    },
    {
      "entropy": 6.034414339065552,
      "epoch": 2.3219435462728306,
      "grad_norm": 2.0625,
      "learning_rate": 0.00044589083011555995,
      "loss": 5.7444,
      "mean_token_accuracy": 0.16780135482549668,
      "num_tokens": 53854897.0,
      "step": 26570
    },
    {
      "entropy": 6.040453338623047,
      "epoch": 2.322380494625535,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004458705584934347,
      "loss": 5.7141,
      "mean_token_accuracy": 0.15906959474086763,
      "num_tokens": 53865152.0,
      "step": 26575
    },
    {
      "entropy": 6.1109795570373535,
      "epoch": 2.32281744297824,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044585028359392256,
      "loss": 5.7194,
      "mean_token_accuracy": 0.16372723430395125,
      "num_tokens": 53875261.0,
      "step": 26580
    },
    {
      "entropy": 6.107522773742676,
      "epoch": 2.323254391330945,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004458300054174125,
      "loss": 5.64,
      "mean_token_accuracy": 0.16566128879785538,
      "num_tokens": 53885281.0,
      "step": 26585
    },
    {
      "entropy": 5.991497659683228,
      "epoch": 2.3236913396836494,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004458097239642933,
      "loss": 5.6621,
      "mean_token_accuracy": 0.16816871911287307,
      "num_tokens": 53895282.0,
      "step": 26590
    },
    {
      "entropy": 6.0337646961212155,
      "epoch": 2.3241282880363543,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004457894392349542,
      "loss": 5.6929,
      "mean_token_accuracy": 0.17028872072696685,
      "num_tokens": 53905843.0,
      "step": 26595
    },
    {
      "entropy": 6.021761083602906,
      "epoch": 2.3245652363890588,
      "grad_norm": 1.375,
      "learning_rate": 0.0004457691512297841,
      "loss": 5.7328,
      "mean_token_accuracy": 0.17171102315187453,
      "num_tokens": 53915270.0,
      "step": 26600
    },
    {
      "entropy": 6.0519756317138675,
      "epoch": 2.3250021847417637,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004457488599491723,
      "loss": 5.757,
      "mean_token_accuracy": 0.16094570755958557,
      "num_tokens": 53924548.0,
      "step": 26605
    },
    {
      "entropy": 6.072415685653686,
      "epoch": 2.325439133094468,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00044572856539350784,
      "loss": 5.727,
      "mean_token_accuracy": 0.169219771027565,
      "num_tokens": 53935696.0,
      "step": 26610
    },
    {
      "entropy": 6.06949372291565,
      "epoch": 2.325876081447173,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044570826756318006,
      "loss": 5.679,
      "mean_token_accuracy": 0.169336698949337,
      "num_tokens": 53945499.0,
      "step": 26615
    },
    {
      "entropy": 6.051640558242798,
      "epoch": 2.3263130297998775,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004456879664585782,
      "loss": 5.6232,
      "mean_token_accuracy": 0.17245375066995622,
      "num_tokens": 53955478.0,
      "step": 26620
    },
    {
      "entropy": 6.06209545135498,
      "epoch": 2.3267499781525824,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004456676620800917,
      "loss": 5.5982,
      "mean_token_accuracy": 0.17011075019836425,
      "num_tokens": 53964908.0,
      "step": 26625
    },
    {
      "entropy": 5.993535757064819,
      "epoch": 2.327186926505287,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004456473544281101,
      "loss": 5.6123,
      "mean_token_accuracy": 0.17369502484798433,
      "num_tokens": 53974850.0,
      "step": 26630
    },
    {
      "entropy": 5.949599409103394,
      "epoch": 2.327623874857992,
      "grad_norm": 1.375,
      "learning_rate": 0.0004456270435030228,
      "loss": 5.6126,
      "mean_token_accuracy": 0.1726825714111328,
      "num_tokens": 53984458.0,
      "step": 26635
    },
    {
      "entropy": 6.011135530471802,
      "epoch": 2.3280608232106967,
      "grad_norm": 1.375,
      "learning_rate": 0.0004456067293052193,
      "loss": 5.6357,
      "mean_token_accuracy": 0.17133185267448425,
      "num_tokens": 53994560.0,
      "step": 26640
    },
    {
      "entropy": 6.0561400890350345,
      "epoch": 2.328497771563401,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044558641183508945,
      "loss": 5.6563,
      "mean_token_accuracy": 0.16989081650972365,
      "num_tokens": 54003698.0,
      "step": 26645
    },
    {
      "entropy": 6.007610940933228,
      "epoch": 2.3289347199161057,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044556609109302277,
      "loss": 5.5722,
      "mean_token_accuracy": 0.18324775695800782,
      "num_tokens": 54013405.0,
      "step": 26650
    },
    {
      "entropy": 6.054485034942627,
      "epoch": 2.3293716682688106,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004455457670794092,
      "loss": 5.6622,
      "mean_token_accuracy": 0.17506738156080245,
      "num_tokens": 54023140.0,
      "step": 26655
    },
    {
      "entropy": 6.038486433029175,
      "epoch": 2.3298086166215155,
      "grad_norm": 2.0625,
      "learning_rate": 0.00044552543979463826,
      "loss": 5.6344,
      "mean_token_accuracy": 0.16677625179290773,
      "num_tokens": 54032506.0,
      "step": 26660
    },
    {
      "entropy": 5.984251689910889,
      "epoch": 2.33024556497422,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004455051092391002,
      "loss": 5.5853,
      "mean_token_accuracy": 0.1770073339343071,
      "num_tokens": 54042113.0,
      "step": 26665
    },
    {
      "entropy": 5.982187271118164,
      "epoch": 2.330682513326925,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00044548477541318473,
      "loss": 5.6548,
      "mean_token_accuracy": 0.17640342861413955,
      "num_tokens": 54051139.0,
      "step": 26670
    },
    {
      "entropy": 6.007686710357666,
      "epoch": 2.3311194616796294,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004454644383172819,
      "loss": 5.6414,
      "mean_token_accuracy": 0.17012815773487092,
      "num_tokens": 54060929.0,
      "step": 26675
    },
    {
      "entropy": 6.122113800048828,
      "epoch": 2.3315564100323343,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004454440979517819,
      "loss": 5.6995,
      "mean_token_accuracy": 0.17117901295423507,
      "num_tokens": 54071654.0,
      "step": 26680
    },
    {
      "entropy": 6.021829414367676,
      "epoch": 2.3319933583850387,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004454237543170747,
      "loss": 5.6209,
      "mean_token_accuracy": 0.17389492690563202,
      "num_tokens": 54081497.0,
      "step": 26685
    },
    {
      "entropy": 6.023815441131592,
      "epoch": 2.3324303067377437,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00044540340741355063,
      "loss": 5.6953,
      "mean_token_accuracy": 0.16818227022886276,
      "num_tokens": 54091899.0,
      "step": 26690
    },
    {
      "entropy": 5.955399131774902,
      "epoch": 2.332867255090448,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004453830572416,
      "loss": 5.6191,
      "mean_token_accuracy": 0.17117690145969391,
      "num_tokens": 54101734.0,
      "step": 26695
    },
    {
      "entropy": 6.047651290893555,
      "epoch": 2.333304203443153,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000445362703801613,
      "loss": 5.6882,
      "mean_token_accuracy": 0.16824776232242583,
      "num_tokens": 54112633.0,
      "step": 26700
    },
    {
      "entropy": 6.030196905136108,
      "epoch": 2.3337411517958575,
      "grad_norm": 1.875,
      "learning_rate": 0.00044534234709398,
      "loss": 5.6311,
      "mean_token_accuracy": 0.16954965740442277,
      "num_tokens": 54122182.0,
      "step": 26705
    },
    {
      "entropy": 5.994605541229248,
      "epoch": 2.3341781001485624,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00044532198711909153,
      "loss": 5.6717,
      "mean_token_accuracy": 0.16682935506105423,
      "num_tokens": 54131410.0,
      "step": 26710
    },
    {
      "entropy": 5.995856475830078,
      "epoch": 2.3346150485012673,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004453016238773382,
      "loss": 5.6279,
      "mean_token_accuracy": 0.17490887641906738,
      "num_tokens": 54141890.0,
      "step": 26715
    },
    {
      "entropy": 5.991449213027954,
      "epoch": 2.335051996853972,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004452812573691104,
      "loss": 5.6657,
      "mean_token_accuracy": 0.17314488291740418,
      "num_tokens": 54151579.0,
      "step": 26720
    },
    {
      "entropy": 5.986635971069336,
      "epoch": 2.3354889452066767,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044526088759479893,
      "loss": 5.7046,
      "mean_token_accuracy": 0.170099700987339,
      "num_tokens": 54161824.0,
      "step": 26725
    },
    {
      "entropy": 5.96722445487976,
      "epoch": 2.335925893559381,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00044524051455479435,
      "loss": 5.5797,
      "mean_token_accuracy": 0.17758048474788665,
      "num_tokens": 54172110.0,
      "step": 26730
    },
    {
      "entropy": 5.967855167388916,
      "epoch": 2.336362841912086,
      "grad_norm": 1.609375,
      "learning_rate": 0.00044522013824948756,
      "loss": 5.5955,
      "mean_token_accuracy": 0.17248726934194564,
      "num_tokens": 54181852.0,
      "step": 26735
    },
    {
      "entropy": 6.02219557762146,
      "epoch": 2.3367997902647906,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00044519975867926933,
      "loss": 5.6946,
      "mean_token_accuracy": 0.16577896773815154,
      "num_tokens": 54192703.0,
      "step": 26740
    },
    {
      "entropy": 5.999609756469726,
      "epoch": 2.3372367386174955,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004451793758445305,
      "loss": 5.5591,
      "mean_token_accuracy": 0.17979228794574736,
      "num_tokens": 54201613.0,
      "step": 26745
    },
    {
      "entropy": 5.994759130477905,
      "epoch": 2.3376736869702,
      "grad_norm": 1.578125,
      "learning_rate": 0.000445158989745662,
      "loss": 5.733,
      "mean_token_accuracy": 0.1647741600871086,
      "num_tokens": 54211717.0,
      "step": 26750
    },
    {
      "entropy": 5.912089204788208,
      "epoch": 2.338110635322905,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00044513860038305505,
      "loss": 5.5684,
      "mean_token_accuracy": 0.17850003242492676,
      "num_tokens": 54221481.0,
      "step": 26755
    },
    {
      "entropy": 6.039092779159546,
      "epoch": 2.3385475836756093,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004451182077571005,
      "loss": 5.651,
      "mean_token_accuracy": 0.17078815549612045,
      "num_tokens": 54231558.0,
      "step": 26760
    },
    {
      "entropy": 5.991443824768067,
      "epoch": 2.3389845320283142,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00044509781186818963,
      "loss": 5.6119,
      "mean_token_accuracy": 0.1743158832192421,
      "num_tokens": 54241150.0,
      "step": 26765
    },
    {
      "entropy": 6.0052282333374025,
      "epoch": 2.339421480381019,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004450774127167136,
      "loss": 5.6538,
      "mean_token_accuracy": 0.17234366238117219,
      "num_tokens": 54251323.0,
      "step": 26770
    },
    {
      "entropy": 6.060701894760132,
      "epoch": 2.3398584287337236,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004450570103030637,
      "loss": 5.6619,
      "mean_token_accuracy": 0.17048828303813934,
      "num_tokens": 54260956.0,
      "step": 26775
    },
    {
      "entropy": 5.99488935470581,
      "epoch": 2.3402953770864285,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044503660462763116,
      "loss": 5.6499,
      "mean_token_accuracy": 0.1727110043168068,
      "num_tokens": 54270382.0,
      "step": 26780
    },
    {
      "entropy": 5.9879655838012695,
      "epoch": 2.340732325439133,
      "grad_norm": 1.625,
      "learning_rate": 0.0004450161956908075,
      "loss": 5.6541,
      "mean_token_accuracy": 0.17102298885583878,
      "num_tokens": 54280380.0,
      "step": 26785
    },
    {
      "entropy": 6.05048828125,
      "epoch": 2.341169273791838,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004449957834929841,
      "loss": 5.6654,
      "mean_token_accuracy": 0.17007970809936523,
      "num_tokens": 54290536.0,
      "step": 26790
    },
    {
      "entropy": 5.967138671875,
      "epoch": 2.3416062221445424,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044497536803455245,
      "loss": 5.589,
      "mean_token_accuracy": 0.17610731720924377,
      "num_tokens": 54299835.0,
      "step": 26795
    },
    {
      "entropy": 5.971677970886231,
      "epoch": 2.3420431704972473,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004449549493159043,
      "loss": 5.6393,
      "mean_token_accuracy": 0.17895786017179488,
      "num_tokens": 54309496.0,
      "step": 26800
    },
    {
      "entropy": 6.015941095352173,
      "epoch": 2.3424801188499518,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004449345273374311,
      "loss": 5.5894,
      "mean_token_accuracy": 0.1737799197435379,
      "num_tokens": 54318591.0,
      "step": 26805
    },
    {
      "entropy": 5.979483366012573,
      "epoch": 2.3429170672026567,
      "grad_norm": 3.125,
      "learning_rate": 0.0004449141020995246,
      "loss": 5.6294,
      "mean_token_accuracy": 0.17278883904218673,
      "num_tokens": 54327995.0,
      "step": 26810
    },
    {
      "entropy": 5.992457056045533,
      "epoch": 2.343354015555361,
      "grad_norm": 2.125,
      "learning_rate": 0.0004448936736025765,
      "loss": 5.6548,
      "mean_token_accuracy": 0.16726170778274535,
      "num_tokens": 54339122.0,
      "step": 26815
    },
    {
      "entropy": 6.004678726196289,
      "epoch": 2.343790963908066,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004448732418469789,
      "loss": 5.6976,
      "mean_token_accuracy": 0.16947846710681916,
      "num_tokens": 54349004.0,
      "step": 26820
    },
    {
      "entropy": 6.031132459640503,
      "epoch": 2.344227912260771,
      "grad_norm": 1.921875,
      "learning_rate": 0.00044485280683312345,
      "loss": 5.6597,
      "mean_token_accuracy": 0.17865059971809388,
      "num_tokens": 54359859.0,
      "step": 26825
    },
    {
      "entropy": 6.040171527862549,
      "epoch": 2.3446648606134755,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004448323685614021,
      "loss": 5.6404,
      "mean_token_accuracy": 0.1665738746523857,
      "num_tokens": 54369386.0,
      "step": 26830
    },
    {
      "entropy": 5.989686822891235,
      "epoch": 2.3451018089661804,
      "grad_norm": 1.421875,
      "learning_rate": 0.00044481192703220695,
      "loss": 5.5975,
      "mean_token_accuracy": 0.17743037194013594,
      "num_tokens": 54379250.0,
      "step": 26835
    },
    {
      "entropy": 6.025388097763061,
      "epoch": 2.345538757318885,
      "grad_norm": 1.625,
      "learning_rate": 0.00044479148224593014,
      "loss": 5.6669,
      "mean_token_accuracy": 0.17191943824291228,
      "num_tokens": 54388980.0,
      "step": 26840
    },
    {
      "entropy": 5.909596729278564,
      "epoch": 2.3459757056715898,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00044477103420296365,
      "loss": 5.5459,
      "mean_token_accuracy": 0.18326443880796434,
      "num_tokens": 54398849.0,
      "step": 26845
    },
    {
      "entropy": 6.068065452575683,
      "epoch": 2.346412654024294,
      "grad_norm": 2.75,
      "learning_rate": 0.0004447505829036998,
      "loss": 5.6899,
      "mean_token_accuracy": 0.16801138818264008,
      "num_tokens": 54409426.0,
      "step": 26850
    },
    {
      "entropy": 6.071161413192749,
      "epoch": 2.346849602376999,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004447301283485308,
      "loss": 5.7211,
      "mean_token_accuracy": 0.16987388283014299,
      "num_tokens": 54420212.0,
      "step": 26855
    },
    {
      "entropy": 6.019332027435302,
      "epoch": 2.3472865507297036,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044470967053784904,
      "loss": 5.7825,
      "mean_token_accuracy": 0.16637797355651857,
      "num_tokens": 54431179.0,
      "step": 26860
    },
    {
      "entropy": 6.097363519668579,
      "epoch": 2.3477234990824085,
      "grad_norm": 1.5,
      "learning_rate": 0.0004446892094720468,
      "loss": 5.6702,
      "mean_token_accuracy": 0.17619675844907762,
      "num_tokens": 54442247.0,
      "step": 26865
    },
    {
      "entropy": 5.992118406295776,
      "epoch": 2.348160447435113,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044466874515151667,
      "loss": 5.5941,
      "mean_token_accuracy": 0.18277213871479034,
      "num_tokens": 54451475.0,
      "step": 26870
    },
    {
      "entropy": 5.976864433288574,
      "epoch": 2.348597395787818,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000444648277576651,
      "loss": 5.7734,
      "mean_token_accuracy": 0.15991430580615998,
      "num_tokens": 54463259.0,
      "step": 26875
    },
    {
      "entropy": 5.999522018432617,
      "epoch": 2.349034344140523,
      "grad_norm": 1.515625,
      "learning_rate": 0.00044462780674784264,
      "loss": 5.6393,
      "mean_token_accuracy": 0.17403579503297806,
      "num_tokens": 54473573.0,
      "step": 26880
    },
    {
      "entropy": 6.065029907226562,
      "epoch": 2.3494712924932273,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000444607332665484,
      "loss": 5.7519,
      "mean_token_accuracy": 0.16335904151201247,
      "num_tokens": 54482907.0,
      "step": 26885
    },
    {
      "entropy": 6.044911050796509,
      "epoch": 2.349908240845932,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004445868553299678,
      "loss": 5.6721,
      "mean_token_accuracy": 0.1704813703894615,
      "num_tokens": 54492993.0,
      "step": 26890
    },
    {
      "entropy": 5.994643306732177,
      "epoch": 2.3503451891986367,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004445663747416868,
      "loss": 5.7185,
      "mean_token_accuracy": 0.16053182631731033,
      "num_tokens": 54503826.0,
      "step": 26895
    },
    {
      "entropy": 5.972176551818848,
      "epoch": 2.3507821375513416,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044454589090103406,
      "loss": 5.6904,
      "mean_token_accuracy": 0.16337574124336243,
      "num_tokens": 54513839.0,
      "step": 26900
    },
    {
      "entropy": 5.991372728347779,
      "epoch": 2.351219085904046,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044452540380840217,
      "loss": 5.5653,
      "mean_token_accuracy": 0.17754670828580857,
      "num_tokens": 54523742.0,
      "step": 26905
    },
    {
      "entropy": 6.061034345626831,
      "epoch": 2.351656034256751,
      "grad_norm": 2.296875,
      "learning_rate": 0.00044450491346418426,
      "loss": 5.7437,
      "mean_token_accuracy": 0.15911646634340287,
      "num_tokens": 54534017.0,
      "step": 26910
    },
    {
      "entropy": 5.986285018920898,
      "epoch": 2.3520929826094554,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004444844198687732,
      "loss": 5.6057,
      "mean_token_accuracy": 0.1752506136894226,
      "num_tokens": 54544445.0,
      "step": 26915
    },
    {
      "entropy": 5.967273950576782,
      "epoch": 2.3525299309621603,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004444639230225622,
      "loss": 5.6267,
      "mean_token_accuracy": 0.17026452720165253,
      "num_tokens": 54554966.0,
      "step": 26920
    },
    {
      "entropy": 5.995234775543213,
      "epoch": 2.352966879314865,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004444434229259444,
      "loss": 5.6128,
      "mean_token_accuracy": 0.17300126403570176,
      "num_tokens": 54565005.0,
      "step": 26925
    },
    {
      "entropy": 6.021780967712402,
      "epoch": 2.3534038276675697,
      "grad_norm": 1.375,
      "learning_rate": 0.0004444229195793129,
      "loss": 5.6535,
      "mean_token_accuracy": 0.17149440348148345,
      "num_tokens": 54575242.0,
      "step": 26930
    },
    {
      "entropy": 5.983388757705688,
      "epoch": 2.3538407760202746,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004444024129830611,
      "loss": 5.6557,
      "mean_token_accuracy": 0.16521713733673096,
      "num_tokens": 54585448.0,
      "step": 26935
    },
    {
      "entropy": 5.923087406158447,
      "epoch": 2.354277724372979,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004443819031375822,
      "loss": 5.6052,
      "mean_token_accuracy": 0.17404352575540544,
      "num_tokens": 54595891.0,
      "step": 26940
    },
    {
      "entropy": 6.0540262222290036,
      "epoch": 2.3547146727256836,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004443613900432697,
      "loss": 5.7176,
      "mean_token_accuracy": 0.16265874356031418,
      "num_tokens": 54607407.0,
      "step": 26945
    },
    {
      "entropy": 6.031851768493652,
      "epoch": 2.3551516210783885,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044434087370051693,
      "loss": 5.6693,
      "mean_token_accuracy": 0.16602350026369095,
      "num_tokens": 54617337.0,
      "step": 26950
    },
    {
      "entropy": 6.011455488204956,
      "epoch": 2.3555885694310934,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004443203541097175,
      "loss": 5.6465,
      "mean_token_accuracy": 0.17428868114948273,
      "num_tokens": 54627914.0,
      "step": 26955
    },
    {
      "entropy": 5.993291425704956,
      "epoch": 2.356025517783798,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004442998312712649,
      "loss": 5.5482,
      "mean_token_accuracy": 0.17556644678115846,
      "num_tokens": 54636709.0,
      "step": 26960
    },
    {
      "entropy": 5.875097942352295,
      "epoch": 2.356462466136503,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00044427930518555294,
      "loss": 5.6224,
      "mean_token_accuracy": 0.17575837373733522,
      "num_tokens": 54646167.0,
      "step": 26965
    },
    {
      "entropy": 5.998673820495606,
      "epoch": 2.3568994144892073,
      "grad_norm": 1.375,
      "learning_rate": 0.0004442587758529751,
      "loss": 5.6683,
      "mean_token_accuracy": 0.16926839649677278,
      "num_tokens": 54656663.0,
      "step": 26970
    },
    {
      "entropy": 6.002322483062744,
      "epoch": 2.357336362841912,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004442382432739253,
      "loss": 5.6778,
      "mean_token_accuracy": 0.16616871505975722,
      "num_tokens": 54667123.0,
      "step": 26975
    },
    {
      "entropy": 5.989703607559204,
      "epoch": 2.3577733111946166,
      "grad_norm": 1.703125,
      "learning_rate": 0.00044421770744879725,
      "loss": 5.6971,
      "mean_token_accuracy": 0.17151135206222534,
      "num_tokens": 54677381.0,
      "step": 26980
    },
    {
      "entropy": 6.024236154556275,
      "epoch": 2.3582102595473216,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000444197168377985,
      "loss": 5.6292,
      "mean_token_accuracy": 0.17418697476387024,
      "num_tokens": 54687190.0,
      "step": 26985
    },
    {
      "entropy": 6.069078636169434,
      "epoch": 2.3586472079000265,
      "grad_norm": 1.671875,
      "learning_rate": 0.00044417662606188236,
      "loss": 5.7061,
      "mean_token_accuracy": 0.1678761750459671,
      "num_tokens": 54697102.0,
      "step": 26990
    },
    {
      "entropy": 6.0085348129272464,
      "epoch": 2.359084156252731,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004441560805008834,
      "loss": 5.6694,
      "mean_token_accuracy": 0.1714668869972229,
      "num_tokens": 54707315.0,
      "step": 26995
    },
    {
      "entropy": 6.029694986343384,
      "epoch": 2.3595211046054354,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004441355316953822,
      "loss": 5.7382,
      "mean_token_accuracy": 0.15984728932380676,
      "num_tokens": 54717873.0,
      "step": 27000
    },
    {
      "epoch": 2.3595211046054354,
      "eval_entropy": 5.764208109146425,
      "eval_loss": 5.728896617889404,
      "eval_mean_token_accuracy": 0.17516870081894903,
      "eval_num_tokens": 54717873.0,
      "eval_runtime": 28.3966,
      "eval_samples_per_second": 1303.465,
      "eval_steps_per_second": 162.942,
      "step": 27000
    },
    {
      "entropy": 6.047735261917114,
      "epoch": 2.3599580529581403,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004441149796457728,
      "loss": 5.724,
      "mean_token_accuracy": 0.16295834332704545,
      "num_tokens": 54727723.0,
      "step": 27005
    },
    {
      "entropy": 5.984262752532959,
      "epoch": 2.3603950013108452,
      "grad_norm": 1.4375,
      "learning_rate": 0.00044409442435244947,
      "loss": 5.6268,
      "mean_token_accuracy": 0.1760528042912483,
      "num_tokens": 54738108.0,
      "step": 27010
    },
    {
      "entropy": 5.983205413818359,
      "epoch": 2.3608319496635497,
      "grad_norm": 1.375,
      "learning_rate": 0.0004440738658158065,
      "loss": 5.7266,
      "mean_token_accuracy": 0.16747768074274064,
      "num_tokens": 54749011.0,
      "step": 27015
    },
    {
      "entropy": 6.076025819778442,
      "epoch": 2.3612688980162546,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004440533040362382,
      "loss": 5.6638,
      "mean_token_accuracy": 0.17263995260000228,
      "num_tokens": 54759922.0,
      "step": 27020
    },
    {
      "entropy": 6.027572393417358,
      "epoch": 2.361705846368959,
      "grad_norm": 1.453125,
      "learning_rate": 0.00044403273901413897,
      "loss": 5.6707,
      "mean_token_accuracy": 0.15966630280017852,
      "num_tokens": 54770134.0,
      "step": 27025
    },
    {
      "entropy": 6.041265296936035,
      "epoch": 2.362142794721664,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044401217074990323,
      "loss": 5.6067,
      "mean_token_accuracy": 0.16911523044109344,
      "num_tokens": 54780131.0,
      "step": 27030
    },
    {
      "entropy": 6.01522045135498,
      "epoch": 2.3625797430743685,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044399159924392545,
      "loss": 5.7149,
      "mean_token_accuracy": 0.1661977142095566,
      "num_tokens": 54790601.0,
      "step": 27035
    },
    {
      "entropy": 6.020757532119751,
      "epoch": 2.3630166914270734,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004439710244966003,
      "loss": 5.6361,
      "mean_token_accuracy": 0.16919801235198975,
      "num_tokens": 54799995.0,
      "step": 27040
    },
    {
      "entropy": 6.078271150588989,
      "epoch": 2.363453639779778,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004439504465083223,
      "loss": 5.6626,
      "mean_token_accuracy": 0.16859346628189087,
      "num_tokens": 54810299.0,
      "step": 27045
    },
    {
      "entropy": 5.987949466705322,
      "epoch": 2.3638905881324828,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004439298652794863,
      "loss": 5.6234,
      "mean_token_accuracy": 0.167210353910923,
      "num_tokens": 54820802.0,
      "step": 27050
    },
    {
      "entropy": 5.972722959518433,
      "epoch": 2.3643275364851872,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044390928081048693,
      "loss": 5.6561,
      "mean_token_accuracy": 0.17254883646965027,
      "num_tokens": 54831077.0,
      "step": 27055
    },
    {
      "entropy": 6.056126260757447,
      "epoch": 2.364764484837892,
      "grad_norm": 1.34375,
      "learning_rate": 0.000443888693101719,
      "loss": 5.6726,
      "mean_token_accuracy": 0.17085867673158645,
      "num_tokens": 54842574.0,
      "step": 27060
    },
    {
      "entropy": 5.981585216522217,
      "epoch": 2.365201433190597,
      "grad_norm": 1.828125,
      "learning_rate": 0.00044386810215357747,
      "loss": 5.6,
      "mean_token_accuracy": 0.17761047333478927,
      "num_tokens": 54852365.0,
      "step": 27065
    },
    {
      "entropy": 6.028675746917725,
      "epoch": 2.3656383815433015,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044384750796645724,
      "loss": 5.6811,
      "mean_token_accuracy": 0.16440631449222565,
      "num_tokens": 54863008.0,
      "step": 27070
    },
    {
      "entropy": 6.012291097640992,
      "epoch": 2.3660753298960064,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004438269105407534,
      "loss": 5.6438,
      "mean_token_accuracy": 0.16873999387025834,
      "num_tokens": 54873134.0,
      "step": 27075
    },
    {
      "entropy": 5.991837835311889,
      "epoch": 2.366512278248711,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00044380630987686086,
      "loss": 5.7007,
      "mean_token_accuracy": 0.17226957827806472,
      "num_tokens": 54883075.0,
      "step": 27080
    },
    {
      "entropy": 6.01791729927063,
      "epoch": 2.366949226601416,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00044378570597517495,
      "loss": 5.6229,
      "mean_token_accuracy": 0.1738975465297699,
      "num_tokens": 54892788.0,
      "step": 27085
    },
    {
      "entropy": 5.931267023086548,
      "epoch": 2.3673861749541203,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004437650988360906,
      "loss": 5.6053,
      "mean_token_accuracy": 0.1750090092420578,
      "num_tokens": 54902708.0,
      "step": 27090
    },
    {
      "entropy": 5.937417030334473,
      "epoch": 2.367823123306825,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004437444884600034,
      "loss": 5.605,
      "mean_token_accuracy": 0.17191918641328813,
      "num_tokens": 54912906.0,
      "step": 27095
    },
    {
      "entropy": 6.019880962371826,
      "epoch": 2.3682600716595297,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044372387484730835,
      "loss": 5.6808,
      "mean_token_accuracy": 0.16479394286870958,
      "num_tokens": 54922802.0,
      "step": 27100
    },
    {
      "entropy": 6.059292793273926,
      "epoch": 2.3686970200122346,
      "grad_norm": 1.515625,
      "learning_rate": 0.00044370325799840106,
      "loss": 5.6911,
      "mean_token_accuracy": 0.16572147160768508,
      "num_tokens": 54933203.0,
      "step": 27105
    },
    {
      "entropy": 6.0310393333435055,
      "epoch": 2.369133968364939,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044368263791367684,
      "loss": 5.6482,
      "mean_token_accuracy": 0.170255246758461,
      "num_tokens": 54944081.0,
      "step": 27110
    },
    {
      "entropy": 6.003881978988647,
      "epoch": 2.369570916717644,
      "grad_norm": 2.578125,
      "learning_rate": 0.00044366201459353124,
      "loss": 5.6534,
      "mean_token_accuracy": 0.17115130722522737,
      "num_tokens": 54953421.0,
      "step": 27115
    },
    {
      "entropy": 5.965193128585815,
      "epoch": 2.370007865070349,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00044364138803835984,
      "loss": 5.666,
      "mean_token_accuracy": 0.16783110350370406,
      "num_tokens": 54964127.0,
      "step": 27120
    },
    {
      "entropy": 5.952135324478149,
      "epoch": 2.3704448134230534,
      "grad_norm": 1.640625,
      "learning_rate": 0.00044362075824855816,
      "loss": 5.6137,
      "mean_token_accuracy": 0.17057058662176133,
      "num_tokens": 54974636.0,
      "step": 27125
    },
    {
      "entropy": 6.100927305221558,
      "epoch": 2.3708817617757583,
      "grad_norm": 1.421875,
      "learning_rate": 0.00044360012522452215,
      "loss": 5.7085,
      "mean_token_accuracy": 0.1713084414601326,
      "num_tokens": 54985542.0,
      "step": 27130
    },
    {
      "entropy": 6.000578880310059,
      "epoch": 2.3713187101284627,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004435794889666472,
      "loss": 5.6237,
      "mean_token_accuracy": 0.17876849621534346,
      "num_tokens": 54994973.0,
      "step": 27135
    },
    {
      "entropy": 5.9691015720367435,
      "epoch": 2.3717556584811677,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044355884947532947,
      "loss": 5.7203,
      "mean_token_accuracy": 0.16475922912359237,
      "num_tokens": 55006510.0,
      "step": 27140
    },
    {
      "entropy": 6.053398180007934,
      "epoch": 2.372192606833872,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004435382067509647,
      "loss": 5.7108,
      "mean_token_accuracy": 0.1684891864657402,
      "num_tokens": 55016375.0,
      "step": 27145
    },
    {
      "entropy": 5.9555364608764645,
      "epoch": 2.372629555186577,
      "grad_norm": 2.375,
      "learning_rate": 0.00044351756079394865,
      "loss": 5.6392,
      "mean_token_accuracy": 0.1787565067410469,
      "num_tokens": 55026213.0,
      "step": 27150
    },
    {
      "entropy": 6.067756271362304,
      "epoch": 2.3730665035392815,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004434969116046776,
      "loss": 5.7412,
      "mean_token_accuracy": 0.16143137067556382,
      "num_tokens": 55036836.0,
      "step": 27155
    },
    {
      "entropy": 6.065904664993286,
      "epoch": 2.3735034518919864,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004434762591835475,
      "loss": 5.5809,
      "mean_token_accuracy": 0.17315057963132857,
      "num_tokens": 55046243.0,
      "step": 27160
    },
    {
      "entropy": 5.994756841659546,
      "epoch": 2.373940400244691,
      "grad_norm": 2.265625,
      "learning_rate": 0.00044345560353095437,
      "loss": 5.6477,
      "mean_token_accuracy": 0.17336818426847458,
      "num_tokens": 55055882.0,
      "step": 27165
    },
    {
      "entropy": 5.921889162063598,
      "epoch": 2.374377348597396,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004434349446472946,
      "loss": 5.5752,
      "mean_token_accuracy": 0.17300212234258652,
      "num_tokens": 55066398.0,
      "step": 27170
    },
    {
      "entropy": 6.066623592376709,
      "epoch": 2.3748142969501007,
      "grad_norm": 1.875,
      "learning_rate": 0.00044341428253296436,
      "loss": 5.7188,
      "mean_token_accuracy": 0.16623372137546538,
      "num_tokens": 55077053.0,
      "step": 27175
    },
    {
      "entropy": 6.019044685363769,
      "epoch": 2.375251245302805,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004433936171883598,
      "loss": 5.6401,
      "mean_token_accuracy": 0.1729683145880699,
      "num_tokens": 55086973.0,
      "step": 27180
    },
    {
      "entropy": 5.979325151443481,
      "epoch": 2.37568819365551,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004433729486138775,
      "loss": 5.6021,
      "mean_token_accuracy": 0.1751595839858055,
      "num_tokens": 55096162.0,
      "step": 27185
    },
    {
      "entropy": 5.962859439849853,
      "epoch": 2.3761251420082146,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044335227680991385,
      "loss": 5.6196,
      "mean_token_accuracy": 0.1745789349079132,
      "num_tokens": 55105852.0,
      "step": 27190
    },
    {
      "entropy": 5.9933881759643555,
      "epoch": 2.3765620903609195,
      "grad_norm": 1.75,
      "learning_rate": 0.0004433316017768653,
      "loss": 5.555,
      "mean_token_accuracy": 0.18254515826702117,
      "num_tokens": 55116092.0,
      "step": 27195
    },
    {
      "entropy": 6.018254995346069,
      "epoch": 2.376999038713624,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004433109235151284,
      "loss": 5.6246,
      "mean_token_accuracy": 0.177516570687294,
      "num_tokens": 55125439.0,
      "step": 27200
    },
    {
      "entropy": 5.9694507122039795,
      "epoch": 2.377435987066329,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044329024202509977,
      "loss": 5.6853,
      "mean_token_accuracy": 0.16751171350479127,
      "num_tokens": 55134684.0,
      "step": 27205
    },
    {
      "entropy": 6.021972227096557,
      "epoch": 2.3778729354190333,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044326955730717614,
      "loss": 5.5997,
      "mean_token_accuracy": 0.17277602702379227,
      "num_tokens": 55143853.0,
      "step": 27210
    },
    {
      "entropy": 6.080524730682373,
      "epoch": 2.3783098837717382,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004432488693617542,
      "loss": 5.8124,
      "mean_token_accuracy": 0.15974631756544114,
      "num_tokens": 55154190.0,
      "step": 27215
    },
    {
      "entropy": 6.039464855194092,
      "epoch": 2.3787468321244427,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044322817818923083,
      "loss": 5.6743,
      "mean_token_accuracy": 0.1718152031302452,
      "num_tokens": 55164158.0,
      "step": 27220
    },
    {
      "entropy": 6.0481367111206055,
      "epoch": 2.3791837804771476,
      "grad_norm": 1.46875,
      "learning_rate": 0.00044320748379000284,
      "loss": 5.7341,
      "mean_token_accuracy": 0.16180138140916825,
      "num_tokens": 55173635.0,
      "step": 27225
    },
    {
      "entropy": 6.0084034442901615,
      "epoch": 2.3796207288298525,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044318678616446717,
      "loss": 5.6365,
      "mean_token_accuracy": 0.18112799525260925,
      "num_tokens": 55182972.0,
      "step": 27230
    },
    {
      "entropy": 6.075549125671387,
      "epoch": 2.380057677182557,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044316608531302085,
      "loss": 5.6919,
      "mean_token_accuracy": 0.16747928857803346,
      "num_tokens": 55192769.0,
      "step": 27235
    },
    {
      "entropy": 5.980849647521973,
      "epoch": 2.380494625535262,
      "grad_norm": 1.46875,
      "learning_rate": 0.00044314538123606083,
      "loss": 5.6749,
      "mean_token_accuracy": 0.17080578058958054,
      "num_tokens": 55203241.0,
      "step": 27240
    },
    {
      "entropy": 5.9540351867675785,
      "epoch": 2.3809315738879664,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044312467393398434,
      "loss": 5.6213,
      "mean_token_accuracy": 0.17798683494329454,
      "num_tokens": 55214272.0,
      "step": 27245
    },
    {
      "entropy": 6.0015465259552006,
      "epoch": 2.3813685222406713,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044310396340718853,
      "loss": 5.5597,
      "mean_token_accuracy": 0.18321134448051452,
      "num_tokens": 55223947.0,
      "step": 27250
    },
    {
      "entropy": 5.992421007156372,
      "epoch": 2.3818054705933758,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044308324965607054,
      "loss": 5.7291,
      "mean_token_accuracy": 0.1617117002606392,
      "num_tokens": 55234906.0,
      "step": 27255
    },
    {
      "entropy": 6.0075665473937985,
      "epoch": 2.3822424189460807,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004430625326810278,
      "loss": 5.5462,
      "mean_token_accuracy": 0.17785390615463256,
      "num_tokens": 55244785.0,
      "step": 27260
    },
    {
      "entropy": 6.062753963470459,
      "epoch": 2.382679367298785,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044304181248245754,
      "loss": 5.7122,
      "mean_token_accuracy": 0.16706502586603164,
      "num_tokens": 55255053.0,
      "step": 27265
    },
    {
      "entropy": 6.029832983016968,
      "epoch": 2.38311631565149,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004430210890607574,
      "loss": 5.725,
      "mean_token_accuracy": 0.16604359149932862,
      "num_tokens": 55264525.0,
      "step": 27270
    },
    {
      "entropy": 6.050643825531006,
      "epoch": 2.3835532640041945,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044300036241632463,
      "loss": 5.6412,
      "mean_token_accuracy": 0.17505367398262023,
      "num_tokens": 55273486.0,
      "step": 27275
    },
    {
      "entropy": 6.015322875976563,
      "epoch": 2.3839902123568995,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044297963254955687,
      "loss": 5.6263,
      "mean_token_accuracy": 0.17410041987895966,
      "num_tokens": 55283164.0,
      "step": 27280
    },
    {
      "entropy": 6.0127105712890625,
      "epoch": 2.3844271607096044,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044295889946085185,
      "loss": 5.6935,
      "mean_token_accuracy": 0.1669090732932091,
      "num_tokens": 55294657.0,
      "step": 27285
    },
    {
      "entropy": 6.050003290176392,
      "epoch": 2.384864109062309,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044293816315060704,
      "loss": 5.689,
      "mean_token_accuracy": 0.1668237879872322,
      "num_tokens": 55304743.0,
      "step": 27290
    },
    {
      "entropy": 5.944294691085815,
      "epoch": 2.3853010574150133,
      "grad_norm": 1.25,
      "learning_rate": 0.0004429174236192203,
      "loss": 5.5931,
      "mean_token_accuracy": 0.1890907406806946,
      "num_tokens": 55314106.0,
      "step": 27295
    },
    {
      "entropy": 6.012021207809449,
      "epoch": 2.385738005767718,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044289668086708935,
      "loss": 5.6406,
      "mean_token_accuracy": 0.16914516389369966,
      "num_tokens": 55324407.0,
      "step": 27300
    },
    {
      "entropy": 6.0448487281799315,
      "epoch": 2.386174954120423,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00044287593489461205,
      "loss": 5.6554,
      "mean_token_accuracy": 0.1706123933196068,
      "num_tokens": 55335853.0,
      "step": 27305
    },
    {
      "entropy": 5.988230419158936,
      "epoch": 2.3866119024731276,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044285518570218635,
      "loss": 5.6322,
      "mean_token_accuracy": 0.17273742258548735,
      "num_tokens": 55345947.0,
      "step": 27310
    },
    {
      "entropy": 6.020481443405151,
      "epoch": 2.3870488508258325,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004428344332902103,
      "loss": 5.6762,
      "mean_token_accuracy": 0.16946894973516463,
      "num_tokens": 55355709.0,
      "step": 27315
    },
    {
      "entropy": 6.016547012329101,
      "epoch": 2.387485799178537,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00044281367765908186,
      "loss": 5.5905,
      "mean_token_accuracy": 0.1695776104927063,
      "num_tokens": 55365597.0,
      "step": 27320
    },
    {
      "entropy": 5.999925422668457,
      "epoch": 2.387922747531242,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044279291880919915,
      "loss": 5.6442,
      "mean_token_accuracy": 0.17121700793504716,
      "num_tokens": 55375508.0,
      "step": 27325
    },
    {
      "entropy": 5.943311023712158,
      "epoch": 2.3883596958839464,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044277215674096035,
      "loss": 5.6299,
      "mean_token_accuracy": 0.1699487864971161,
      "num_tokens": 55384928.0,
      "step": 27330
    },
    {
      "entropy": 6.007876300811768,
      "epoch": 2.3887966442366513,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004427513914547637,
      "loss": 5.6496,
      "mean_token_accuracy": 0.1745507687330246,
      "num_tokens": 55394805.0,
      "step": 27335
    },
    {
      "entropy": 6.008000755310059,
      "epoch": 2.3892335925893557,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004427306229510074,
      "loss": 5.6755,
      "mean_token_accuracy": 0.17142999172210693,
      "num_tokens": 55405360.0,
      "step": 27340
    },
    {
      "entropy": 6.034095668792725,
      "epoch": 2.3896705409420607,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00044270985123008984,
      "loss": 5.6316,
      "mean_token_accuracy": 0.1645196259021759,
      "num_tokens": 55415124.0,
      "step": 27345
    },
    {
      "entropy": 6.010285377502441,
      "epoch": 2.390107489294765,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004426890762924096,
      "loss": 5.6968,
      "mean_token_accuracy": 0.1687926545739174,
      "num_tokens": 55425621.0,
      "step": 27350
    },
    {
      "entropy": 6.08832950592041,
      "epoch": 2.39054443764747,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004426682981383648,
      "loss": 5.714,
      "mean_token_accuracy": 0.16404400318861007,
      "num_tokens": 55435812.0,
      "step": 27355
    },
    {
      "entropy": 6.011795091629028,
      "epoch": 2.390981386000175,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004426475167683543,
      "loss": 5.66,
      "mean_token_accuracy": 0.16874019354581832,
      "num_tokens": 55445462.0,
      "step": 27360
    },
    {
      "entropy": 6.015990495681763,
      "epoch": 2.3914183343528794,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044262673218277654,
      "loss": 5.663,
      "mean_token_accuracy": 0.16298411786556244,
      "num_tokens": 55455992.0,
      "step": 27365
    },
    {
      "entropy": 6.097350072860718,
      "epoch": 2.3918552827055843,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044260594438203025,
      "loss": 5.7205,
      "mean_token_accuracy": 0.16504674702882766,
      "num_tokens": 55466715.0,
      "step": 27370
    },
    {
      "entropy": 6.053145217895508,
      "epoch": 2.392292231058289,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044258515336651416,
      "loss": 5.6478,
      "mean_token_accuracy": 0.1743471220135689,
      "num_tokens": 55476280.0,
      "step": 27375
    },
    {
      "entropy": 5.9903394222259525,
      "epoch": 2.3927291794109937,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044256435913662696,
      "loss": 5.6088,
      "mean_token_accuracy": 0.17398712038993835,
      "num_tokens": 55486105.0,
      "step": 27380
    },
    {
      "entropy": 6.024464893341064,
      "epoch": 2.393166127763698,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004425435616927676,
      "loss": 5.6509,
      "mean_token_accuracy": 0.16936458349227906,
      "num_tokens": 55496152.0,
      "step": 27385
    },
    {
      "entropy": 6.022613477706909,
      "epoch": 2.393603076116403,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004425227610353349,
      "loss": 5.6741,
      "mean_token_accuracy": 0.17778374999761581,
      "num_tokens": 55507104.0,
      "step": 27390
    },
    {
      "entropy": 6.006140661239624,
      "epoch": 2.3940400244691076,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004425019571647279,
      "loss": 5.6543,
      "mean_token_accuracy": 0.1717411607503891,
      "num_tokens": 55517114.0,
      "step": 27395
    },
    {
      "entropy": 5.951012420654297,
      "epoch": 2.3944769728218125,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044248115008134557,
      "loss": 5.5641,
      "mean_token_accuracy": 0.17964337021112442,
      "num_tokens": 55526719.0,
      "step": 27400
    },
    {
      "entropy": 5.992446231842041,
      "epoch": 2.394913921174517,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044246033978558697,
      "loss": 5.6109,
      "mean_token_accuracy": 0.17304812073707582,
      "num_tokens": 55536760.0,
      "step": 27405
    },
    {
      "entropy": 6.00459303855896,
      "epoch": 2.395350869527222,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044243952627785136,
      "loss": 5.6657,
      "mean_token_accuracy": 0.17079958766698838,
      "num_tokens": 55547193.0,
      "step": 27410
    },
    {
      "entropy": 6.012344455718994,
      "epoch": 2.395787817879927,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044241870955853793,
      "loss": 5.6041,
      "mean_token_accuracy": 0.17567792981863023,
      "num_tokens": 55557257.0,
      "step": 27415
    },
    {
      "entropy": 5.960874462127686,
      "epoch": 2.3962247662326313,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044239788962804584,
      "loss": 5.6354,
      "mean_token_accuracy": 0.1709110274910927,
      "num_tokens": 55568191.0,
      "step": 27420
    },
    {
      "entropy": 5.999865293502808,
      "epoch": 2.396661714585336,
      "grad_norm": 1.671875,
      "learning_rate": 0.00044237706648677456,
      "loss": 5.6639,
      "mean_token_accuracy": 0.16470031142234803,
      "num_tokens": 55578983.0,
      "step": 27425
    },
    {
      "entropy": 5.988319396972656,
      "epoch": 2.3970986629380406,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004423562401351234,
      "loss": 5.6121,
      "mean_token_accuracy": 0.17255409210920333,
      "num_tokens": 55588772.0,
      "step": 27430
    },
    {
      "entropy": 5.897207832336425,
      "epoch": 2.3975356112907455,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004423354105734919,
      "loss": 5.6361,
      "mean_token_accuracy": 0.1715884417295456,
      "num_tokens": 55599514.0,
      "step": 27435
    },
    {
      "entropy": 6.012020063400269,
      "epoch": 2.39797255964345,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004423145778022795,
      "loss": 5.6697,
      "mean_token_accuracy": 0.16502425223588943,
      "num_tokens": 55609183.0,
      "step": 27440
    },
    {
      "entropy": 6.049776220321656,
      "epoch": 2.398409507996155,
      "grad_norm": 1.5,
      "learning_rate": 0.0004422937418218859,
      "loss": 5.6634,
      "mean_token_accuracy": 0.16621574014425278,
      "num_tokens": 55618836.0,
      "step": 27445
    },
    {
      "entropy": 6.025172376632691,
      "epoch": 2.3988464563488594,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044227290263271066,
      "loss": 5.6975,
      "mean_token_accuracy": 0.16778405457735063,
      "num_tokens": 55629314.0,
      "step": 27450
    },
    {
      "entropy": 6.002670383453369,
      "epoch": 2.3992834047015643,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004422520602351535,
      "loss": 5.5663,
      "mean_token_accuracy": 0.18581517040729523,
      "num_tokens": 55640542.0,
      "step": 27455
    },
    {
      "entropy": 5.973333072662354,
      "epoch": 2.399720353054269,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004422312146296141,
      "loss": 5.7084,
      "mean_token_accuracy": 0.16489446759223939,
      "num_tokens": 55650165.0,
      "step": 27460
    },
    {
      "entropy": 6.035307359695435,
      "epoch": 2.4001573014069737,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044221036581649245,
      "loss": 5.6611,
      "mean_token_accuracy": 0.16691832095384598,
      "num_tokens": 55658962.0,
      "step": 27465
    },
    {
      "entropy": 6.03214750289917,
      "epoch": 2.4005942497596786,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004421895137961884,
      "loss": 5.6448,
      "mean_token_accuracy": 0.16530183106660842,
      "num_tokens": 55668357.0,
      "step": 27470
    },
    {
      "entropy": 6.052408027648926,
      "epoch": 2.401031198112383,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00044216865856910176,
      "loss": 5.7549,
      "mean_token_accuracy": 0.16897477954626083,
      "num_tokens": 55678313.0,
      "step": 27475
    },
    {
      "entropy": 6.048983001708985,
      "epoch": 2.401468146465088,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00044214780013563277,
      "loss": 5.666,
      "mean_token_accuracy": 0.16566652208566665,
      "num_tokens": 55689332.0,
      "step": 27480
    },
    {
      "entropy": 6.017721462249756,
      "epoch": 2.4019050948177925,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044212693849618136,
      "loss": 5.6562,
      "mean_token_accuracy": 0.17405372858047485,
      "num_tokens": 55699397.0,
      "step": 27485
    },
    {
      "entropy": 6.022109889984131,
      "epoch": 2.4023420431704974,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004421060736511476,
      "loss": 5.7029,
      "mean_token_accuracy": 0.17003467082977294,
      "num_tokens": 55708548.0,
      "step": 27490
    },
    {
      "entropy": 6.053746223449707,
      "epoch": 2.402778991523202,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044208520560093183,
      "loss": 5.7038,
      "mean_token_accuracy": 0.16823964416980744,
      "num_tokens": 55718043.0,
      "step": 27495
    },
    {
      "entropy": 6.05175371170044,
      "epoch": 2.4032159398759068,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004420643343459343,
      "loss": 5.6608,
      "mean_token_accuracy": 0.16894715577363967,
      "num_tokens": 55727889.0,
      "step": 27500
    },
    {
      "entropy": 5.996723937988281,
      "epoch": 2.4036528882286112,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004420434598865553,
      "loss": 5.6864,
      "mean_token_accuracy": 0.17481441497802735,
      "num_tokens": 55738993.0,
      "step": 27505
    },
    {
      "entropy": 6.021278762817383,
      "epoch": 2.404089836581316,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004420225822231952,
      "loss": 5.6716,
      "mean_token_accuracy": 0.17193639278411865,
      "num_tokens": 55748858.0,
      "step": 27510
    },
    {
      "entropy": 5.992381143569946,
      "epoch": 2.4045267849340206,
      "grad_norm": 1.609375,
      "learning_rate": 0.00044200170135625446,
      "loss": 5.6439,
      "mean_token_accuracy": 0.17655548751354216,
      "num_tokens": 55758639.0,
      "step": 27515
    },
    {
      "entropy": 6.079120016098022,
      "epoch": 2.4049637332867255,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004419808172861336,
      "loss": 5.7015,
      "mean_token_accuracy": 0.16405318677425385,
      "num_tokens": 55769632.0,
      "step": 27520
    },
    {
      "entropy": 6.058025741577149,
      "epoch": 2.4054006816394304,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004419599300132331,
      "loss": 5.7203,
      "mean_token_accuracy": 0.1673765942454338,
      "num_tokens": 55780147.0,
      "step": 27525
    },
    {
      "entropy": 6.009837484359741,
      "epoch": 2.405837629992135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044193903953795367,
      "loss": 5.6052,
      "mean_token_accuracy": 0.17039939463138581,
      "num_tokens": 55790933.0,
      "step": 27530
    },
    {
      "entropy": 5.97081298828125,
      "epoch": 2.40627457834484,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044191814586069605,
      "loss": 5.6024,
      "mean_token_accuracy": 0.17834389805793763,
      "num_tokens": 55800479.0,
      "step": 27535
    },
    {
      "entropy": 6.027687120437622,
      "epoch": 2.4067115266975443,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004418972489818609,
      "loss": 5.6392,
      "mean_token_accuracy": 0.17343633621931076,
      "num_tokens": 55810613.0,
      "step": 27540
    },
    {
      "entropy": 6.0783837795257565,
      "epoch": 2.407148475050249,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00044187634890184904,
      "loss": 5.7042,
      "mean_token_accuracy": 0.1689218834042549,
      "num_tokens": 55821507.0,
      "step": 27545
    },
    {
      "entropy": 6.030051231384277,
      "epoch": 2.4075854234029537,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004418554456210614,
      "loss": 5.7438,
      "mean_token_accuracy": 0.1606538251042366,
      "num_tokens": 55831878.0,
      "step": 27550
    },
    {
      "entropy": 6.02212700843811,
      "epoch": 2.4080223717556586,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044183453913989885,
      "loss": 5.5921,
      "mean_token_accuracy": 0.17469872683286666,
      "num_tokens": 55841314.0,
      "step": 27555
    },
    {
      "entropy": 6.008976888656616,
      "epoch": 2.408459320108363,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004418136294587624,
      "loss": 5.6383,
      "mean_token_accuracy": 0.17528133541345597,
      "num_tokens": 55851056.0,
      "step": 27560
    },
    {
      "entropy": 5.9451805591583256,
      "epoch": 2.408896268461068,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004417927165780531,
      "loss": 5.6324,
      "mean_token_accuracy": 0.1716329574584961,
      "num_tokens": 55860705.0,
      "step": 27565
    },
    {
      "entropy": 6.017947244644165,
      "epoch": 2.4093332168137724,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004417718004981723,
      "loss": 5.6637,
      "mean_token_accuracy": 0.17125621438026428,
      "num_tokens": 55870749.0,
      "step": 27570
    },
    {
      "entropy": 6.0561610698699955,
      "epoch": 2.4097701651664774,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004417508812195207,
      "loss": 5.7473,
      "mean_token_accuracy": 0.16355765014886856,
      "num_tokens": 55880135.0,
      "step": 27575
    },
    {
      "entropy": 6.096548223495484,
      "epoch": 2.4102071135191823,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004417299587425001,
      "loss": 5.5937,
      "mean_token_accuracy": 0.18011222779750824,
      "num_tokens": 55890357.0,
      "step": 27580
    },
    {
      "entropy": 6.060089540481568,
      "epoch": 2.4106440618718867,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004417090330675114,
      "loss": 5.6786,
      "mean_token_accuracy": 0.16804808229207993,
      "num_tokens": 55900080.0,
      "step": 27585
    },
    {
      "entropy": 5.934854984283447,
      "epoch": 2.4110810102245916,
      "grad_norm": 1.4375,
      "learning_rate": 0.00044168810419495597,
      "loss": 5.6722,
      "mean_token_accuracy": 0.170474012196064,
      "num_tokens": 55911326.0,
      "step": 27590
    },
    {
      "entropy": 6.011664915084839,
      "epoch": 2.411517958577296,
      "grad_norm": 1.25,
      "learning_rate": 0.0004416671721252355,
      "loss": 5.6005,
      "mean_token_accuracy": 0.17549705654382705,
      "num_tokens": 55921872.0,
      "step": 27595
    },
    {
      "entropy": 5.9432604789733885,
      "epoch": 2.411954906930001,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044164623685875133,
      "loss": 5.6414,
      "mean_token_accuracy": 0.18062955290079116,
      "num_tokens": 55932539.0,
      "step": 27600
    },
    {
      "entropy": 5.994942045211792,
      "epoch": 2.4123918552827055,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000441625298395905,
      "loss": 5.673,
      "mean_token_accuracy": 0.17705821692943574,
      "num_tokens": 55942087.0,
      "step": 27605
    },
    {
      "entropy": 6.048709678649902,
      "epoch": 2.4128288036354104,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004416043567370981,
      "loss": 5.6471,
      "mean_token_accuracy": 0.17217118740081788,
      "num_tokens": 55951604.0,
      "step": 27610
    },
    {
      "entropy": 6.006827735900879,
      "epoch": 2.413265751988115,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004415834118827323,
      "loss": 5.6292,
      "mean_token_accuracy": 0.17297839224338532,
      "num_tokens": 55961687.0,
      "step": 27615
    },
    {
      "entropy": 6.032395935058593,
      "epoch": 2.41370270034082,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004415624638332095,
      "loss": 5.6949,
      "mean_token_accuracy": 0.16734401434659957,
      "num_tokens": 55972088.0,
      "step": 27620
    },
    {
      "entropy": 5.992652177810669,
      "epoch": 2.4141396486935243,
      "grad_norm": 1.671875,
      "learning_rate": 0.00044154151258893125,
      "loss": 5.6687,
      "mean_token_accuracy": 0.16224369108676912,
      "num_tokens": 55981857.0,
      "step": 27625
    },
    {
      "entropy": 5.989962482452393,
      "epoch": 2.414576597046229,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004415205581502996,
      "loss": 5.5713,
      "mean_token_accuracy": 0.18086562901735306,
      "num_tokens": 55992196.0,
      "step": 27630
    },
    {
      "entropy": 5.978525829315186,
      "epoch": 2.415013545398934,
      "grad_norm": 3.25,
      "learning_rate": 0.00044149960051771643,
      "loss": 5.5936,
      "mean_token_accuracy": 0.17399648427963257,
      "num_tokens": 56002295.0,
      "step": 27635
    },
    {
      "entropy": 5.9840062141418455,
      "epoch": 2.4154504937516386,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044147863969158353,
      "loss": 5.6193,
      "mean_token_accuracy": 0.17308146506547928,
      "num_tokens": 56011847.0,
      "step": 27640
    },
    {
      "entropy": 5.937113952636719,
      "epoch": 2.415887442104343,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004414576756723032,
      "loss": 5.5515,
      "mean_token_accuracy": 0.18193933516740798,
      "num_tokens": 56021717.0,
      "step": 27645
    },
    {
      "entropy": 6.02287187576294,
      "epoch": 2.416324390457048,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004414367084602774,
      "loss": 5.6739,
      "mean_token_accuracy": 0.1681233361363411,
      "num_tokens": 56031845.0,
      "step": 27650
    },
    {
      "entropy": 5.972611284255981,
      "epoch": 2.416761338809753,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004414157380559083,
      "loss": 5.6486,
      "mean_token_accuracy": 0.17439601570367813,
      "num_tokens": 56042082.0,
      "step": 27655
    },
    {
      "entropy": 6.031040048599243,
      "epoch": 2.4171982871624573,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044139476445959817,
      "loss": 5.651,
      "mean_token_accuracy": 0.16867455393075942,
      "num_tokens": 56052592.0,
      "step": 27660
    },
    {
      "entropy": 6.003866529464721,
      "epoch": 2.4176352355151622,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004413737876717492,
      "loss": 5.7312,
      "mean_token_accuracy": 0.15980170369148256,
      "num_tokens": 56062500.0,
      "step": 27665
    },
    {
      "entropy": 5.9655615329742435,
      "epoch": 2.4180721838678667,
      "grad_norm": 1.375,
      "learning_rate": 0.0004413528076927639,
      "loss": 5.5576,
      "mean_token_accuracy": 0.1742238685488701,
      "num_tokens": 56072912.0,
      "step": 27670
    },
    {
      "entropy": 6.0139732837677,
      "epoch": 2.4185091322205716,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004413318245230446,
      "loss": 5.6711,
      "mean_token_accuracy": 0.16769876331090927,
      "num_tokens": 56082591.0,
      "step": 27675
    },
    {
      "entropy": 5.9267514705657955,
      "epoch": 2.418946080573276,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004413108381629936,
      "loss": 5.5511,
      "mean_token_accuracy": 0.17916042357683182,
      "num_tokens": 56091738.0,
      "step": 27680
    },
    {
      "entropy": 5.983007335662842,
      "epoch": 2.419383028925981,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004412898486130137,
      "loss": 5.6683,
      "mean_token_accuracy": 0.16942031383514405,
      "num_tokens": 56102011.0,
      "step": 27685
    },
    {
      "entropy": 6.0146331787109375,
      "epoch": 2.4198199772786855,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00044126885587350735,
      "loss": 5.6587,
      "mean_token_accuracy": 0.17423245757818223,
      "num_tokens": 56112901.0,
      "step": 27690
    },
    {
      "entropy": 5.998768663406372,
      "epoch": 2.4202569256313904,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004412478599448772,
      "loss": 5.6497,
      "mean_token_accuracy": 0.17302301079034804,
      "num_tokens": 56123598.0,
      "step": 27695
    },
    {
      "entropy": 5.964513683319092,
      "epoch": 2.420693873984095,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000441226860827526,
      "loss": 5.6527,
      "mean_token_accuracy": 0.16832312494516372,
      "num_tokens": 56133686.0,
      "step": 27700
    },
    {
      "entropy": 6.048338556289673,
      "epoch": 2.4211308223367998,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044120585852185654,
      "loss": 5.671,
      "mean_token_accuracy": 0.16454783380031585,
      "num_tokens": 56143722.0,
      "step": 27705
    },
    {
      "entropy": 5.992341804504394,
      "epoch": 2.4215677706895047,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044118485302827156,
      "loss": 5.5886,
      "mean_token_accuracy": 0.18175728917121886,
      "num_tokens": 56153781.0,
      "step": 27710
    },
    {
      "entropy": 5.91708779335022,
      "epoch": 2.422004719042209,
      "grad_norm": 1.71875,
      "learning_rate": 0.00044116384434717405,
      "loss": 5.6169,
      "mean_token_accuracy": 0.17654758095741271,
      "num_tokens": 56163398.0,
      "step": 27715
    },
    {
      "entropy": 6.053183555603027,
      "epoch": 2.422441667394914,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004411428324789669,
      "loss": 5.7092,
      "mean_token_accuracy": 0.17204767912626268,
      "num_tokens": 56173418.0,
      "step": 27720
    },
    {
      "entropy": 6.050770854949951,
      "epoch": 2.4228786157476185,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044112181742405324,
      "loss": 5.7562,
      "mean_token_accuracy": 0.16434993892908095,
      "num_tokens": 56184547.0,
      "step": 27725
    },
    {
      "entropy": 6.052869129180908,
      "epoch": 2.4233155641003234,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000441100799182836,
      "loss": 5.6445,
      "mean_token_accuracy": 0.16755487471818925,
      "num_tokens": 56195404.0,
      "step": 27730
    },
    {
      "entropy": 5.966994524002075,
      "epoch": 2.423752512453028,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004410797777557184,
      "loss": 5.5982,
      "mean_token_accuracy": 0.17669935524463654,
      "num_tokens": 56205552.0,
      "step": 27735
    },
    {
      "entropy": 5.960805988311767,
      "epoch": 2.424189460805733,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044105875314310374,
      "loss": 5.62,
      "mean_token_accuracy": 0.17208915054798127,
      "num_tokens": 56215581.0,
      "step": 27740
    },
    {
      "entropy": 5.963318634033203,
      "epoch": 2.4246264091584373,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044103772534539505,
      "loss": 5.5878,
      "mean_token_accuracy": 0.1774356782436371,
      "num_tokens": 56225428.0,
      "step": 27745
    },
    {
      "entropy": 6.006903171539307,
      "epoch": 2.425063357511142,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004410166943629959,
      "loss": 5.6577,
      "mean_token_accuracy": 0.17280308306217193,
      "num_tokens": 56235284.0,
      "step": 27750
    },
    {
      "entropy": 5.976521348953247,
      "epoch": 2.4255003058638467,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004409956601963095,
      "loss": 5.6117,
      "mean_token_accuracy": 0.17664215415716172,
      "num_tokens": 56244444.0,
      "step": 27755
    },
    {
      "entropy": 6.016897058486938,
      "epoch": 2.4259372542165516,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004409746228457393,
      "loss": 5.6547,
      "mean_token_accuracy": 0.17274448424577712,
      "num_tokens": 56254983.0,
      "step": 27760
    },
    {
      "entropy": 6.010185766220093,
      "epoch": 2.4263742025692565,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004409535823116889,
      "loss": 5.6216,
      "mean_token_accuracy": 0.17817620784044266,
      "num_tokens": 56264307.0,
      "step": 27765
    },
    {
      "entropy": 5.916494607925415,
      "epoch": 2.426811150921961,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004409325385945619,
      "loss": 5.6679,
      "mean_token_accuracy": 0.1691889449954033,
      "num_tokens": 56274418.0,
      "step": 27770
    },
    {
      "entropy": 6.134376573562622,
      "epoch": 2.427248099274666,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004409114916947618,
      "loss": 5.7462,
      "mean_token_accuracy": 0.1622008629143238,
      "num_tokens": 56285665.0,
      "step": 27775
    },
    {
      "entropy": 6.039281797409058,
      "epoch": 2.4276850476273704,
      "grad_norm": 1.453125,
      "learning_rate": 0.00044089044161269237,
      "loss": 5.6453,
      "mean_token_accuracy": 0.17028720825910568,
      "num_tokens": 56296320.0,
      "step": 27780
    },
    {
      "entropy": 6.0147847652435305,
      "epoch": 2.4281219959800753,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004408693883487574,
      "loss": 5.7287,
      "mean_token_accuracy": 0.16610631644725798,
      "num_tokens": 56308009.0,
      "step": 27785
    },
    {
      "entropy": 5.975602102279663,
      "epoch": 2.4285589443327797,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004408483319033606,
      "loss": 5.656,
      "mean_token_accuracy": 0.16994627565145493,
      "num_tokens": 56318341.0,
      "step": 27790
    },
    {
      "entropy": 5.954961538314819,
      "epoch": 2.4289958926854847,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004408272722769059,
      "loss": 5.5755,
      "mean_token_accuracy": 0.18136348128318786,
      "num_tokens": 56328808.0,
      "step": 27795
    },
    {
      "entropy": 6.1171950340271,
      "epoch": 2.429432841038189,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004408062094697972,
      "loss": 5.6555,
      "mean_token_accuracy": 0.17305833399295806,
      "num_tokens": 56339043.0,
      "step": 27800
    },
    {
      "entropy": 5.966913223266602,
      "epoch": 2.429869789390894,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044078514348243846,
      "loss": 5.6471,
      "mean_token_accuracy": 0.17264187335968018,
      "num_tokens": 56349518.0,
      "step": 27805
    },
    {
      "entropy": 5.946867752075195,
      "epoch": 2.4303067377435985,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044076407431523394,
      "loss": 5.6024,
      "mean_token_accuracy": 0.1668172687292099,
      "num_tokens": 56358739.0,
      "step": 27810
    },
    {
      "entropy": 6.082629823684693,
      "epoch": 2.4307436860963034,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004407430019685875,
      "loss": 5.6543,
      "mean_token_accuracy": 0.17315979599952697,
      "num_tokens": 56368049.0,
      "step": 27815
    },
    {
      "entropy": 5.998693323135376,
      "epoch": 2.4311806344490083,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00044072192644290355,
      "loss": 5.6422,
      "mean_token_accuracy": 0.1718129888176918,
      "num_tokens": 56377607.0,
      "step": 27820
    },
    {
      "entropy": 5.931372451782226,
      "epoch": 2.431617582801713,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004407008477385862,
      "loss": 5.6391,
      "mean_token_accuracy": 0.1787662535905838,
      "num_tokens": 56387188.0,
      "step": 27825
    },
    {
      "entropy": 6.059385108947754,
      "epoch": 2.4320545311544177,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044067976585603964,
      "loss": 5.748,
      "mean_token_accuracy": 0.16023964136838914,
      "num_tokens": 56396389.0,
      "step": 27830
    },
    {
      "entropy": 6.0558065414428714,
      "epoch": 2.432491479507122,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004406586807956684,
      "loss": 5.6844,
      "mean_token_accuracy": 0.16940981298685073,
      "num_tokens": 56406909.0,
      "step": 27835
    },
    {
      "entropy": 6.007445478439331,
      "epoch": 2.432928427859827,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044063759255787686,
      "loss": 5.6644,
      "mean_token_accuracy": 0.16882779598236083,
      "num_tokens": 56416373.0,
      "step": 27840
    },
    {
      "entropy": 6.12060227394104,
      "epoch": 2.4333653762125316,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044061650114306954,
      "loss": 5.707,
      "mean_token_accuracy": 0.16283225566148757,
      "num_tokens": 56426759.0,
      "step": 27845
    },
    {
      "entropy": 6.045754480361938,
      "epoch": 2.4338023245652365,
      "grad_norm": 1.46875,
      "learning_rate": 0.00044059540655165097,
      "loss": 5.6861,
      "mean_token_accuracy": 0.16749546825885772,
      "num_tokens": 56436783.0,
      "step": 27850
    },
    {
      "entropy": 6.040624570846558,
      "epoch": 2.434239272917941,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044057430878402564,
      "loss": 5.6366,
      "mean_token_accuracy": 0.17278213798999786,
      "num_tokens": 56447191.0,
      "step": 27855
    },
    {
      "entropy": 5.967563962936401,
      "epoch": 2.434676221270646,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004405532078405984,
      "loss": 5.7107,
      "mean_token_accuracy": 0.16418252736330033,
      "num_tokens": 56457119.0,
      "step": 27860
    },
    {
      "entropy": 6.067411231994629,
      "epoch": 2.4351131696233503,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004405321037217739,
      "loss": 5.5861,
      "mean_token_accuracy": 0.17680654376745225,
      "num_tokens": 56467139.0,
      "step": 27865
    },
    {
      "entropy": 5.965267086029053,
      "epoch": 2.4355501179760553,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004405109964279568,
      "loss": 5.6415,
      "mean_token_accuracy": 0.17415418326854706,
      "num_tokens": 56477188.0,
      "step": 27870
    },
    {
      "entropy": 6.0387907981872555,
      "epoch": 2.43598706632876,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044048988595955216,
      "loss": 5.78,
      "mean_token_accuracy": 0.1641570433974266,
      "num_tokens": 56487594.0,
      "step": 27875
    },
    {
      "entropy": 6.122031927108765,
      "epoch": 2.4364240146814646,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004404687723169647,
      "loss": 5.6767,
      "mean_token_accuracy": 0.1735115170478821,
      "num_tokens": 56498008.0,
      "step": 27880
    },
    {
      "entropy": 5.937547254562378,
      "epoch": 2.4368609630341695,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044044765550059955,
      "loss": 5.5826,
      "mean_token_accuracy": 0.17975291311740876,
      "num_tokens": 56508409.0,
      "step": 27885
    },
    {
      "entropy": 5.958007192611694,
      "epoch": 2.437297911386874,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004404265355108616,
      "loss": 5.6662,
      "mean_token_accuracy": 0.16936213225126268,
      "num_tokens": 56519513.0,
      "step": 27890
    },
    {
      "entropy": 5.976481008529663,
      "epoch": 2.437734859739579,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004404054123481561,
      "loss": 5.5982,
      "mean_token_accuracy": 0.17300918251276015,
      "num_tokens": 56528549.0,
      "step": 27895
    },
    {
      "entropy": 5.93858494758606,
      "epoch": 2.4381718080922834,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044038428601288817,
      "loss": 5.6375,
      "mean_token_accuracy": 0.17153674215078354,
      "num_tokens": 56537859.0,
      "step": 27900
    },
    {
      "entropy": 5.987246417999268,
      "epoch": 2.4386087564449883,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044036315650546287,
      "loss": 5.6881,
      "mean_token_accuracy": 0.16419751346111297,
      "num_tokens": 56547326.0,
      "step": 27905
    },
    {
      "entropy": 6.0768290042877195,
      "epoch": 2.439045704797693,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044034202382628565,
      "loss": 5.6959,
      "mean_token_accuracy": 0.16866328716278076,
      "num_tokens": 56557460.0,
      "step": 27910
    },
    {
      "entropy": 5.908795404434204,
      "epoch": 2.4394826531503977,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004403208879757617,
      "loss": 5.6202,
      "mean_token_accuracy": 0.1790868803858757,
      "num_tokens": 56566928.0,
      "step": 27915
    },
    {
      "entropy": 5.93289589881897,
      "epoch": 2.439919601503102,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004402997489542966,
      "loss": 5.577,
      "mean_token_accuracy": 0.18507134765386582,
      "num_tokens": 56577576.0,
      "step": 27920
    },
    {
      "entropy": 6.052718210220337,
      "epoch": 2.440356549855807,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004402786067622955,
      "loss": 5.7133,
      "mean_token_accuracy": 0.16322027295827865,
      "num_tokens": 56587744.0,
      "step": 27925
    },
    {
      "entropy": 6.002193117141724,
      "epoch": 2.440793498208512,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004402574614001643,
      "loss": 5.6473,
      "mean_token_accuracy": 0.17169027179479598,
      "num_tokens": 56598272.0,
      "step": 27930
    },
    {
      "entropy": 5.961315155029297,
      "epoch": 2.4412304465612165,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044023631286830835,
      "loss": 5.5954,
      "mean_token_accuracy": 0.17470410019159316,
      "num_tokens": 56609346.0,
      "step": 27935
    },
    {
      "entropy": 6.050959873199463,
      "epoch": 2.441667394913921,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044021516116713334,
      "loss": 5.6094,
      "mean_token_accuracy": 0.1719526693224907,
      "num_tokens": 56619210.0,
      "step": 27940
    },
    {
      "entropy": 5.988172769546509,
      "epoch": 2.442104343266626,
      "grad_norm": 1.234375,
      "learning_rate": 0.000440194006297045,
      "loss": 5.6769,
      "mean_token_accuracy": 0.16551533490419387,
      "num_tokens": 56629135.0,
      "step": 27945
    },
    {
      "entropy": 6.024645709991455,
      "epoch": 2.4425412916193308,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000440172848258449,
      "loss": 5.6261,
      "mean_token_accuracy": 0.17546496391296387,
      "num_tokens": 56638900.0,
      "step": 27950
    },
    {
      "entropy": 6.021108865737915,
      "epoch": 2.4429782399720352,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044015168705175117,
      "loss": 5.6552,
      "mean_token_accuracy": 0.17264468371868133,
      "num_tokens": 56648605.0,
      "step": 27955
    },
    {
      "entropy": 5.998253011703492,
      "epoch": 2.44341518832474,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004401305226773576,
      "loss": 5.6976,
      "mean_token_accuracy": 0.17142298370599746,
      "num_tokens": 56659556.0,
      "step": 27960
    },
    {
      "entropy": 6.077078342437744,
      "epoch": 2.4438521366774446,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004401093551356739,
      "loss": 5.7636,
      "mean_token_accuracy": 0.1622890204191208,
      "num_tokens": 56669727.0,
      "step": 27965
    },
    {
      "entropy": 6.096043109893799,
      "epoch": 2.4442890850301495,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004400881844271064,
      "loss": 5.7143,
      "mean_token_accuracy": 0.16447331607341767,
      "num_tokens": 56681308.0,
      "step": 27970
    },
    {
      "entropy": 6.06246509552002,
      "epoch": 2.444726033382854,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044006701055206095,
      "loss": 5.6829,
      "mean_token_accuracy": 0.16964756399393083,
      "num_tokens": 56691089.0,
      "step": 27975
    },
    {
      "entropy": 5.965799236297608,
      "epoch": 2.445162981735559,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004400458335109437,
      "loss": 5.617,
      "mean_token_accuracy": 0.17372521460056306,
      "num_tokens": 56701599.0,
      "step": 27980
    },
    {
      "entropy": 5.89811224937439,
      "epoch": 2.4455999300882634,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044002465330416095,
      "loss": 5.5393,
      "mean_token_accuracy": 0.1803772568702698,
      "num_tokens": 56711890.0,
      "step": 27985
    },
    {
      "entropy": 5.991268730163574,
      "epoch": 2.4460368784409683,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004400034699321189,
      "loss": 5.5966,
      "mean_token_accuracy": 0.17606013715267183,
      "num_tokens": 56722003.0,
      "step": 27990
    },
    {
      "entropy": 6.003815650939941,
      "epoch": 2.4464738267936728,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004399822833952238,
      "loss": 5.5636,
      "mean_token_accuracy": 0.1702982649207115,
      "num_tokens": 56731636.0,
      "step": 27995
    },
    {
      "entropy": 5.926470804214477,
      "epoch": 2.4469107751463777,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000439961093693882,
      "loss": 5.5659,
      "mean_token_accuracy": 0.17543820589780806,
      "num_tokens": 56741973.0,
      "step": 28000
    },
    {
      "entropy": 6.0338152885437015,
      "epoch": 2.4473477234990826,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004399399008285001,
      "loss": 5.7126,
      "mean_token_accuracy": 0.16747092306613923,
      "num_tokens": 56751441.0,
      "step": 28005
    },
    {
      "entropy": 5.997309637069702,
      "epoch": 2.447784671851787,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004399187047994844,
      "loss": 5.6558,
      "mean_token_accuracy": 0.1760654106736183,
      "num_tokens": 56761156.0,
      "step": 28010
    },
    {
      "entropy": 6.024370002746582,
      "epoch": 2.448221620204492,
      "grad_norm": 1.5625,
      "learning_rate": 0.00043989750560724153,
      "loss": 5.6605,
      "mean_token_accuracy": 0.17268238514661788,
      "num_tokens": 56771414.0,
      "step": 28015
    },
    {
      "entropy": 5.903719329833985,
      "epoch": 2.4486585685571964,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004398763032521781,
      "loss": 5.5305,
      "mean_token_accuracy": 0.1855832204222679,
      "num_tokens": 56781572.0,
      "step": 28020
    },
    {
      "entropy": 6.020181941986084,
      "epoch": 2.4490955169099013,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043985509773470077,
      "loss": 5.6994,
      "mean_token_accuracy": 0.1636117197573185,
      "num_tokens": 56791299.0,
      "step": 28025
    },
    {
      "entropy": 6.016559410095215,
      "epoch": 2.449532465262606,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004398338890552163,
      "loss": 5.6166,
      "mean_token_accuracy": 0.17167791575193406,
      "num_tokens": 56802485.0,
      "step": 28030
    },
    {
      "entropy": 6.018938112258911,
      "epoch": 2.4499694136153107,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004398126772141315,
      "loss": 5.6662,
      "mean_token_accuracy": 0.16695892363786696,
      "num_tokens": 56811694.0,
      "step": 28035
    },
    {
      "entropy": 5.927751302719116,
      "epoch": 2.450406361968015,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00043979146221185313,
      "loss": 5.5765,
      "mean_token_accuracy": 0.1728035107254982,
      "num_tokens": 56821814.0,
      "step": 28040
    },
    {
      "entropy": 6.0037867546081545,
      "epoch": 2.45084331032072,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004397702440487882,
      "loss": 5.6488,
      "mean_token_accuracy": 0.1744418054819107,
      "num_tokens": 56832257.0,
      "step": 28045
    },
    {
      "entropy": 6.038809061050415,
      "epoch": 2.4512802586734246,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004397490227253436,
      "loss": 5.6592,
      "mean_token_accuracy": 0.17095112800598145,
      "num_tokens": 56841789.0,
      "step": 28050
    },
    {
      "entropy": 5.974933004379272,
      "epoch": 2.4517172070261295,
      "grad_norm": 1.625,
      "learning_rate": 0.0004397277982419265,
      "loss": 5.6287,
      "mean_token_accuracy": 0.16899534910917283,
      "num_tokens": 56851612.0,
      "step": 28055
    },
    {
      "entropy": 6.010353708267212,
      "epoch": 2.4521541553788344,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043970657059894385,
      "loss": 5.5986,
      "mean_token_accuracy": 0.17605460435152054,
      "num_tokens": 56861812.0,
      "step": 28060
    },
    {
      "entropy": 5.994963884353638,
      "epoch": 2.452591103731539,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000439685339796803,
      "loss": 5.5996,
      "mean_token_accuracy": 0.17618847340345384,
      "num_tokens": 56871873.0,
      "step": 28065
    },
    {
      "entropy": 6.001564693450928,
      "epoch": 2.453028052084244,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004396641058359109,
      "loss": 5.6753,
      "mean_token_accuracy": 0.1620353117585182,
      "num_tokens": 56883579.0,
      "step": 28070
    },
    {
      "entropy": 5.955346012115479,
      "epoch": 2.4534650004369483,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043964286871667495,
      "loss": 5.6191,
      "mean_token_accuracy": 0.17050646245479584,
      "num_tokens": 56894305.0,
      "step": 28075
    },
    {
      "entropy": 6.036136245727539,
      "epoch": 2.453901948789653,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004396216284395026,
      "loss": 5.6227,
      "mean_token_accuracy": 0.17327746152877807,
      "num_tokens": 56904393.0,
      "step": 28080
    },
    {
      "entropy": 6.019481086730957,
      "epoch": 2.4543388971423576,
      "grad_norm": 1.125,
      "learning_rate": 0.0004396003850048011,
      "loss": 5.6366,
      "mean_token_accuracy": 0.1772239997982979,
      "num_tokens": 56914600.0,
      "step": 28085
    },
    {
      "entropy": 5.982816600799561,
      "epoch": 2.4547758454950626,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004395791384129779,
      "loss": 5.7103,
      "mean_token_accuracy": 0.1768913149833679,
      "num_tokens": 56925948.0,
      "step": 28090
    },
    {
      "entropy": 5.979525566101074,
      "epoch": 2.455212793847767,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043955788866444073,
      "loss": 5.5722,
      "mean_token_accuracy": 0.1765099436044693,
      "num_tokens": 56935395.0,
      "step": 28095
    },
    {
      "entropy": 5.990030813217163,
      "epoch": 2.455649742200472,
      "grad_norm": 1.375,
      "learning_rate": 0.00043953663575959696,
      "loss": 5.5615,
      "mean_token_accuracy": 0.17788636386394502,
      "num_tokens": 56944769.0,
      "step": 28100
    },
    {
      "entropy": 5.884201097488403,
      "epoch": 2.4560866905531764,
      "grad_norm": 1.375,
      "learning_rate": 0.0004395153796988543,
      "loss": 5.5742,
      "mean_token_accuracy": 0.17701289653778077,
      "num_tokens": 56954352.0,
      "step": 28105
    },
    {
      "entropy": 5.937060928344726,
      "epoch": 2.4565236389058813,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004394941204826205,
      "loss": 5.6115,
      "mean_token_accuracy": 0.17051715701818465,
      "num_tokens": 56963759.0,
      "step": 28110
    },
    {
      "entropy": 6.010041284561157,
      "epoch": 2.4569605872585862,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043947285811130315,
      "loss": 5.5885,
      "mean_token_accuracy": 0.17390285879373552,
      "num_tokens": 56974962.0,
      "step": 28115
    },
    {
      "entropy": 5.98431544303894,
      "epoch": 2.4573975356112907,
      "grad_norm": 1.625,
      "learning_rate": 0.00043945159258531035,
      "loss": 5.6728,
      "mean_token_accuracy": 0.16596126556396484,
      "num_tokens": 56985637.0,
      "step": 28120
    },
    {
      "entropy": 5.983330821990966,
      "epoch": 2.4578344839639956,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004394303239050498,
      "loss": 5.6425,
      "mean_token_accuracy": 0.17557602375745773,
      "num_tokens": 56995779.0,
      "step": 28125
    },
    {
      "entropy": 5.951988983154297,
      "epoch": 2.4582714323167,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004394090520709294,
      "loss": 5.5843,
      "mean_token_accuracy": 0.17242403328418732,
      "num_tokens": 57005506.0,
      "step": 28130
    },
    {
      "entropy": 5.943587923049927,
      "epoch": 2.458708380669405,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043938777708335726,
      "loss": 5.7087,
      "mean_token_accuracy": 0.1656632110476494,
      "num_tokens": 57016001.0,
      "step": 28135
    },
    {
      "entropy": 5.9858238220214846,
      "epoch": 2.4591453290221095,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00043936649894274146,
      "loss": 5.6406,
      "mean_token_accuracy": 0.17126996219158172,
      "num_tokens": 57026418.0,
      "step": 28140
    },
    {
      "entropy": 6.032602453231812,
      "epoch": 2.4595822773748144,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00043934521764949,
      "loss": 5.6667,
      "mean_token_accuracy": 0.17216914594173433,
      "num_tokens": 57037170.0,
      "step": 28145
    },
    {
      "entropy": 6.018949174880982,
      "epoch": 2.460019225727519,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004393239332040113,
      "loss": 5.6048,
      "mean_token_accuracy": 0.17756972312927247,
      "num_tokens": 57046775.0,
      "step": 28150
    },
    {
      "entropy": 5.962317037582397,
      "epoch": 2.4604561740802238,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043930264560671336,
      "loss": 5.6014,
      "mean_token_accuracy": 0.17720289081335067,
      "num_tokens": 57056360.0,
      "step": 28155
    },
    {
      "entropy": 5.971512746810913,
      "epoch": 2.4608931224329282,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004392813548580045,
      "loss": 5.5936,
      "mean_token_accuracy": 0.17654372453689576,
      "num_tokens": 57066355.0,
      "step": 28160
    },
    {
      "entropy": 5.9403704643249515,
      "epoch": 2.461330070785633,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004392600609582933,
      "loss": 5.6362,
      "mean_token_accuracy": 0.1771702617406845,
      "num_tokens": 57077158.0,
      "step": 28165
    },
    {
      "entropy": 6.018401098251343,
      "epoch": 2.461767019138338,
      "grad_norm": 1.21875,
      "learning_rate": 0.000439238763907988,
      "loss": 5.6952,
      "mean_token_accuracy": 0.1724142238497734,
      "num_tokens": 57088005.0,
      "step": 28170
    },
    {
      "entropy": 6.028969430923462,
      "epoch": 2.4622039674910425,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004392174637074972,
      "loss": 5.6206,
      "mean_token_accuracy": 0.17320394068956374,
      "num_tokens": 57098399.0,
      "step": 28175
    },
    {
      "entropy": 5.9929211139678955,
      "epoch": 2.4626409158437474,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004391961603572294,
      "loss": 5.617,
      "mean_token_accuracy": 0.17210791856050492,
      "num_tokens": 57108949.0,
      "step": 28180
    },
    {
      "entropy": 5.96373872756958,
      "epoch": 2.463077864196452,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043917485385759316,
      "loss": 5.6494,
      "mean_token_accuracy": 0.17213456928730012,
      "num_tokens": 57119441.0,
      "step": 28185
    },
    {
      "entropy": 5.988091564178466,
      "epoch": 2.463514812549157,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004391535442089971,
      "loss": 5.6201,
      "mean_token_accuracy": 0.18320077359676362,
      "num_tokens": 57129507.0,
      "step": 28190
    },
    {
      "entropy": 5.987861442565918,
      "epoch": 2.4639517609018613,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004391322314118502,
      "loss": 5.6359,
      "mean_token_accuracy": 0.16836163252592087,
      "num_tokens": 57140490.0,
      "step": 28195
    },
    {
      "entropy": 5.941593170166016,
      "epoch": 2.464388709254566,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043911091546656105,
      "loss": 5.5418,
      "mean_token_accuracy": 0.18127829432487488,
      "num_tokens": 57150706.0,
      "step": 28200
    },
    {
      "entropy": 5.913116455078125,
      "epoch": 2.4648256576072707,
      "grad_norm": 1.90625,
      "learning_rate": 0.00043908959637353864,
      "loss": 5.5989,
      "mean_token_accuracy": 0.17579455971717833,
      "num_tokens": 57160259.0,
      "step": 28205
    },
    {
      "entropy": 5.996926259994507,
      "epoch": 2.4652626059599756,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043906827413319166,
      "loss": 5.5877,
      "mean_token_accuracy": 0.18287759125232697,
      "num_tokens": 57171831.0,
      "step": 28210
    },
    {
      "entropy": 6.023711156845093,
      "epoch": 2.46569955431268,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004390469487459292,
      "loss": 5.7053,
      "mean_token_accuracy": 0.16661300808191298,
      "num_tokens": 57182172.0,
      "step": 28215
    },
    {
      "entropy": 5.948973321914673,
      "epoch": 2.466136502665385,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043902562021216033,
      "loss": 5.5942,
      "mean_token_accuracy": 0.18057496547698976,
      "num_tokens": 57191633.0,
      "step": 28220
    },
    {
      "entropy": 6.053310823440552,
      "epoch": 2.46657345101809,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004390042885322942,
      "loss": 5.619,
      "mean_token_accuracy": 0.17669338285923003,
      "num_tokens": 57200953.0,
      "step": 28225
    },
    {
      "entropy": 5.9851995468139645,
      "epoch": 2.4670103993707944,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043898295370673986,
      "loss": 5.6899,
      "mean_token_accuracy": 0.16727260500192642,
      "num_tokens": 57210893.0,
      "step": 28230
    },
    {
      "entropy": 5.92959156036377,
      "epoch": 2.4674473477234993,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043896161573590646,
      "loss": 5.6152,
      "mean_token_accuracy": 0.1800549790263176,
      "num_tokens": 57221458.0,
      "step": 28235
    },
    {
      "entropy": 6.033135271072387,
      "epoch": 2.4678842960762037,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004389402746202034,
      "loss": 5.675,
      "mean_token_accuracy": 0.16674021631479263,
      "num_tokens": 57231450.0,
      "step": 28240
    },
    {
      "entropy": 6.047430515289307,
      "epoch": 2.4683212444289087,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043891893036004,
      "loss": 5.6397,
      "mean_token_accuracy": 0.17231406569480895,
      "num_tokens": 57242820.0,
      "step": 28245
    },
    {
      "entropy": 6.066785860061645,
      "epoch": 2.468758192781613,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004388975829558256,
      "loss": 5.6783,
      "mean_token_accuracy": 0.16788724809885025,
      "num_tokens": 57253465.0,
      "step": 28250
    },
    {
      "entropy": 5.913343620300293,
      "epoch": 2.469195141134318,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043887623240796965,
      "loss": 5.6276,
      "mean_token_accuracy": 0.1743263393640518,
      "num_tokens": 57263521.0,
      "step": 28255
    },
    {
      "entropy": 5.986600780487061,
      "epoch": 2.4696320894870225,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004388548787168817,
      "loss": 5.6268,
      "mean_token_accuracy": 0.16649169772863387,
      "num_tokens": 57273568.0,
      "step": 28260
    },
    {
      "entropy": 6.040337133407593,
      "epoch": 2.4700690378397274,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004388335218829713,
      "loss": 5.6966,
      "mean_token_accuracy": 0.17046983242034913,
      "num_tokens": 57283960.0,
      "step": 28265
    },
    {
      "entropy": 5.954203796386719,
      "epoch": 2.470505986192432,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004388121619066482,
      "loss": 5.642,
      "mean_token_accuracy": 0.17315568774938583,
      "num_tokens": 57294672.0,
      "step": 28270
    },
    {
      "entropy": 6.022079706192017,
      "epoch": 2.470942934545137,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004387907987883219,
      "loss": 5.6326,
      "mean_token_accuracy": 0.17701055705547333,
      "num_tokens": 57305594.0,
      "step": 28275
    },
    {
      "entropy": 6.004144620895386,
      "epoch": 2.4713798828978417,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004387694325284023,
      "loss": 5.5873,
      "mean_token_accuracy": 0.1838662639260292,
      "num_tokens": 57315250.0,
      "step": 28280
    },
    {
      "entropy": 6.017524147033692,
      "epoch": 2.471816831250546,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004387480631272991,
      "loss": 5.6502,
      "mean_token_accuracy": 0.16906920075416565,
      "num_tokens": 57326750.0,
      "step": 28285
    },
    {
      "entropy": 6.0033032417297365,
      "epoch": 2.4722537796032507,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004387266905854223,
      "loss": 5.6164,
      "mean_token_accuracy": 0.17399057000875473,
      "num_tokens": 57336151.0,
      "step": 28290
    },
    {
      "entropy": 5.962000751495362,
      "epoch": 2.4726907279559556,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004387053149031819,
      "loss": 5.598,
      "mean_token_accuracy": 0.17377203851938247,
      "num_tokens": 57346761.0,
      "step": 28295
    },
    {
      "entropy": 5.995953750610352,
      "epoch": 2.4731276763086605,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004386839360809876,
      "loss": 5.6625,
      "mean_token_accuracy": 0.17435790002346038,
      "num_tokens": 57356654.0,
      "step": 28300
    },
    {
      "entropy": 6.022018718719482,
      "epoch": 2.473564624661365,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004386625541192497,
      "loss": 5.6737,
      "mean_token_accuracy": 0.16744315922260283,
      "num_tokens": 57367888.0,
      "step": 28305
    },
    {
      "entropy": 5.975557947158814,
      "epoch": 2.47400157301407,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043864116901837826,
      "loss": 5.5943,
      "mean_token_accuracy": 0.18178453296422958,
      "num_tokens": 57377167.0,
      "step": 28310
    },
    {
      "entropy": 5.911859226226807,
      "epoch": 2.4744385213667743,
      "grad_norm": 1.65625,
      "learning_rate": 0.00043861978077878343,
      "loss": 5.6022,
      "mean_token_accuracy": 0.1760938674211502,
      "num_tokens": 57389076.0,
      "step": 28315
    },
    {
      "entropy": 5.99954743385315,
      "epoch": 2.4748754697194792,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004385983894008755,
      "loss": 5.6524,
      "mean_token_accuracy": 0.17480300813913346,
      "num_tokens": 57401136.0,
      "step": 28320
    },
    {
      "entropy": 6.008118343353272,
      "epoch": 2.4753124180721837,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00043857699488506467,
      "loss": 5.6041,
      "mean_token_accuracy": 0.17411221861839293,
      "num_tokens": 57410548.0,
      "step": 28325
    },
    {
      "entropy": 6.006905555725098,
      "epoch": 2.4757493664248886,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004385555972317615,
      "loss": 5.651,
      "mean_token_accuracy": 0.1698381319642067,
      "num_tokens": 57420371.0,
      "step": 28330
    },
    {
      "entropy": 6.023439264297485,
      "epoch": 2.476186314777593,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043853419644137616,
      "loss": 5.6644,
      "mean_token_accuracy": 0.1714356392621994,
      "num_tokens": 57430079.0,
      "step": 28335
    },
    {
      "entropy": 5.977290153503418,
      "epoch": 2.476623263130298,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00043851279251431937,
      "loss": 5.627,
      "mean_token_accuracy": 0.17431796044111253,
      "num_tokens": 57440763.0,
      "step": 28340
    },
    {
      "entropy": 6.093556928634643,
      "epoch": 2.4770602114830025,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00043849138545100136,
      "loss": 5.7019,
      "mean_token_accuracy": 0.16683005541563034,
      "num_tokens": 57451008.0,
      "step": 28345
    },
    {
      "entropy": 5.9849354267120365,
      "epoch": 2.4774971598357074,
      "grad_norm": 1.703125,
      "learning_rate": 0.00043846997525183305,
      "loss": 5.6835,
      "mean_token_accuracy": 0.16616218239068986,
      "num_tokens": 57460852.0,
      "step": 28350
    },
    {
      "entropy": 5.981824970245361,
      "epoch": 2.4779341081884123,
      "grad_norm": 1.546875,
      "learning_rate": 0.00043844856191722486,
      "loss": 5.6509,
      "mean_token_accuracy": 0.16692606657743453,
      "num_tokens": 57470027.0,
      "step": 28355
    },
    {
      "entropy": 6.004398536682129,
      "epoch": 2.478371056541117,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043842714544758764,
      "loss": 5.5591,
      "mean_token_accuracy": 0.17595591098070146,
      "num_tokens": 57479846.0,
      "step": 28360
    },
    {
      "entropy": 6.005037450790406,
      "epoch": 2.4788080048938217,
      "grad_norm": 2.703125,
      "learning_rate": 0.00043840572584333216,
      "loss": 5.7047,
      "mean_token_accuracy": 0.16821253746747972,
      "num_tokens": 57490569.0,
      "step": 28365
    },
    {
      "entropy": 5.969551658630371,
      "epoch": 2.479244953246526,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004383843031048692,
      "loss": 5.6465,
      "mean_token_accuracy": 0.1697237804532051,
      "num_tokens": 57501048.0,
      "step": 28370
    },
    {
      "entropy": 5.965248966217041,
      "epoch": 2.479681901599231,
      "grad_norm": 2.0625,
      "learning_rate": 0.00043836287723260977,
      "loss": 5.6914,
      "mean_token_accuracy": 0.16196770071983338,
      "num_tokens": 57511510.0,
      "step": 28375
    },
    {
      "entropy": 6.000234746932984,
      "epoch": 2.4801188499519355,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00043834144822696463,
      "loss": 5.6457,
      "mean_token_accuracy": 0.1731523022055626,
      "num_tokens": 57522229.0,
      "step": 28380
    },
    {
      "entropy": 6.03197283744812,
      "epoch": 2.4805557983046405,
      "grad_norm": 1.546875,
      "learning_rate": 0.000438320016088345,
      "loss": 5.5978,
      "mean_token_accuracy": 0.1698383942246437,
      "num_tokens": 57532618.0,
      "step": 28385
    },
    {
      "entropy": 5.970089054107666,
      "epoch": 2.480992746657345,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004382985808171619,
      "loss": 5.6428,
      "mean_token_accuracy": 0.17189929634332657,
      "num_tokens": 57542698.0,
      "step": 28390
    },
    {
      "entropy": 5.900430202484131,
      "epoch": 2.48142969501005,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043827714241382644,
      "loss": 5.5286,
      "mean_token_accuracy": 0.17623188942670823,
      "num_tokens": 57552556.0,
      "step": 28395
    },
    {
      "entropy": 6.08671293258667,
      "epoch": 2.4818666433627543,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004382557008787498,
      "loss": 5.6784,
      "mean_token_accuracy": 0.17108576893806457,
      "num_tokens": 57562830.0,
      "step": 28400
    },
    {
      "entropy": 6.056606245040894,
      "epoch": 2.4823035917154592,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004382342562123433,
      "loss": 5.6149,
      "mean_token_accuracy": 0.17269092053174973,
      "num_tokens": 57572584.0,
      "step": 28405
    },
    {
      "entropy": 5.902246761322021,
      "epoch": 2.482740540068164,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004382128084150182,
      "loss": 5.6037,
      "mean_token_accuracy": 0.1712510883808136,
      "num_tokens": 57582571.0,
      "step": 28410
    },
    {
      "entropy": 5.951892614364624,
      "epoch": 2.4831774884208686,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00043819135748718595,
      "loss": 5.6631,
      "mean_token_accuracy": 0.17755152881145478,
      "num_tokens": 57593267.0,
      "step": 28415
    },
    {
      "entropy": 5.980672979354859,
      "epoch": 2.4836144367735735,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043816990342925797,
      "loss": 5.6615,
      "mean_token_accuracy": 0.17158745974302292,
      "num_tokens": 57603708.0,
      "step": 28420
    },
    {
      "entropy": 6.0133928775787355,
      "epoch": 2.484051385126278,
      "grad_norm": 1.46875,
      "learning_rate": 0.00043814844624164576,
      "loss": 5.7076,
      "mean_token_accuracy": 0.16409794390201568,
      "num_tokens": 57613523.0,
      "step": 28425
    },
    {
      "entropy": 6.014401292800903,
      "epoch": 2.484488333478983,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004381269859247607,
      "loss": 5.6294,
      "mean_token_accuracy": 0.1740129381418228,
      "num_tokens": 57624181.0,
      "step": 28430
    },
    {
      "entropy": 5.97234525680542,
      "epoch": 2.4849252818316874,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043810552247901475,
      "loss": 5.5736,
      "mean_token_accuracy": 0.17146258801221848,
      "num_tokens": 57633674.0,
      "step": 28435
    },
    {
      "entropy": 6.016128969192505,
      "epoch": 2.4853622301843923,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043808405590481934,
      "loss": 5.6462,
      "mean_token_accuracy": 0.17299685180187224,
      "num_tokens": 57644723.0,
      "step": 28440
    },
    {
      "entropy": 5.98001708984375,
      "epoch": 2.4857991785370968,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004380625862025863,
      "loss": 5.6027,
      "mean_token_accuracy": 0.17238631397485732,
      "num_tokens": 57654274.0,
      "step": 28445
    },
    {
      "entropy": 5.942051410675049,
      "epoch": 2.4862361268898017,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004380411133727274,
      "loss": 5.5758,
      "mean_token_accuracy": 0.1725737124681473,
      "num_tokens": 57664572.0,
      "step": 28450
    },
    {
      "entropy": 6.010201501846313,
      "epoch": 2.486673075242506,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043801963741565454,
      "loss": 5.6926,
      "mean_token_accuracy": 0.16342156529426574,
      "num_tokens": 57674048.0,
      "step": 28455
    },
    {
      "entropy": 6.05449743270874,
      "epoch": 2.487110023595211,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004379981583317796,
      "loss": 5.657,
      "mean_token_accuracy": 0.1672518879175186,
      "num_tokens": 57685488.0,
      "step": 28460
    },
    {
      "entropy": 6.006172513961792,
      "epoch": 2.487546971947916,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043797667612151454,
      "loss": 5.658,
      "mean_token_accuracy": 0.174440698325634,
      "num_tokens": 57696044.0,
      "step": 28465
    },
    {
      "entropy": 5.960581541061401,
      "epoch": 2.4879839203006204,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043795519078527144,
      "loss": 5.6117,
      "mean_token_accuracy": 0.1818651333451271,
      "num_tokens": 57705702.0,
      "step": 28470
    },
    {
      "entropy": 6.008641386032105,
      "epoch": 2.4884208686533253,
      "grad_norm": 1.671875,
      "learning_rate": 0.00043793370232346246,
      "loss": 5.6919,
      "mean_token_accuracy": 0.16951791495084761,
      "num_tokens": 57715628.0,
      "step": 28475
    },
    {
      "entropy": 6.030720520019531,
      "epoch": 2.48885781700603,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00043791221073649964,
      "loss": 5.6437,
      "mean_token_accuracy": 0.17061752527952195,
      "num_tokens": 57725319.0,
      "step": 28480
    },
    {
      "entropy": 6.0224357604980465,
      "epoch": 2.4892947653587347,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043789071602479525,
      "loss": 5.6628,
      "mean_token_accuracy": 0.16748576760292053,
      "num_tokens": 57736483.0,
      "step": 28485
    },
    {
      "entropy": 6.0158308982849125,
      "epoch": 2.489731713711439,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004378692181887616,
      "loss": 5.6533,
      "mean_token_accuracy": 0.16433672457933426,
      "num_tokens": 57746667.0,
      "step": 28490
    },
    {
      "entropy": 5.913653469085693,
      "epoch": 2.490168662064144,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00043784771722881095,
      "loss": 5.5274,
      "mean_token_accuracy": 0.18881852477788924,
      "num_tokens": 57755971.0,
      "step": 28495
    },
    {
      "entropy": 5.903770971298218,
      "epoch": 2.4906056104168486,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004378262131453558,
      "loss": 5.594,
      "mean_token_accuracy": 0.18210698366165162,
      "num_tokens": 57765858.0,
      "step": 28500
    },
    {
      "entropy": 5.975754690170288,
      "epoch": 2.4910425587695535,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004378047059388086,
      "loss": 5.6115,
      "mean_token_accuracy": 0.17565581649541856,
      "num_tokens": 57775154.0,
      "step": 28505
    },
    {
      "entropy": 5.996017551422119,
      "epoch": 2.491479507122258,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004377831956095817,
      "loss": 5.6023,
      "mean_token_accuracy": 0.17599532455205918,
      "num_tokens": 57785416.0,
      "step": 28510
    },
    {
      "entropy": 5.97162356376648,
      "epoch": 2.491916455474963,
      "grad_norm": 1.265625,
      "learning_rate": 0.000437761682158088,
      "loss": 5.6149,
      "mean_token_accuracy": 0.17283039093017577,
      "num_tokens": 57795483.0,
      "step": 28515
    },
    {
      "entropy": 6.039251279830933,
      "epoch": 2.492353403827668,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043774016558473976,
      "loss": 5.6393,
      "mean_token_accuracy": 0.17336104363203048,
      "num_tokens": 57805772.0,
      "step": 28520
    },
    {
      "entropy": 6.021987533569336,
      "epoch": 2.4927903521803723,
      "grad_norm": 2.25,
      "learning_rate": 0.00043771864588994995,
      "loss": 5.6522,
      "mean_token_accuracy": 0.16953013986349105,
      "num_tokens": 57815514.0,
      "step": 28525
    },
    {
      "entropy": 5.9368823051452635,
      "epoch": 2.493227300533077,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004376971230741313,
      "loss": 5.5971,
      "mean_token_accuracy": 0.17565522640943526,
      "num_tokens": 57826312.0,
      "step": 28530
    },
    {
      "entropy": 5.966194009780883,
      "epoch": 2.4936642488857816,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004376755971376965,
      "loss": 5.683,
      "mean_token_accuracy": 0.16923390179872513,
      "num_tokens": 57836718.0,
      "step": 28535
    },
    {
      "entropy": 6.024297571182251,
      "epoch": 2.4941011972384866,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004376540680810585,
      "loss": 5.5997,
      "mean_token_accuracy": 0.1765248328447342,
      "num_tokens": 57846833.0,
      "step": 28540
    },
    {
      "entropy": 6.04984941482544,
      "epoch": 2.494538145591191,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004376325359046304,
      "loss": 5.6946,
      "mean_token_accuracy": 0.16629765331745147,
      "num_tokens": 57856716.0,
      "step": 28545
    },
    {
      "entropy": 6.010496139526367,
      "epoch": 2.494975093943896,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00043761100060882486,
      "loss": 5.6799,
      "mean_token_accuracy": 0.16717909127473832,
      "num_tokens": 57867079.0,
      "step": 28550
    },
    {
      "entropy": 6.002038431167603,
      "epoch": 2.4954120422966004,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004375894621940552,
      "loss": 5.6071,
      "mean_token_accuracy": 0.1737575799226761,
      "num_tokens": 57876593.0,
      "step": 28555
    },
    {
      "entropy": 5.954208326339722,
      "epoch": 2.4958489906493053,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004375679206607344,
      "loss": 5.622,
      "mean_token_accuracy": 0.178703773021698,
      "num_tokens": 57886834.0,
      "step": 28560
    },
    {
      "entropy": 5.913920164108276,
      "epoch": 2.49628593900201,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004375463760092758,
      "loss": 5.5637,
      "mean_token_accuracy": 0.180254927277565,
      "num_tokens": 57896766.0,
      "step": 28565
    },
    {
      "entropy": 6.003846502304077,
      "epoch": 2.4967228873547147,
      "grad_norm": 1.375,
      "learning_rate": 0.0004375248282400925,
      "loss": 5.6155,
      "mean_token_accuracy": 0.1747686192393303,
      "num_tokens": 57907259.0,
      "step": 28570
    },
    {
      "entropy": 6.052454280853271,
      "epoch": 2.4971598357074196,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004375032773535977,
      "loss": 5.7428,
      "mean_token_accuracy": 0.16533606201410295,
      "num_tokens": 57917940.0,
      "step": 28575
    },
    {
      "entropy": 6.047459316253662,
      "epoch": 2.497596784060124,
      "grad_norm": 1.265625,
      "learning_rate": 0.000437481723350205,
      "loss": 5.6036,
      "mean_token_accuracy": 0.17271875590085983,
      "num_tokens": 57927256.0,
      "step": 28580
    },
    {
      "entropy": 5.970742845535279,
      "epoch": 2.4980337324128286,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043746016623032776,
      "loss": 5.5863,
      "mean_token_accuracy": 0.1759258523583412,
      "num_tokens": 57936718.0,
      "step": 28585
    },
    {
      "entropy": 6.0256139755249025,
      "epoch": 2.4984706807655335,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004374386059943794,
      "loss": 5.6462,
      "mean_token_accuracy": 0.16761403381824494,
      "num_tokens": 57947548.0,
      "step": 28590
    },
    {
      "entropy": 5.978464603424072,
      "epoch": 2.4989076291182384,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004374170426427734,
      "loss": 5.569,
      "mean_token_accuracy": 0.17800707668066024,
      "num_tokens": 57957598.0,
      "step": 28595
    },
    {
      "entropy": 5.9802226543426515,
      "epoch": 2.499344577470943,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043739547617592344,
      "loss": 5.7031,
      "mean_token_accuracy": 0.16666433066129685,
      "num_tokens": 57967790.0,
      "step": 28600
    },
    {
      "entropy": 5.993328475952149,
      "epoch": 2.4997815258236478,
      "grad_norm": 1.40625,
      "learning_rate": 0.00043737390659424316,
      "loss": 5.6385,
      "mean_token_accuracy": 0.1776592254638672,
      "num_tokens": 57977904.0,
      "step": 28605
    },
    {
      "entropy": 5.949097967147827,
      "epoch": 2.5002184741763522,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043735233389814623,
      "loss": 5.4751,
      "mean_token_accuracy": 0.18423134535551072,
      "num_tokens": 57987882.0,
      "step": 28610
    },
    {
      "entropy": 5.939853382110596,
      "epoch": 2.500655422529057,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043733075808804643,
      "loss": 5.6263,
      "mean_token_accuracy": 0.17872802019119263,
      "num_tokens": 57997376.0,
      "step": 28615
    },
    {
      "entropy": 5.898761653900147,
      "epoch": 2.5010923708817616,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043730917916435776,
      "loss": 5.5576,
      "mean_token_accuracy": 0.18288896530866622,
      "num_tokens": 58008115.0,
      "step": 28620
    },
    {
      "entropy": 6.037384271621704,
      "epoch": 2.5015293192344665,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004372875971274939,
      "loss": 5.6943,
      "mean_token_accuracy": 0.16299022734165192,
      "num_tokens": 58017922.0,
      "step": 28625
    },
    {
      "entropy": 5.997530126571656,
      "epoch": 2.5019662675871714,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004372660119778689,
      "loss": 5.6479,
      "mean_token_accuracy": 0.17369375228881836,
      "num_tokens": 58028208.0,
      "step": 28630
    },
    {
      "entropy": 5.997331237792968,
      "epoch": 2.502403215939876,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004372444237158968,
      "loss": 5.6555,
      "mean_token_accuracy": 0.17137689292430877,
      "num_tokens": 58037237.0,
      "step": 28635
    },
    {
      "entropy": 5.97856559753418,
      "epoch": 2.5028401642925804,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004372228323419916,
      "loss": 5.6123,
      "mean_token_accuracy": 0.1790345698595047,
      "num_tokens": 58048083.0,
      "step": 28640
    },
    {
      "entropy": 6.0352319240570065,
      "epoch": 2.5032771126452853,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004372012378565674,
      "loss": 5.6648,
      "mean_token_accuracy": 0.1713245540857315,
      "num_tokens": 58058593.0,
      "step": 28645
    },
    {
      "entropy": 5.982099771499634,
      "epoch": 2.50371406099799,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043717964026003856,
      "loss": 5.6094,
      "mean_token_accuracy": 0.18245284855365754,
      "num_tokens": 58068996.0,
      "step": 28650
    },
    {
      "entropy": 5.980434894561768,
      "epoch": 2.5041510093506947,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004371580395528192,
      "loss": 5.6087,
      "mean_token_accuracy": 0.17165557295084,
      "num_tokens": 58078546.0,
      "step": 28655
    },
    {
      "entropy": 6.0645339012146,
      "epoch": 2.5045879577033996,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043713643573532377,
      "loss": 5.7053,
      "mean_token_accuracy": 0.15952103585004807,
      "num_tokens": 58087824.0,
      "step": 28660
    },
    {
      "entropy": 6.006406545639038,
      "epoch": 2.505024906056104,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043711482880796643,
      "loss": 5.6619,
      "mean_token_accuracy": 0.17847764641046523,
      "num_tokens": 58097940.0,
      "step": 28665
    },
    {
      "entropy": 6.046171188354492,
      "epoch": 2.505461854408809,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004370932187711618,
      "loss": 5.6615,
      "mean_token_accuracy": 0.16259069591760636,
      "num_tokens": 58109274.0,
      "step": 28670
    },
    {
      "entropy": 6.0193727016448975,
      "epoch": 2.5058988027615134,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004370716056253242,
      "loss": 5.626,
      "mean_token_accuracy": 0.1686664342880249,
      "num_tokens": 58119061.0,
      "step": 28675
    },
    {
      "entropy": 5.966940832138062,
      "epoch": 2.5063357511142184,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043704998937086844,
      "loss": 5.6508,
      "mean_token_accuracy": 0.17158881574869156,
      "num_tokens": 58128949.0,
      "step": 28680
    },
    {
      "entropy": 5.940167093276978,
      "epoch": 2.5067726994669233,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004370283700082088,
      "loss": 5.5261,
      "mean_token_accuracy": 0.18095172494649886,
      "num_tokens": 58139807.0,
      "step": 28685
    },
    {
      "entropy": 5.965424919128418,
      "epoch": 2.5072096478196277,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004370067475377602,
      "loss": 5.6637,
      "mean_token_accuracy": 0.1696872115135193,
      "num_tokens": 58150878.0,
      "step": 28690
    },
    {
      "entropy": 5.91997389793396,
      "epoch": 2.507646596172332,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004369851219599373,
      "loss": 5.639,
      "mean_token_accuracy": 0.1776092156767845,
      "num_tokens": 58160805.0,
      "step": 28695
    },
    {
      "entropy": 6.045497274398803,
      "epoch": 2.508083544525037,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00043696349327515484,
      "loss": 5.5696,
      "mean_token_accuracy": 0.17523164600133895,
      "num_tokens": 58170298.0,
      "step": 28700
    },
    {
      "entropy": 5.919219064712524,
      "epoch": 2.508520492877742,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043694186148382773,
      "loss": 5.5987,
      "mean_token_accuracy": 0.18149090856313704,
      "num_tokens": 58180497.0,
      "step": 28705
    },
    {
      "entropy": 6.01473445892334,
      "epoch": 2.5089574412304465,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043692022658637085,
      "loss": 5.6623,
      "mean_token_accuracy": 0.17179598808288574,
      "num_tokens": 58190871.0,
      "step": 28710
    },
    {
      "entropy": 5.962814903259277,
      "epoch": 2.5093943895831514,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00043689858858319923,
      "loss": 5.6189,
      "mean_token_accuracy": 0.17653909027576448,
      "num_tokens": 58201654.0,
      "step": 28715
    },
    {
      "entropy": 5.997442674636841,
      "epoch": 2.509831337935856,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043687694747472785,
      "loss": 5.6774,
      "mean_token_accuracy": 0.16882194876670836,
      "num_tokens": 58211647.0,
      "step": 28720
    },
    {
      "entropy": 5.934496450424194,
      "epoch": 2.510268286288561,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043685530326137173,
      "loss": 5.6138,
      "mean_token_accuracy": 0.17824828028678893,
      "num_tokens": 58221945.0,
      "step": 28725
    },
    {
      "entropy": 5.976836252212524,
      "epoch": 2.5107052346412653,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043683365594354614,
      "loss": 5.6455,
      "mean_token_accuracy": 0.1717928320169449,
      "num_tokens": 58231702.0,
      "step": 28730
    },
    {
      "entropy": 6.016206359863281,
      "epoch": 2.51114218299397,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004368120055216662,
      "loss": 5.6855,
      "mean_token_accuracy": 0.1669629320502281,
      "num_tokens": 58242822.0,
      "step": 28735
    },
    {
      "entropy": 5.985897636413574,
      "epoch": 2.511579131346675,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004367903519961472,
      "loss": 5.5818,
      "mean_token_accuracy": 0.17776113599538804,
      "num_tokens": 58253221.0,
      "step": 28740
    },
    {
      "entropy": 5.974702548980713,
      "epoch": 2.5120160796993796,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004367686953674045,
      "loss": 5.5917,
      "mean_token_accuracy": 0.17786476463079454,
      "num_tokens": 58263739.0,
      "step": 28745
    },
    {
      "entropy": 5.943215084075928,
      "epoch": 2.512453028052084,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043674703563585345,
      "loss": 5.5839,
      "mean_token_accuracy": 0.1823539540171623,
      "num_tokens": 58273526.0,
      "step": 28750
    },
    {
      "entropy": 5.914198970794677,
      "epoch": 2.512889976404789,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004367253728019095,
      "loss": 5.6282,
      "mean_token_accuracy": 0.1774004280567169,
      "num_tokens": 58283373.0,
      "step": 28755
    },
    {
      "entropy": 6.003595399856567,
      "epoch": 2.513326924757494,
      "grad_norm": 2.078125,
      "learning_rate": 0.00043670370686598814,
      "loss": 5.6624,
      "mean_token_accuracy": 0.1755555108189583,
      "num_tokens": 58293365.0,
      "step": 28760
    },
    {
      "entropy": 5.966490411758423,
      "epoch": 2.5137638731101983,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043668203782850495,
      "loss": 5.6192,
      "mean_token_accuracy": 0.18070915788412095,
      "num_tokens": 58303750.0,
      "step": 28765
    },
    {
      "entropy": 6.0271117210388185,
      "epoch": 2.5142008214629032,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004366603656898755,
      "loss": 5.6257,
      "mean_token_accuracy": 0.17847390174865724,
      "num_tokens": 58313209.0,
      "step": 28770
    },
    {
      "entropy": 5.988869380950928,
      "epoch": 2.5146377698156077,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00043663869045051565,
      "loss": 5.6417,
      "mean_token_accuracy": 0.17619580775499344,
      "num_tokens": 58322641.0,
      "step": 28775
    },
    {
      "entropy": 5.963724279403687,
      "epoch": 2.5150747181683126,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043661701211084096,
      "loss": 5.5168,
      "mean_token_accuracy": 0.18464192152023315,
      "num_tokens": 58331865.0,
      "step": 28780
    },
    {
      "entropy": 5.967606496810913,
      "epoch": 2.515511666521017,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043659533067126735,
      "loss": 5.6662,
      "mean_token_accuracy": 0.17112652212381363,
      "num_tokens": 58341188.0,
      "step": 28785
    },
    {
      "entropy": 5.96037974357605,
      "epoch": 2.515948614873722,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004365736461322105,
      "loss": 5.6856,
      "mean_token_accuracy": 0.1683497056365013,
      "num_tokens": 58351856.0,
      "step": 28790
    },
    {
      "entropy": 6.036629629135132,
      "epoch": 2.5163855632264265,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004365519584940866,
      "loss": 5.6684,
      "mean_token_accuracy": 0.1731512054800987,
      "num_tokens": 58363066.0,
      "step": 28795
    },
    {
      "entropy": 6.060342693328858,
      "epoch": 2.5168225115791314,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004365302677573114,
      "loss": 5.6457,
      "mean_token_accuracy": 0.16644304096698762,
      "num_tokens": 58374646.0,
      "step": 28800
    },
    {
      "entropy": 5.963460445404053,
      "epoch": 2.517259459931836,
      "grad_norm": 1.125,
      "learning_rate": 0.000436508573922301,
      "loss": 5.6209,
      "mean_token_accuracy": 0.16897981315851213,
      "num_tokens": 58385192.0,
      "step": 28805
    },
    {
      "entropy": 6.024902868270874,
      "epoch": 2.5176964082845408,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043648687698947153,
      "loss": 5.6362,
      "mean_token_accuracy": 0.17826849222183228,
      "num_tokens": 58394237.0,
      "step": 28810
    },
    {
      "entropy": 5.905282115936279,
      "epoch": 2.5181333566372457,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043646517695923923,
      "loss": 5.5973,
      "mean_token_accuracy": 0.1762307569384575,
      "num_tokens": 58404229.0,
      "step": 28815
    },
    {
      "entropy": 5.934435319900513,
      "epoch": 2.51857030498995,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004364434738320201,
      "loss": 5.6476,
      "mean_token_accuracy": 0.1701701819896698,
      "num_tokens": 58414706.0,
      "step": 28820
    },
    {
      "entropy": 6.059801292419434,
      "epoch": 2.5190072533426546,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043642176760823066,
      "loss": 5.6571,
      "mean_token_accuracy": 0.16879345029592513,
      "num_tokens": 58425699.0,
      "step": 28825
    },
    {
      "entropy": 6.049916172027588,
      "epoch": 2.5194442016953595,
      "grad_norm": 1.34375,
      "learning_rate": 0.000436400058288287,
      "loss": 5.6645,
      "mean_token_accuracy": 0.16226862370967865,
      "num_tokens": 58435923.0,
      "step": 28830
    },
    {
      "entropy": 5.937602806091308,
      "epoch": 2.5198811500480645,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004363783458726058,
      "loss": 5.6239,
      "mean_token_accuracy": 0.1761593446135521,
      "num_tokens": 58445569.0,
      "step": 28835
    },
    {
      "entropy": 6.013532829284668,
      "epoch": 2.520318098400769,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043635663036160324,
      "loss": 5.6166,
      "mean_token_accuracy": 0.17860966175794601,
      "num_tokens": 58456686.0,
      "step": 28840
    },
    {
      "entropy": 5.919421863555908,
      "epoch": 2.520755046753474,
      "grad_norm": 1.296875,
      "learning_rate": 0.000436334911755696,
      "loss": 5.5468,
      "mean_token_accuracy": 0.18152692914009094,
      "num_tokens": 58466148.0,
      "step": 28845
    },
    {
      "entropy": 5.982247543334961,
      "epoch": 2.5211919951061783,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004363131900553006,
      "loss": 5.6005,
      "mean_token_accuracy": 0.17178924083709718,
      "num_tokens": 58475161.0,
      "step": 28850
    },
    {
      "entropy": 5.968374395370484,
      "epoch": 2.521628943458883,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004362914652608338,
      "loss": 5.558,
      "mean_token_accuracy": 0.17729087322950363,
      "num_tokens": 58483631.0,
      "step": 28855
    },
    {
      "entropy": 5.873214340209961,
      "epoch": 2.5220658918115877,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004362697373727121,
      "loss": 5.5745,
      "mean_token_accuracy": 0.17598396986722947,
      "num_tokens": 58493063.0,
      "step": 28860
    },
    {
      "entropy": 5.9819293975830075,
      "epoch": 2.5225028401642926,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004362480063913524,
      "loss": 5.721,
      "mean_token_accuracy": 0.16623589247465134,
      "num_tokens": 58503658.0,
      "step": 28865
    },
    {
      "entropy": 6.016240358352661,
      "epoch": 2.5229397885169975,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00043622627231717126,
      "loss": 5.5899,
      "mean_token_accuracy": 0.17875068187713622,
      "num_tokens": 58515083.0,
      "step": 28870
    },
    {
      "entropy": 6.052314949035645,
      "epoch": 2.523376736869702,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004362045351505859,
      "loss": 5.6452,
      "mean_token_accuracy": 0.17229576259851456,
      "num_tokens": 58524984.0,
      "step": 28875
    },
    {
      "entropy": 6.06729211807251,
      "epoch": 2.5238136852224065,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004361827948920131,
      "loss": 5.7455,
      "mean_token_accuracy": 0.1676253482699394,
      "num_tokens": 58535651.0,
      "step": 28880
    },
    {
      "entropy": 5.983251619338989,
      "epoch": 2.5242506335751114,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043616105154186974,
      "loss": 5.6965,
      "mean_token_accuracy": 0.1705407604575157,
      "num_tokens": 58546380.0,
      "step": 28885
    },
    {
      "entropy": 5.926673460006714,
      "epoch": 2.5246875819278163,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043613930510057305,
      "loss": 5.5866,
      "mean_token_accuracy": 0.18434925824403764,
      "num_tokens": 58555639.0,
      "step": 28890
    },
    {
      "entropy": 5.964972972869873,
      "epoch": 2.5251245302805208,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043611755556853993,
      "loss": 5.6281,
      "mean_token_accuracy": 0.17637918293476104,
      "num_tokens": 58564948.0,
      "step": 28895
    },
    {
      "entropy": 6.0194010734558105,
      "epoch": 2.5255614786332257,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004360958029461878,
      "loss": 5.6995,
      "mean_token_accuracy": 0.16772203743457795,
      "num_tokens": 58574146.0,
      "step": 28900
    },
    {
      "entropy": 6.052498817443848,
      "epoch": 2.52599842698593,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004360740472339336,
      "loss": 5.6126,
      "mean_token_accuracy": 0.18399252295494078,
      "num_tokens": 58583438.0,
      "step": 28905
    },
    {
      "entropy": 5.996318197250366,
      "epoch": 2.526435375338635,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043605228843219494,
      "loss": 5.5991,
      "mean_token_accuracy": 0.17371416985988616,
      "num_tokens": 58594552.0,
      "step": 28910
    },
    {
      "entropy": 5.983938503265381,
      "epoch": 2.5268723236913395,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004360305265413889,
      "loss": 5.6145,
      "mean_token_accuracy": 0.17394714653491974,
      "num_tokens": 58604162.0,
      "step": 28915
    },
    {
      "entropy": 5.9691709041595455,
      "epoch": 2.5273092720440444,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004360087615619331,
      "loss": 5.5581,
      "mean_token_accuracy": 0.17459675818681716,
      "num_tokens": 58613487.0,
      "step": 28920
    },
    {
      "entropy": 5.880557250976563,
      "epoch": 2.5277462203967493,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043598699349424474,
      "loss": 5.5057,
      "mean_token_accuracy": 0.19234222322702407,
      "num_tokens": 58623252.0,
      "step": 28925
    },
    {
      "entropy": 5.958805131912231,
      "epoch": 2.528183168749454,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004359652223387415,
      "loss": 5.6957,
      "mean_token_accuracy": 0.16120317429304123,
      "num_tokens": 58633585.0,
      "step": 28930
    },
    {
      "entropy": 5.941791152954101,
      "epoch": 2.5286201171021583,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004359434480958411,
      "loss": 5.651,
      "mean_token_accuracy": 0.16809140741825104,
      "num_tokens": 58643202.0,
      "step": 28935
    },
    {
      "entropy": 5.954914188385009,
      "epoch": 2.529057065454863,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004359216707659608,
      "loss": 5.5626,
      "mean_token_accuracy": 0.1858238935470581,
      "num_tokens": 58653153.0,
      "step": 28940
    },
    {
      "entropy": 6.002790069580078,
      "epoch": 2.529494013807568,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004358998903495188,
      "loss": 5.5329,
      "mean_token_accuracy": 0.18161595314741136,
      "num_tokens": 58663835.0,
      "step": 28945
    },
    {
      "entropy": 6.016327142715454,
      "epoch": 2.5299309621602726,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004358781068469324,
      "loss": 5.6247,
      "mean_token_accuracy": 0.17680606991052628,
      "num_tokens": 58673114.0,
      "step": 28950
    },
    {
      "entropy": 5.98763484954834,
      "epoch": 2.5303679105129775,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043585632025861963,
      "loss": 5.6053,
      "mean_token_accuracy": 0.17487743198871614,
      "num_tokens": 58683121.0,
      "step": 28955
    },
    {
      "entropy": 6.03760838508606,
      "epoch": 2.530804858865682,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004358345305849984,
      "loss": 5.6504,
      "mean_token_accuracy": 0.17299544364213942,
      "num_tokens": 58693656.0,
      "step": 28960
    },
    {
      "entropy": 5.93443193435669,
      "epoch": 2.531241807218387,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043581273782648664,
      "loss": 5.5879,
      "mean_token_accuracy": 0.1799925833940506,
      "num_tokens": 58703061.0,
      "step": 28965
    },
    {
      "entropy": 5.997078704833984,
      "epoch": 2.5316787555710913,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043579094198350224,
      "loss": 5.6405,
      "mean_token_accuracy": 0.17408106476068497,
      "num_tokens": 58712463.0,
      "step": 28970
    },
    {
      "entropy": 6.012002754211426,
      "epoch": 2.5321157039237963,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043576914305646334,
      "loss": 5.5836,
      "mean_token_accuracy": 0.17737848162651063,
      "num_tokens": 58722214.0,
      "step": 28975
    },
    {
      "entropy": 5.946855163574218,
      "epoch": 2.532552652276501,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000435747341045788,
      "loss": 5.6586,
      "mean_token_accuracy": 0.16823919415473937,
      "num_tokens": 58732042.0,
      "step": 28980
    },
    {
      "entropy": 6.0291601657867435,
      "epoch": 2.5329896006292056,
      "grad_norm": 1.796875,
      "learning_rate": 0.00043572553595189444,
      "loss": 5.7303,
      "mean_token_accuracy": 0.16864857971668243,
      "num_tokens": 58743139.0,
      "step": 28985
    },
    {
      "entropy": 6.071452665328979,
      "epoch": 2.53342654898191,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043570372777520096,
      "loss": 5.6704,
      "mean_token_accuracy": 0.1679088518023491,
      "num_tokens": 58752994.0,
      "step": 28990
    },
    {
      "entropy": 5.95704140663147,
      "epoch": 2.533863497334615,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043568191651612573,
      "loss": 5.628,
      "mean_token_accuracy": 0.16869681924581528,
      "num_tokens": 58762664.0,
      "step": 28995
    },
    {
      "entropy": 5.990674114227295,
      "epoch": 2.53430044568732,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043566010217508706,
      "loss": 5.6551,
      "mean_token_accuracy": 0.16618743240833284,
      "num_tokens": 58772992.0,
      "step": 29000
    },
    {
      "entropy": 6.008784198760987,
      "epoch": 2.5347373940400244,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004356382847525035,
      "loss": 5.6405,
      "mean_token_accuracy": 0.16756619960069657,
      "num_tokens": 58783157.0,
      "step": 29005
    },
    {
      "entropy": 5.954902029037475,
      "epoch": 2.5351743423927293,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004356164642487935,
      "loss": 5.6753,
      "mean_token_accuracy": 0.173319011926651,
      "num_tokens": 58794020.0,
      "step": 29010
    },
    {
      "entropy": 5.944209146499634,
      "epoch": 2.535611290745434,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004355946406643755,
      "loss": 5.5848,
      "mean_token_accuracy": 0.18117177188396455,
      "num_tokens": 58804524.0,
      "step": 29015
    },
    {
      "entropy": 5.983925247192383,
      "epoch": 2.5360482390981387,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043557281399966814,
      "loss": 5.6295,
      "mean_token_accuracy": 0.17257027924060822,
      "num_tokens": 58814278.0,
      "step": 29020
    },
    {
      "entropy": 5.9239161014556885,
      "epoch": 2.536485187450843,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004355509842550901,
      "loss": 5.5789,
      "mean_token_accuracy": 0.17318172007799149,
      "num_tokens": 58824258.0,
      "step": 29025
    },
    {
      "entropy": 5.961574602127075,
      "epoch": 2.536922135803548,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004355291514310601,
      "loss": 5.6492,
      "mean_token_accuracy": 0.17762927263975142,
      "num_tokens": 58834316.0,
      "step": 29030
    },
    {
      "entropy": 6.010121011734009,
      "epoch": 2.537359084156253,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043550731552799676,
      "loss": 5.6647,
      "mean_token_accuracy": 0.1687854066491127,
      "num_tokens": 58844201.0,
      "step": 29035
    },
    {
      "entropy": 5.985983943939209,
      "epoch": 2.5377960325089575,
      "grad_norm": 1.125,
      "learning_rate": 0.00043548547654631917,
      "loss": 5.5811,
      "mean_token_accuracy": 0.17567051947116852,
      "num_tokens": 58853675.0,
      "step": 29040
    },
    {
      "entropy": 5.991980075836182,
      "epoch": 2.538232980861662,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004354636344864459,
      "loss": 5.6122,
      "mean_token_accuracy": 0.1724163204431534,
      "num_tokens": 58863316.0,
      "step": 29045
    },
    {
      "entropy": 5.959676504135132,
      "epoch": 2.538669929214367,
      "grad_norm": 1.375,
      "learning_rate": 0.0004354417893487961,
      "loss": 5.5716,
      "mean_token_accuracy": 0.17300185263156892,
      "num_tokens": 58873479.0,
      "step": 29050
    },
    {
      "entropy": 5.895291709899903,
      "epoch": 2.5391068775670718,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004354199411337886,
      "loss": 5.603,
      "mean_token_accuracy": 0.17313025891780853,
      "num_tokens": 58882540.0,
      "step": 29055
    },
    {
      "entropy": 5.993014812469482,
      "epoch": 2.5395438259197762,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004353980898418427,
      "loss": 5.5918,
      "mean_token_accuracy": 0.16783300787210464,
      "num_tokens": 58891565.0,
      "step": 29060
    },
    {
      "entropy": 5.9833837985992435,
      "epoch": 2.539980774272481,
      "grad_norm": 1.25,
      "learning_rate": 0.0004353762354733773,
      "loss": 5.5961,
      "mean_token_accuracy": 0.17049848139286042,
      "num_tokens": 58900689.0,
      "step": 29065
    },
    {
      "entropy": 5.927656078338623,
      "epoch": 2.5404177226251856,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043535437802881175,
      "loss": 5.5648,
      "mean_token_accuracy": 0.18412479013204575,
      "num_tokens": 58910588.0,
      "step": 29070
    },
    {
      "entropy": 5.975254964828491,
      "epoch": 2.5408546709778905,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004353325175085651,
      "loss": 5.6882,
      "mean_token_accuracy": 0.16953627318143843,
      "num_tokens": 58921189.0,
      "step": 29075
    },
    {
      "entropy": 5.96263017654419,
      "epoch": 2.541291619330595,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043531065391305685,
      "loss": 5.6042,
      "mean_token_accuracy": 0.17303879261016847,
      "num_tokens": 58930941.0,
      "step": 29080
    },
    {
      "entropy": 6.009849166870117,
      "epoch": 2.5417285676833,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004352887872427063,
      "loss": 5.7294,
      "mean_token_accuracy": 0.17154210954904556,
      "num_tokens": 58940371.0,
      "step": 29085
    },
    {
      "entropy": 6.047749423980713,
      "epoch": 2.5421655160360044,
      "grad_norm": 1.578125,
      "learning_rate": 0.00043526691749793273,
      "loss": 5.6177,
      "mean_token_accuracy": 0.16984915137290954,
      "num_tokens": 58950446.0,
      "step": 29090
    },
    {
      "entropy": 5.943446683883667,
      "epoch": 2.5426024643887093,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004352450446791557,
      "loss": 5.5889,
      "mean_token_accuracy": 0.1760086476802826,
      "num_tokens": 58960757.0,
      "step": 29095
    },
    {
      "entropy": 6.026783084869384,
      "epoch": 2.5430394127414138,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043522316878679475,
      "loss": 5.6648,
      "mean_token_accuracy": 0.1779208555817604,
      "num_tokens": 58971621.0,
      "step": 29100
    },
    {
      "entropy": 6.066559362411499,
      "epoch": 2.5434763610941187,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043520128982126944,
      "loss": 5.6437,
      "mean_token_accuracy": 0.1737103894352913,
      "num_tokens": 58982505.0,
      "step": 29105
    },
    {
      "entropy": 5.951644325256348,
      "epoch": 2.5439133094468236,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004351794077829995,
      "loss": 5.5481,
      "mean_token_accuracy": 0.18005869090557097,
      "num_tokens": 58992270.0,
      "step": 29110
    },
    {
      "entropy": 5.863481760025024,
      "epoch": 2.544350257799528,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004351575226724046,
      "loss": 5.5108,
      "mean_token_accuracy": 0.18389444798231125,
      "num_tokens": 59002510.0,
      "step": 29115
    },
    {
      "entropy": 6.003634929656982,
      "epoch": 2.544787206152233,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004351356344899045,
      "loss": 5.63,
      "mean_token_accuracy": 0.17261735200881959,
      "num_tokens": 59012942.0,
      "step": 29120
    },
    {
      "entropy": 6.0261882781982425,
      "epoch": 2.5452241545049374,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043511374323591904,
      "loss": 5.7096,
      "mean_token_accuracy": 0.17125186026096345,
      "num_tokens": 59023509.0,
      "step": 29125
    },
    {
      "entropy": 6.035990047454834,
      "epoch": 2.5456611028576424,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043509184891086795,
      "loss": 5.6021,
      "mean_token_accuracy": 0.1765693321824074,
      "num_tokens": 59033782.0,
      "step": 29130
    },
    {
      "entropy": 5.962306165695191,
      "epoch": 2.546098051210347,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043506995151517135,
      "loss": 5.6472,
      "mean_token_accuracy": 0.16781223118305205,
      "num_tokens": 59045791.0,
      "step": 29135
    },
    {
      "entropy": 5.959169244766235,
      "epoch": 2.5465349995630517,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043504805104924925,
      "loss": 5.5857,
      "mean_token_accuracy": 0.1767916202545166,
      "num_tokens": 59056578.0,
      "step": 29140
    },
    {
      "entropy": 5.997250556945801,
      "epoch": 2.546971947915756,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004350261475135217,
      "loss": 5.6492,
      "mean_token_accuracy": 0.16920868456363677,
      "num_tokens": 59066712.0,
      "step": 29145
    },
    {
      "entropy": 5.942590713500977,
      "epoch": 2.547408896268461,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004350042409084088,
      "loss": 5.5771,
      "mean_token_accuracy": 0.17894797623157502,
      "num_tokens": 59077557.0,
      "step": 29150
    },
    {
      "entropy": 5.92939977645874,
      "epoch": 2.5478458446211656,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00043498233123433064,
      "loss": 5.6115,
      "mean_token_accuracy": 0.17229997217655182,
      "num_tokens": 59086649.0,
      "step": 29155
    },
    {
      "entropy": 6.014968967437744,
      "epoch": 2.5482827929738705,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043496041849170757,
      "loss": 5.5921,
      "mean_token_accuracy": 0.17355646193027496,
      "num_tokens": 59096838.0,
      "step": 29160
    },
    {
      "entropy": 6.03086895942688,
      "epoch": 2.5487197413265754,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004349385026809599,
      "loss": 5.6321,
      "mean_token_accuracy": 0.17674482017755508,
      "num_tokens": 59107884.0,
      "step": 29165
    },
    {
      "entropy": 5.939611434936523,
      "epoch": 2.54915668967928,
      "grad_norm": 1.65625,
      "learning_rate": 0.00043491658380250784,
      "loss": 5.6647,
      "mean_token_accuracy": 0.1678069442510605,
      "num_tokens": 59118315.0,
      "step": 29170
    },
    {
      "entropy": 5.920858669281006,
      "epoch": 2.5495936380319844,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043489466185677193,
      "loss": 5.6019,
      "mean_token_accuracy": 0.17994168549776077,
      "num_tokens": 59128331.0,
      "step": 29175
    },
    {
      "entropy": 5.994144248962402,
      "epoch": 2.5500305863846893,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004348727368441727,
      "loss": 5.6237,
      "mean_token_accuracy": 0.1779268503189087,
      "num_tokens": 59138643.0,
      "step": 29180
    },
    {
      "entropy": 5.990560007095337,
      "epoch": 2.550467534737394,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004348508087651306,
      "loss": 5.5641,
      "mean_token_accuracy": 0.1786769822239876,
      "num_tokens": 59148364.0,
      "step": 29185
    },
    {
      "entropy": 5.947463035583496,
      "epoch": 2.5509044830900987,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043482887762006614,
      "loss": 5.6234,
      "mean_token_accuracy": 0.1770188421010971,
      "num_tokens": 59158317.0,
      "step": 29190
    },
    {
      "entropy": 6.030345630645752,
      "epoch": 2.5513414314428036,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043480694340940017,
      "loss": 5.6897,
      "mean_token_accuracy": 0.17361495941877364,
      "num_tokens": 59168879.0,
      "step": 29195
    },
    {
      "entropy": 6.009251117706299,
      "epoch": 2.551778379795508,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004347850061335532,
      "loss": 5.6094,
      "mean_token_accuracy": 0.17313436716794967,
      "num_tokens": 59179248.0,
      "step": 29200
    },
    {
      "entropy": 5.882672929763794,
      "epoch": 2.552215328148213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004347630657929461,
      "loss": 5.5441,
      "mean_token_accuracy": 0.18215697407722473,
      "num_tokens": 59189217.0,
      "step": 29205
    },
    {
      "entropy": 6.008963918685913,
      "epoch": 2.5526522765009174,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043474112238799974,
      "loss": 5.6271,
      "mean_token_accuracy": 0.1767999142408371,
      "num_tokens": 59199857.0,
      "step": 29210
    },
    {
      "entropy": 6.035866117477417,
      "epoch": 2.5530892248536223,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043471917591913486,
      "loss": 5.6335,
      "mean_token_accuracy": 0.1753063127398491,
      "num_tokens": 59210119.0,
      "step": 29215
    },
    {
      "entropy": 5.931250762939453,
      "epoch": 2.5535261732063272,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043469722638677256,
      "loss": 5.5585,
      "mean_token_accuracy": 0.18011242747306824,
      "num_tokens": 59219258.0,
      "step": 29220
    },
    {
      "entropy": 5.937494945526123,
      "epoch": 2.5539631215590317,
      "grad_norm": 1.9375,
      "learning_rate": 0.00043467527379133377,
      "loss": 5.6521,
      "mean_token_accuracy": 0.17350521832704544,
      "num_tokens": 59229664.0,
      "step": 29225
    },
    {
      "entropy": 6.058416557312012,
      "epoch": 2.554400069911736,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004346533181332395,
      "loss": 5.6582,
      "mean_token_accuracy": 0.1698213368654251,
      "num_tokens": 59239699.0,
      "step": 29230
    },
    {
      "entropy": 5.971835899353027,
      "epoch": 2.554837018264441,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043463135941291105,
      "loss": 5.5972,
      "mean_token_accuracy": 0.1750577434897423,
      "num_tokens": 59249891.0,
      "step": 29235
    },
    {
      "entropy": 5.903539514541626,
      "epoch": 2.555273966617146,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004346093976307694,
      "loss": 5.6423,
      "mean_token_accuracy": 0.17244429588317872,
      "num_tokens": 59260092.0,
      "step": 29240
    },
    {
      "entropy": 5.917753267288208,
      "epoch": 2.5557109149698505,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004345874327872359,
      "loss": 5.6528,
      "mean_token_accuracy": 0.1717403382062912,
      "num_tokens": 59268844.0,
      "step": 29245
    },
    {
      "entropy": 6.043127727508545,
      "epoch": 2.5561478633225554,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004345654648827318,
      "loss": 5.7112,
      "mean_token_accuracy": 0.16484513133764267,
      "num_tokens": 59279275.0,
      "step": 29250
    },
    {
      "entropy": 6.003256464004517,
      "epoch": 2.55658481167526,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004345434939176784,
      "loss": 5.6511,
      "mean_token_accuracy": 0.17599427551031113,
      "num_tokens": 59290596.0,
      "step": 29255
    },
    {
      "entropy": 5.959814262390137,
      "epoch": 2.5570217600279648,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043452151989249727,
      "loss": 5.5916,
      "mean_token_accuracy": 0.17266078144311905,
      "num_tokens": 59300642.0,
      "step": 29260
    },
    {
      "entropy": 5.935699081420898,
      "epoch": 2.5574587083806692,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004344995428076098,
      "loss": 5.6312,
      "mean_token_accuracy": 0.17535073906183243,
      "num_tokens": 59310064.0,
      "step": 29265
    },
    {
      "entropy": 6.057871770858765,
      "epoch": 2.557895656733374,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043447756266343755,
      "loss": 5.622,
      "mean_token_accuracy": 0.17525675445795058,
      "num_tokens": 59319114.0,
      "step": 29270
    },
    {
      "entropy": 5.992635822296142,
      "epoch": 2.558332605086079,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00043445557946040207,
      "loss": 5.6976,
      "mean_token_accuracy": 0.1697056546807289,
      "num_tokens": 59329414.0,
      "step": 29275
    },
    {
      "entropy": 6.02472677230835,
      "epoch": 2.5587695534387835,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043443359319892495,
      "loss": 5.7068,
      "mean_token_accuracy": 0.16786182224750518,
      "num_tokens": 59339980.0,
      "step": 29280
    },
    {
      "entropy": 6.065290117263794,
      "epoch": 2.559206501791488,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000434411603879428,
      "loss": 5.6822,
      "mean_token_accuracy": 0.16451524794101716,
      "num_tokens": 59349754.0,
      "step": 29285
    },
    {
      "entropy": 5.909274578094482,
      "epoch": 2.559643450144193,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000434389611502333,
      "loss": 5.4799,
      "mean_token_accuracy": 0.1913463890552521,
      "num_tokens": 59358542.0,
      "step": 29290
    },
    {
      "entropy": 5.958402395248413,
      "epoch": 2.560080398496898,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004343676160680617,
      "loss": 5.6137,
      "mean_token_accuracy": 0.17632492482662201,
      "num_tokens": 59367743.0,
      "step": 29295
    },
    {
      "entropy": 6.016168069839478,
      "epoch": 2.5605173468496023,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000434345617577036,
      "loss": 5.6254,
      "mean_token_accuracy": 0.17186663001775743,
      "num_tokens": 59377622.0,
      "step": 29300
    },
    {
      "entropy": 5.909173536300659,
      "epoch": 2.560954295202307,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043432361602967787,
      "loss": 5.5816,
      "mean_token_accuracy": 0.17598945796489715,
      "num_tokens": 59388661.0,
      "step": 29305
    },
    {
      "entropy": 5.996257162094116,
      "epoch": 2.5613912435550117,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043430161142640934,
      "loss": 5.6629,
      "mean_token_accuracy": 0.17084756940603257,
      "num_tokens": 59399935.0,
      "step": 29310
    },
    {
      "entropy": 6.013975000381469,
      "epoch": 2.5618281919077166,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004342796037676524,
      "loss": 5.598,
      "mean_token_accuracy": 0.1791263073682785,
      "num_tokens": 59409513.0,
      "step": 29315
    },
    {
      "entropy": 5.9609959602355955,
      "epoch": 2.562265140260421,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004342575930538292,
      "loss": 5.6037,
      "mean_token_accuracy": 0.17323299497365952,
      "num_tokens": 59420158.0,
      "step": 29320
    },
    {
      "entropy": 5.9467151165008545,
      "epoch": 2.562702088613126,
      "grad_norm": 1.609375,
      "learning_rate": 0.00043423557928536197,
      "loss": 5.656,
      "mean_token_accuracy": 0.1731094390153885,
      "num_tokens": 59429440.0,
      "step": 29325
    },
    {
      "entropy": 5.898823261260986,
      "epoch": 2.563139036965831,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004342135624626728,
      "loss": 5.5363,
      "mean_token_accuracy": 0.18207371085882187,
      "num_tokens": 59439014.0,
      "step": 29330
    },
    {
      "entropy": 5.926129341125488,
      "epoch": 2.5635759853185354,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004341915425861841,
      "loss": 5.5765,
      "mean_token_accuracy": 0.17837532609701157,
      "num_tokens": 59448714.0,
      "step": 29335
    },
    {
      "entropy": 5.935424089431763,
      "epoch": 2.56401293367124,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004341695196563183,
      "loss": 5.6326,
      "mean_token_accuracy": 0.17459529191255568,
      "num_tokens": 59459577.0,
      "step": 29340
    },
    {
      "entropy": 6.051880836486816,
      "epoch": 2.5644498820239447,
      "grad_norm": 2.0,
      "learning_rate": 0.00043414749367349755,
      "loss": 5.6156,
      "mean_token_accuracy": 0.17785526812076569,
      "num_tokens": 59470313.0,
      "step": 29345
    },
    {
      "entropy": 5.9955606937408445,
      "epoch": 2.5648868303766497,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004341254646381446,
      "loss": 5.6389,
      "mean_token_accuracy": 0.17633092999458314,
      "num_tokens": 59479484.0,
      "step": 29350
    },
    {
      "entropy": 5.962207698822022,
      "epoch": 2.565323778729354,
      "grad_norm": 1.25,
      "learning_rate": 0.00043410343255068177,
      "loss": 5.5973,
      "mean_token_accuracy": 0.17009349316358566,
      "num_tokens": 59490290.0,
      "step": 29355
    },
    {
      "entropy": 5.968140983581543,
      "epoch": 2.565760727082059,
      "grad_norm": 1.5625,
      "learning_rate": 0.00043408139741153175,
      "loss": 5.6482,
      "mean_token_accuracy": 0.17647232860326767,
      "num_tokens": 59501080.0,
      "step": 29360
    },
    {
      "entropy": 5.966657400131226,
      "epoch": 2.5661976754347635,
      "grad_norm": 1.375,
      "learning_rate": 0.00043405935922111725,
      "loss": 5.5241,
      "mean_token_accuracy": 0.18910010755062104,
      "num_tokens": 59511453.0,
      "step": 29365
    },
    {
      "entropy": 5.93217453956604,
      "epoch": 2.5666346237874684,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004340373179798608,
      "loss": 5.6482,
      "mean_token_accuracy": 0.17876284420490265,
      "num_tokens": 59521438.0,
      "step": 29370
    },
    {
      "entropy": 5.985105228424072,
      "epoch": 2.567071572140173,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004340152736881853,
      "loss": 5.6889,
      "mean_token_accuracy": 0.17289944142103195,
      "num_tokens": 59531763.0,
      "step": 29375
    },
    {
      "entropy": 6.105340385437012,
      "epoch": 2.567508520492878,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043399322634651345,
      "loss": 5.7579,
      "mean_token_accuracy": 0.15994392037391664,
      "num_tokens": 59542857.0,
      "step": 29380
    },
    {
      "entropy": 6.017859268188476,
      "epoch": 2.5679454688455827,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004339711759552684,
      "loss": 5.5885,
      "mean_token_accuracy": 0.1801576569676399,
      "num_tokens": 59552353.0,
      "step": 29385
    },
    {
      "entropy": 6.013512420654297,
      "epoch": 2.568382417198287,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004339491225148727,
      "loss": 5.657,
      "mean_token_accuracy": 0.17234538346529008,
      "num_tokens": 59562377.0,
      "step": 29390
    },
    {
      "entropy": 6.004199695587158,
      "epoch": 2.5688193655509917,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043392706602574967,
      "loss": 5.6273,
      "mean_token_accuracy": 0.17525392323732375,
      "num_tokens": 59572141.0,
      "step": 29395
    },
    {
      "entropy": 5.92354941368103,
      "epoch": 2.5692563139036966,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043390500648832214,
      "loss": 5.5734,
      "mean_token_accuracy": 0.17035203129053117,
      "num_tokens": 59581883.0,
      "step": 29400
    },
    {
      "entropy": 5.972730445861816,
      "epoch": 2.5696932622564015,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004338829439030134,
      "loss": 5.6514,
      "mean_token_accuracy": 0.17054153233766556,
      "num_tokens": 59591798.0,
      "step": 29405
    },
    {
      "entropy": 6.0133678913116455,
      "epoch": 2.570130210609106,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004338608782702465,
      "loss": 5.5577,
      "mean_token_accuracy": 0.17745942175388335,
      "num_tokens": 59601525.0,
      "step": 29410
    },
    {
      "entropy": 5.942092418670654,
      "epoch": 2.570567158961811,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004338388095904447,
      "loss": 5.6878,
      "mean_token_accuracy": 0.1720588654279709,
      "num_tokens": 59611335.0,
      "step": 29415
    },
    {
      "entropy": 5.95778112411499,
      "epoch": 2.5710041073145153,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004338167378640314,
      "loss": 5.6551,
      "mean_token_accuracy": 0.17268840670585633,
      "num_tokens": 59620962.0,
      "step": 29420
    },
    {
      "entropy": 6.010981178283691,
      "epoch": 2.5714410556672203,
      "grad_norm": 1.25,
      "learning_rate": 0.0004337946630914297,
      "loss": 5.6022,
      "mean_token_accuracy": 0.17115475237369537,
      "num_tokens": 59631568.0,
      "step": 29425
    },
    {
      "entropy": 5.9606321334838865,
      "epoch": 2.5718780040199247,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043377258527306316,
      "loss": 5.6657,
      "mean_token_accuracy": 0.17167529165744783,
      "num_tokens": 59642249.0,
      "step": 29430
    },
    {
      "entropy": 6.0266594886779785,
      "epoch": 2.5723149523726296,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004337505044093554,
      "loss": 5.63,
      "mean_token_accuracy": 0.16800073832273482,
      "num_tokens": 59652516.0,
      "step": 29435
    },
    {
      "entropy": 5.881336784362793,
      "epoch": 2.572751900725334,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043372842050072965,
      "loss": 5.4968,
      "mean_token_accuracy": 0.18849113434553147,
      "num_tokens": 59662487.0,
      "step": 29440
    },
    {
      "entropy": 5.911009311676025,
      "epoch": 2.573188849078039,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004337063335476097,
      "loss": 5.6528,
      "mean_token_accuracy": 0.1694091811776161,
      "num_tokens": 59674061.0,
      "step": 29445
    },
    {
      "entropy": 5.949256563186646,
      "epoch": 2.5736257974307435,
      "grad_norm": 1.125,
      "learning_rate": 0.000433684243550419,
      "loss": 5.655,
      "mean_token_accuracy": 0.17843039482831954,
      "num_tokens": 59684159.0,
      "step": 29450
    },
    {
      "entropy": 6.002117252349853,
      "epoch": 2.5740627457834484,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004336621505095815,
      "loss": 5.6523,
      "mean_token_accuracy": 0.1705852121114731,
      "num_tokens": 59693742.0,
      "step": 29455
    },
    {
      "entropy": 6.033664417266846,
      "epoch": 2.5744996941361533,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043364005442552073,
      "loss": 5.7043,
      "mean_token_accuracy": 0.16491157412528992,
      "num_tokens": 59703613.0,
      "step": 29460
    },
    {
      "entropy": 5.986129283905029,
      "epoch": 2.574936642488858,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004336179552986606,
      "loss": 5.568,
      "mean_token_accuracy": 0.1793495461344719,
      "num_tokens": 59713409.0,
      "step": 29465
    },
    {
      "entropy": 5.9399878025054935,
      "epoch": 2.5753735908415623,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043359585312942496,
      "loss": 5.5533,
      "mean_token_accuracy": 0.18508943021297455,
      "num_tokens": 59723502.0,
      "step": 29470
    },
    {
      "entropy": 5.966472482681274,
      "epoch": 2.575810539194267,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043357374791823784,
      "loss": 5.6635,
      "mean_token_accuracy": 0.17423918694257737,
      "num_tokens": 59734069.0,
      "step": 29475
    },
    {
      "entropy": 5.9849381923675535,
      "epoch": 2.576247487546972,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043355163966552303,
      "loss": 5.6202,
      "mean_token_accuracy": 0.1690603017807007,
      "num_tokens": 59743793.0,
      "step": 29480
    },
    {
      "entropy": 5.932826709747315,
      "epoch": 2.5766844358996766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043352952837170467,
      "loss": 5.6677,
      "mean_token_accuracy": 0.16972599625587464,
      "num_tokens": 59754524.0,
      "step": 29485
    },
    {
      "entropy": 6.071797323226929,
      "epoch": 2.5771213842523815,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000433507414037207,
      "loss": 5.6192,
      "mean_token_accuracy": 0.17663384228944778,
      "num_tokens": 59765437.0,
      "step": 29490
    },
    {
      "entropy": 5.992762088775635,
      "epoch": 2.577558332605086,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000433485296662454,
      "loss": 5.6225,
      "mean_token_accuracy": 0.17389067709445954,
      "num_tokens": 59776136.0,
      "step": 29495
    },
    {
      "entropy": 5.9018027782440186,
      "epoch": 2.577995280957791,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043346317624787007,
      "loss": 5.5832,
      "mean_token_accuracy": 0.17390791475772857,
      "num_tokens": 59786375.0,
      "step": 29500
    },
    {
      "entropy": 5.904481840133667,
      "epoch": 2.5784322293104953,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004334410527938793,
      "loss": 5.5294,
      "mean_token_accuracy": 0.17999818325042724,
      "num_tokens": 59796937.0,
      "step": 29505
    },
    {
      "entropy": 5.973866605758667,
      "epoch": 2.5788691776632002,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043341892630090607,
      "loss": 5.6176,
      "mean_token_accuracy": 0.1739523485302925,
      "num_tokens": 59806625.0,
      "step": 29510
    },
    {
      "entropy": 6.005253505706787,
      "epoch": 2.579306126015905,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004333967967693749,
      "loss": 5.6626,
      "mean_token_accuracy": 0.1706141635775566,
      "num_tokens": 59816618.0,
      "step": 29515
    },
    {
      "entropy": 6.0140830993652346,
      "epoch": 2.5797430743686096,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004333746641997102,
      "loss": 5.633,
      "mean_token_accuracy": 0.17376273423433303,
      "num_tokens": 59827666.0,
      "step": 29520
    },
    {
      "entropy": 5.992721891403198,
      "epoch": 2.580180022721314,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004333525285923363,
      "loss": 5.6189,
      "mean_token_accuracy": 0.17484813034534455,
      "num_tokens": 59837704.0,
      "step": 29525
    },
    {
      "entropy": 5.9370622634887695,
      "epoch": 2.580616971074019,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004333303899476781,
      "loss": 5.5594,
      "mean_token_accuracy": 0.17588390856981279,
      "num_tokens": 59848186.0,
      "step": 29530
    },
    {
      "entropy": 5.969654941558838,
      "epoch": 2.581053919426724,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043330824826616,
      "loss": 5.5928,
      "mean_token_accuracy": 0.1726814717054367,
      "num_tokens": 59858453.0,
      "step": 29535
    },
    {
      "entropy": 5.911748838424683,
      "epoch": 2.5814908677794284,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004332861035482067,
      "loss": 5.6083,
      "mean_token_accuracy": 0.17169446051120757,
      "num_tokens": 59867863.0,
      "step": 29540
    },
    {
      "entropy": 5.945579671859742,
      "epoch": 2.5819278161321333,
      "grad_norm": 1.390625,
      "learning_rate": 0.000433263955794243,
      "loss": 5.6103,
      "mean_token_accuracy": 0.1772844448685646,
      "num_tokens": 59879932.0,
      "step": 29545
    },
    {
      "entropy": 5.996226024627686,
      "epoch": 2.5823647644848378,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004332418050046937,
      "loss": 5.6066,
      "mean_token_accuracy": 0.17027337700128556,
      "num_tokens": 59889586.0,
      "step": 29550
    },
    {
      "entropy": 5.921569299697876,
      "epoch": 2.5828017128375427,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004332196511799837,
      "loss": 5.5481,
      "mean_token_accuracy": 0.17535588443279265,
      "num_tokens": 59898789.0,
      "step": 29555
    },
    {
      "entropy": 5.87989296913147,
      "epoch": 2.583238661190247,
      "grad_norm": 2.625,
      "learning_rate": 0.0004331974943205379,
      "loss": 5.6335,
      "mean_token_accuracy": 0.17424348443746568,
      "num_tokens": 59909346.0,
      "step": 29560
    },
    {
      "entropy": 5.97916808128357,
      "epoch": 2.583675609542952,
      "grad_norm": 1.71875,
      "learning_rate": 0.00043317533442678124,
      "loss": 5.5479,
      "mean_token_accuracy": 0.1852991372346878,
      "num_tokens": 59919979.0,
      "step": 29565
    },
    {
      "entropy": 5.9714008331298825,
      "epoch": 2.584112557895657,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00043315317149913876,
      "loss": 5.6059,
      "mean_token_accuracy": 0.17655448168516158,
      "num_tokens": 59929855.0,
      "step": 29570
    },
    {
      "entropy": 5.937791204452514,
      "epoch": 2.5845495062483614,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043313100553803564,
      "loss": 5.591,
      "mean_token_accuracy": 0.1741088718175888,
      "num_tokens": 59939148.0,
      "step": 29575
    },
    {
      "entropy": 5.876013660430909,
      "epoch": 2.584986454601066,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004331088365438969,
      "loss": 5.5266,
      "mean_token_accuracy": 0.17343011498451233,
      "num_tokens": 59948645.0,
      "step": 29580
    },
    {
      "entropy": 5.971442031860351,
      "epoch": 2.585423402953771,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004330866645171479,
      "loss": 5.5873,
      "mean_token_accuracy": 0.17552874088287354,
      "num_tokens": 59958777.0,
      "step": 29585
    },
    {
      "entropy": 5.935768079757691,
      "epoch": 2.5858603513064757,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004330644894582138,
      "loss": 5.5473,
      "mean_token_accuracy": 0.18644680976867675,
      "num_tokens": 59969176.0,
      "step": 29590
    },
    {
      "entropy": 5.881566190719605,
      "epoch": 2.58629729965918,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00043304231136751995,
      "loss": 5.637,
      "mean_token_accuracy": 0.172166208922863,
      "num_tokens": 59979263.0,
      "step": 29595
    },
    {
      "entropy": 5.966000699996949,
      "epoch": 2.586734248011885,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043302013024549174,
      "loss": 5.5661,
      "mean_token_accuracy": 0.18013789206743241,
      "num_tokens": 59989313.0,
      "step": 29600
    },
    {
      "entropy": 5.993509721755982,
      "epoch": 2.5871711963645896,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043299794609255474,
      "loss": 5.6194,
      "mean_token_accuracy": 0.17288853973150253,
      "num_tokens": 59999410.0,
      "step": 29605
    },
    {
      "entropy": 5.897624158859253,
      "epoch": 2.5876081447172945,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004329757589091342,
      "loss": 5.5308,
      "mean_token_accuracy": 0.18117574602365494,
      "num_tokens": 60010206.0,
      "step": 29610
    },
    {
      "entropy": 5.8199857711792,
      "epoch": 2.588045093069999,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043295356869565586,
      "loss": 5.4486,
      "mean_token_accuracy": 0.19439005702733994,
      "num_tokens": 60018692.0,
      "step": 29615
    },
    {
      "entropy": 5.958795738220215,
      "epoch": 2.588482041422704,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043293137545254536,
      "loss": 5.6705,
      "mean_token_accuracy": 0.16851996779441833,
      "num_tokens": 60028486.0,
      "step": 29620
    },
    {
      "entropy": 5.996899509429932,
      "epoch": 2.588918989775409,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004329091791802283,
      "loss": 5.6396,
      "mean_token_accuracy": 0.1714769959449768,
      "num_tokens": 60039347.0,
      "step": 29625
    },
    {
      "entropy": 5.917147064208985,
      "epoch": 2.5893559381281133,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004328869798791305,
      "loss": 5.5311,
      "mean_token_accuracy": 0.18793608993291855,
      "num_tokens": 60049891.0,
      "step": 29630
    },
    {
      "entropy": 5.997017908096313,
      "epoch": 2.5897928864808177,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043286477754967757,
      "loss": 5.6334,
      "mean_token_accuracy": 0.17115118205547333,
      "num_tokens": 60061264.0,
      "step": 29635
    },
    {
      "entropy": 5.990096235275269,
      "epoch": 2.5902298348335226,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043284257219229557,
      "loss": 5.6435,
      "mean_token_accuracy": 0.17224315404891968,
      "num_tokens": 60072453.0,
      "step": 29640
    },
    {
      "entropy": 5.904420614242554,
      "epoch": 2.5906667831862276,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043282036380741035,
      "loss": 5.6095,
      "mean_token_accuracy": 0.1747308850288391,
      "num_tokens": 60083619.0,
      "step": 29645
    },
    {
      "entropy": 5.999918222427368,
      "epoch": 2.591103731538932,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004327981523954478,
      "loss": 5.5357,
      "mean_token_accuracy": 0.17849100679159163,
      "num_tokens": 60093231.0,
      "step": 29650
    },
    {
      "entropy": 5.897391033172608,
      "epoch": 2.591540679891637,
      "grad_norm": 1.296875,
      "learning_rate": 0.000432775937956834,
      "loss": 5.5526,
      "mean_token_accuracy": 0.1783018544316292,
      "num_tokens": 60103017.0,
      "step": 29655
    },
    {
      "entropy": 5.973471403121948,
      "epoch": 2.5919776282443414,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004327537204919951,
      "loss": 5.598,
      "mean_token_accuracy": 0.16742797195911407,
      "num_tokens": 60113095.0,
      "step": 29660
    },
    {
      "entropy": 6.019942378997802,
      "epoch": 2.5924145765970463,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004327315000013571,
      "loss": 5.6532,
      "mean_token_accuracy": 0.1745111048221588,
      "num_tokens": 60123506.0,
      "step": 29665
    },
    {
      "entropy": 5.96159520149231,
      "epoch": 2.592851524949751,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004327092764853463,
      "loss": 5.7011,
      "mean_token_accuracy": 0.1678940862417221,
      "num_tokens": 60134349.0,
      "step": 29670
    },
    {
      "entropy": 5.950967454910279,
      "epoch": 2.5932884733024557,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043268704994438897,
      "loss": 5.5622,
      "mean_token_accuracy": 0.174437814950943,
      "num_tokens": 60143970.0,
      "step": 29675
    },
    {
      "entropy": 5.92673077583313,
      "epoch": 2.5937254216551606,
      "grad_norm": 1.375,
      "learning_rate": 0.0004326648203789113,
      "loss": 5.5855,
      "mean_token_accuracy": 0.17579474151134492,
      "num_tokens": 60154904.0,
      "step": 29680
    },
    {
      "entropy": 5.919591999053955,
      "epoch": 2.594162370007865,
      "grad_norm": 1.25,
      "learning_rate": 0.0004326425877893399,
      "loss": 5.5839,
      "mean_token_accuracy": 0.17657359391450883,
      "num_tokens": 60164747.0,
      "step": 29685
    },
    {
      "entropy": 6.030140829086304,
      "epoch": 2.5945993183605696,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043262035217610104,
      "loss": 5.6186,
      "mean_token_accuracy": 0.17685027718544005,
      "num_tokens": 60175487.0,
      "step": 29690
    },
    {
      "entropy": 5.938058471679687,
      "epoch": 2.5950362667132745,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043259811353962114,
      "loss": 5.593,
      "mean_token_accuracy": 0.1757149264216423,
      "num_tokens": 60185216.0,
      "step": 29695
    },
    {
      "entropy": 5.97583703994751,
      "epoch": 2.5954732150659794,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004325758718803269,
      "loss": 5.6084,
      "mean_token_accuracy": 0.1690000221133232,
      "num_tokens": 60195495.0,
      "step": 29700
    },
    {
      "entropy": 6.0079092502594,
      "epoch": 2.595910163418684,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043255362719864487,
      "loss": 5.6651,
      "mean_token_accuracy": 0.1664616197347641,
      "num_tokens": 60205556.0,
      "step": 29705
    },
    {
      "entropy": 5.958883762359619,
      "epoch": 2.5963471117713888,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00043253137949500166,
      "loss": 5.6037,
      "mean_token_accuracy": 0.1763804465532303,
      "num_tokens": 60215904.0,
      "step": 29710
    },
    {
      "entropy": 5.895038080215454,
      "epoch": 2.5967840601240932,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043250912876982415,
      "loss": 5.5306,
      "mean_token_accuracy": 0.18261654078960418,
      "num_tokens": 60224762.0,
      "step": 29715
    },
    {
      "entropy": 5.920426988601685,
      "epoch": 2.597221008476798,
      "grad_norm": 1.09375,
      "learning_rate": 0.000432486875023539,
      "loss": 5.5698,
      "mean_token_accuracy": 0.1776355117559433,
      "num_tokens": 60235284.0,
      "step": 29720
    },
    {
      "entropy": 5.985507869720459,
      "epoch": 2.5976579568295026,
      "grad_norm": 2.953125,
      "learning_rate": 0.000432464618256573,
      "loss": 5.6305,
      "mean_token_accuracy": 0.17464576065540313,
      "num_tokens": 60245939.0,
      "step": 29725
    },
    {
      "entropy": 6.040480422973633,
      "epoch": 2.5980949051822075,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004324423584693531,
      "loss": 5.6258,
      "mean_token_accuracy": 0.17130244970321656,
      "num_tokens": 60256903.0,
      "step": 29730
    },
    {
      "entropy": 5.9878827095031735,
      "epoch": 2.5985318535349125,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004324200956623063,
      "loss": 5.5699,
      "mean_token_accuracy": 0.17011653482913972,
      "num_tokens": 60267046.0,
      "step": 29735
    },
    {
      "entropy": 5.976599550247192,
      "epoch": 2.598968801887617,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043239782983585964,
      "loss": 5.6221,
      "mean_token_accuracy": 0.17416119575500488,
      "num_tokens": 60276858.0,
      "step": 29740
    },
    {
      "entropy": 5.931429100036621,
      "epoch": 2.5994057502403214,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043237556099044,
      "loss": 5.5899,
      "mean_token_accuracy": 0.17704679816961288,
      "num_tokens": 60285874.0,
      "step": 29745
    },
    {
      "entropy": 5.955851364135742,
      "epoch": 2.5998426985930263,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004323532891264748,
      "loss": 5.655,
      "mean_token_accuracy": 0.17108396142721177,
      "num_tokens": 60295580.0,
      "step": 29750
    },
    {
      "entropy": 6.025412273406983,
      "epoch": 2.600279646945731,
      "grad_norm": 1.25,
      "learning_rate": 0.0004323310142443909,
      "loss": 5.6259,
      "mean_token_accuracy": 0.17492730021476746,
      "num_tokens": 60305425.0,
      "step": 29755
    },
    {
      "entropy": 5.9703082084655765,
      "epoch": 2.6007165952984357,
      "grad_norm": 1.25,
      "learning_rate": 0.00043230873634461577,
      "loss": 5.5918,
      "mean_token_accuracy": 0.17354216277599335,
      "num_tokens": 60315578.0,
      "step": 29760
    },
    {
      "entropy": 5.976793050765991,
      "epoch": 2.6011535436511406,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004322864554275767,
      "loss": 5.5861,
      "mean_token_accuracy": 0.17891765981912613,
      "num_tokens": 60325345.0,
      "step": 29765
    },
    {
      "entropy": 5.977735280990601,
      "epoch": 2.601590492003845,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043226417149370093,
      "loss": 5.553,
      "mean_token_accuracy": 0.17383327335119247,
      "num_tokens": 60334608.0,
      "step": 29770
    },
    {
      "entropy": 5.941929578781128,
      "epoch": 2.60202744035655,
      "grad_norm": 2.109375,
      "learning_rate": 0.00043224188454341593,
      "loss": 5.5862,
      "mean_token_accuracy": 0.17646297067403793,
      "num_tokens": 60346909.0,
      "step": 29775
    },
    {
      "entropy": 5.86066427230835,
      "epoch": 2.6024643887092545,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004322195945771493,
      "loss": 5.5528,
      "mean_token_accuracy": 0.18539124280214309,
      "num_tokens": 60355829.0,
      "step": 29780
    },
    {
      "entropy": 5.866851806640625,
      "epoch": 2.6029013370619594,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004321973015953284,
      "loss": 5.5556,
      "mean_token_accuracy": 0.17392218559980394,
      "num_tokens": 60364954.0,
      "step": 29785
    },
    {
      "entropy": 6.032665586471557,
      "epoch": 2.603338285414664,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043217500559838084,
      "loss": 5.7184,
      "mean_token_accuracy": 0.16491560339927674,
      "num_tokens": 60375575.0,
      "step": 29790
    },
    {
      "entropy": 6.020511722564697,
      "epoch": 2.6037752337673687,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043215270658673437,
      "loss": 5.632,
      "mean_token_accuracy": 0.17421837002038956,
      "num_tokens": 60385534.0,
      "step": 29795
    },
    {
      "entropy": 5.936035633087158,
      "epoch": 2.604212182120073,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004321304045608166,
      "loss": 5.6056,
      "mean_token_accuracy": 0.17142608314752578,
      "num_tokens": 60395517.0,
      "step": 29800
    },
    {
      "entropy": 6.031725120544434,
      "epoch": 2.604649130472778,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004321080995210554,
      "loss": 5.6742,
      "mean_token_accuracy": 0.17873824536800384,
      "num_tokens": 60404781.0,
      "step": 29805
    },
    {
      "entropy": 5.953365135192871,
      "epoch": 2.605086078825483,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004320857914678785,
      "loss": 5.6386,
      "mean_token_accuracy": 0.17621469497680664,
      "num_tokens": 60415323.0,
      "step": 29810
    },
    {
      "entropy": 5.998104238510132,
      "epoch": 2.6055230271781875,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004320634804017138,
      "loss": 5.7301,
      "mean_token_accuracy": 0.16171030700206757,
      "num_tokens": 60425328.0,
      "step": 29815
    },
    {
      "entropy": 5.940479803085327,
      "epoch": 2.605959975530892,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004320411663229892,
      "loss": 5.6025,
      "mean_token_accuracy": 0.17368533611297607,
      "num_tokens": 60435391.0,
      "step": 29820
    },
    {
      "entropy": 5.998672103881836,
      "epoch": 2.606396923883597,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004320188492321328,
      "loss": 5.6354,
      "mean_token_accuracy": 0.16907297670841218,
      "num_tokens": 60445155.0,
      "step": 29825
    },
    {
      "entropy": 6.011406803131104,
      "epoch": 2.606833872236302,
      "grad_norm": 2.125,
      "learning_rate": 0.0004319965291295726,
      "loss": 5.5634,
      "mean_token_accuracy": 0.17386059910058976,
      "num_tokens": 60455655.0,
      "step": 29830
    },
    {
      "entropy": 5.909736490249633,
      "epoch": 2.6072708205890063,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043197420601573665,
      "loss": 5.4738,
      "mean_token_accuracy": 0.18409329950809478,
      "num_tokens": 60465080.0,
      "step": 29835
    },
    {
      "entropy": 5.9246070861816404,
      "epoch": 2.607707768941711,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043195187989105327,
      "loss": 5.614,
      "mean_token_accuracy": 0.17697376161813735,
      "num_tokens": 60474970.0,
      "step": 29840
    },
    {
      "entropy": 6.001937770843506,
      "epoch": 2.6081447172944157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043192955075595045,
      "loss": 5.5695,
      "mean_token_accuracy": 0.178043369948864,
      "num_tokens": 60486599.0,
      "step": 29845
    },
    {
      "entropy": 5.9902966022491455,
      "epoch": 2.6085816656471206,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004319072186108567,
      "loss": 5.658,
      "mean_token_accuracy": 0.17048973590135574,
      "num_tokens": 60496849.0,
      "step": 29850
    },
    {
      "entropy": 5.980178785324097,
      "epoch": 2.609018613999825,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004318848834562003,
      "loss": 5.64,
      "mean_token_accuracy": 0.17113014161586762,
      "num_tokens": 60506708.0,
      "step": 29855
    },
    {
      "entropy": 5.845941162109375,
      "epoch": 2.60945556235253,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043186254529240957,
      "loss": 5.4964,
      "mean_token_accuracy": 0.19307905733585357,
      "num_tokens": 60516538.0,
      "step": 29860
    },
    {
      "entropy": 5.935975408554077,
      "epoch": 2.609892510705235,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043184020411991315,
      "loss": 5.5281,
      "mean_token_accuracy": 0.1846722424030304,
      "num_tokens": 60526206.0,
      "step": 29865
    },
    {
      "entropy": 5.9683661460876465,
      "epoch": 2.6103294590579393,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004318178599391393,
      "loss": 5.7057,
      "mean_token_accuracy": 0.16501474380493164,
      "num_tokens": 60536878.0,
      "step": 29870
    },
    {
      "entropy": 6.043201923370361,
      "epoch": 2.610766407410644,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043179551275051675,
      "loss": 5.7634,
      "mean_token_accuracy": 0.16625409424304963,
      "num_tokens": 60547709.0,
      "step": 29875
    },
    {
      "entropy": 6.06393871307373,
      "epoch": 2.6112033557633487,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004317731625544742,
      "loss": 5.5348,
      "mean_token_accuracy": 0.17900736927986144,
      "num_tokens": 60557257.0,
      "step": 29880
    },
    {
      "entropy": 5.965143489837646,
      "epoch": 2.6116403041160536,
      "grad_norm": 1.71875,
      "learning_rate": 0.00043175080935144015,
      "loss": 5.6696,
      "mean_token_accuracy": 0.17515866458415985,
      "num_tokens": 60568663.0,
      "step": 29885
    },
    {
      "entropy": 5.912416553497314,
      "epoch": 2.612077252468758,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004317284531418434,
      "loss": 5.5852,
      "mean_token_accuracy": 0.18065462857484818,
      "num_tokens": 60579168.0,
      "step": 29890
    },
    {
      "entropy": 5.984638977050781,
      "epoch": 2.612514200821463,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00043170609392611294,
      "loss": 5.6611,
      "mean_token_accuracy": 0.17112548649311066,
      "num_tokens": 60589893.0,
      "step": 29895
    },
    {
      "entropy": 5.9198899269104,
      "epoch": 2.6129511491741675,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043168373170467735,
      "loss": 5.5323,
      "mean_token_accuracy": 0.18360403776168824,
      "num_tokens": 60599406.0,
      "step": 29900
    },
    {
      "entropy": 5.855995845794678,
      "epoch": 2.6133880975268724,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043166136647796577,
      "loss": 5.6339,
      "mean_token_accuracy": 0.1887843444943428,
      "num_tokens": 60610435.0,
      "step": 29905
    },
    {
      "entropy": 6.0017396926879885,
      "epoch": 2.613825045879577,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000431638998246407,
      "loss": 5.6017,
      "mean_token_accuracy": 0.17320692390203477,
      "num_tokens": 60621405.0,
      "step": 29910
    },
    {
      "entropy": 6.022151803970337,
      "epoch": 2.614261994232282,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004316166270104303,
      "loss": 5.6035,
      "mean_token_accuracy": 0.17173578441143036,
      "num_tokens": 60632119.0,
      "step": 29915
    },
    {
      "entropy": 6.008257055282593,
      "epoch": 2.6146989425849867,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043159425277046454,
      "loss": 5.5651,
      "mean_token_accuracy": 0.17841157615184783,
      "num_tokens": 60642163.0,
      "step": 29920
    },
    {
      "entropy": 5.8869000434875485,
      "epoch": 2.615135890937691,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000431571875526939,
      "loss": 5.5565,
      "mean_token_accuracy": 0.17900291383266448,
      "num_tokens": 60652050.0,
      "step": 29925
    },
    {
      "entropy": 5.907185745239258,
      "epoch": 2.6155728392903956,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004315494952802828,
      "loss": 5.5596,
      "mean_token_accuracy": 0.1765395447611809,
      "num_tokens": 60661843.0,
      "step": 29930
    },
    {
      "entropy": 5.970516967773437,
      "epoch": 2.6160097876431005,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043152711203092524,
      "loss": 5.5219,
      "mean_token_accuracy": 0.18412061184644699,
      "num_tokens": 60671618.0,
      "step": 29935
    },
    {
      "entropy": 6.00994644165039,
      "epoch": 2.6164467359958055,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004315047257792957,
      "loss": 5.7093,
      "mean_token_accuracy": 0.16883492320775986,
      "num_tokens": 60682930.0,
      "step": 29940
    },
    {
      "entropy": 6.0838274478912355,
      "epoch": 2.61688368434851,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004314823365258235,
      "loss": 5.6394,
      "mean_token_accuracy": 0.16916418373584746,
      "num_tokens": 60692864.0,
      "step": 29945
    },
    {
      "entropy": 6.002289247512818,
      "epoch": 2.617320632701215,
      "grad_norm": 1.46875,
      "learning_rate": 0.00043145994427093806,
      "loss": 5.6155,
      "mean_token_accuracy": 0.1739310011267662,
      "num_tokens": 60702934.0,
      "step": 29950
    },
    {
      "entropy": 5.96088662147522,
      "epoch": 2.6177575810539193,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043143754901506893,
      "loss": 5.5754,
      "mean_token_accuracy": 0.17063530832529067,
      "num_tokens": 60712289.0,
      "step": 29955
    },
    {
      "entropy": 5.85986180305481,
      "epoch": 2.6181945294066242,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043141515075864563,
      "loss": 5.5455,
      "mean_token_accuracy": 0.1795621931552887,
      "num_tokens": 60723046.0,
      "step": 29960
    },
    {
      "entropy": 6.000193786621094,
      "epoch": 2.6186314777593287,
      "grad_norm": 1.25,
      "learning_rate": 0.00043139274950209776,
      "loss": 5.6234,
      "mean_token_accuracy": 0.1681890904903412,
      "num_tokens": 60731969.0,
      "step": 29965
    },
    {
      "entropy": 6.071108388900757,
      "epoch": 2.6190684261120336,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043137034524585506,
      "loss": 5.7047,
      "mean_token_accuracy": 0.1664678394794464,
      "num_tokens": 60743021.0,
      "step": 29970
    },
    {
      "entropy": 6.009530401229858,
      "epoch": 2.6195053744647385,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004313479379903472,
      "loss": 5.6104,
      "mean_token_accuracy": 0.17357812970876693,
      "num_tokens": 60751854.0,
      "step": 29975
    },
    {
      "entropy": 6.04745717048645,
      "epoch": 2.619942322817443,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004313255277360039,
      "loss": 5.5787,
      "mean_token_accuracy": 0.17341050803661345,
      "num_tokens": 60761572.0,
      "step": 29980
    },
    {
      "entropy": 5.930404329299927,
      "epoch": 2.6203792711701475,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004313031144832551,
      "loss": 5.5532,
      "mean_token_accuracy": 0.17552249729633332,
      "num_tokens": 60771618.0,
      "step": 29985
    },
    {
      "entropy": 5.946342849731446,
      "epoch": 2.6208162195228524,
      "grad_norm": 1.484375,
      "learning_rate": 0.00043128069823253064,
      "loss": 5.5514,
      "mean_token_accuracy": 0.1755268469452858,
      "num_tokens": 60781038.0,
      "step": 29990
    },
    {
      "entropy": 5.952841854095459,
      "epoch": 2.6212531678755573,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00043125827898426056,
      "loss": 5.6352,
      "mean_token_accuracy": 0.16931041330099106,
      "num_tokens": 60792433.0,
      "step": 29995
    },
    {
      "entropy": 5.943544197082519,
      "epoch": 2.6216901162282618,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043123585673887475,
      "loss": 5.5635,
      "mean_token_accuracy": 0.17772556096315384,
      "num_tokens": 60801995.0,
      "step": 30000
    },
    {
      "epoch": 2.6216901162282618,
      "eval_entropy": 5.750227942743955,
      "eval_loss": 5.656104564666748,
      "eval_mean_token_accuracy": 0.18080137163700155,
      "eval_num_tokens": 60801995.0,
      "eval_runtime": 142.367,
      "eval_samples_per_second": 259.99,
      "eval_steps_per_second": 32.501,
      "step": 30000
    },
    {
      "entropy": 5.973458242416382,
      "epoch": 2.6221270645809667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004312134314968034,
      "loss": 5.6036,
      "mean_token_accuracy": 0.17698222994804383,
      "num_tokens": 60811784.0,
      "step": 30005
    },
    {
      "entropy": 5.9458301067352295,
      "epoch": 2.622564012933671,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043119100325847654,
      "loss": 5.5519,
      "mean_token_accuracy": 0.181108558177948,
      "num_tokens": 60822018.0,
      "step": 30010
    },
    {
      "entropy": 5.864708948135376,
      "epoch": 2.623000961286376,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004311685720243243,
      "loss": 5.6007,
      "mean_token_accuracy": 0.1777154251933098,
      "num_tokens": 60832204.0,
      "step": 30015
    },
    {
      "entropy": 5.966365623474121,
      "epoch": 2.6234379096390805,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004311461377947772,
      "loss": 5.57,
      "mean_token_accuracy": 0.17675431519746782,
      "num_tokens": 60843293.0,
      "step": 30020
    },
    {
      "entropy": 6.034214448928833,
      "epoch": 2.6238748579917854,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004311237005702652,
      "loss": 5.557,
      "mean_token_accuracy": 0.17210794538259505,
      "num_tokens": 60852323.0,
      "step": 30025
    },
    {
      "entropy": 5.927505159378052,
      "epoch": 2.6243118063444903,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004311012603512189,
      "loss": 5.5802,
      "mean_token_accuracy": 0.17559949159622193,
      "num_tokens": 60862788.0,
      "step": 30030
    },
    {
      "entropy": 5.930975151062012,
      "epoch": 2.624748754697195,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043107881713806863,
      "loss": 5.6055,
      "mean_token_accuracy": 0.17772463113069534,
      "num_tokens": 60872793.0,
      "step": 30035
    },
    {
      "entropy": 5.927507829666138,
      "epoch": 2.6251857030498993,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00043105637093124484,
      "loss": 5.5516,
      "mean_token_accuracy": 0.18308677524328232,
      "num_tokens": 60882712.0,
      "step": 30040
    },
    {
      "entropy": 6.013456773757935,
      "epoch": 2.625622651402604,
      "grad_norm": 1.65625,
      "learning_rate": 0.00043103392173117814,
      "loss": 5.6104,
      "mean_token_accuracy": 0.1777219444513321,
      "num_tokens": 60892400.0,
      "step": 30045
    },
    {
      "entropy": 5.962447881698608,
      "epoch": 2.626059599755309,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043101146953829895,
      "loss": 5.5502,
      "mean_token_accuracy": 0.1834649547934532,
      "num_tokens": 60902480.0,
      "step": 30050
    },
    {
      "entropy": 5.913914346694947,
      "epoch": 2.6264965481080136,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004309890143530382,
      "loss": 5.609,
      "mean_token_accuracy": 0.17889604419469834,
      "num_tokens": 60912183.0,
      "step": 30055
    },
    {
      "entropy": 6.042778730392456,
      "epoch": 2.6269334964607185,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004309665561758262,
      "loss": 5.6003,
      "mean_token_accuracy": 0.17944021373987198,
      "num_tokens": 60922166.0,
      "step": 30060
    },
    {
      "entropy": 5.922384643554688,
      "epoch": 2.627370444813423,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043094409500709404,
      "loss": 5.5386,
      "mean_token_accuracy": 0.1790107935667038,
      "num_tokens": 60931980.0,
      "step": 30065
    },
    {
      "entropy": 5.925993347167969,
      "epoch": 2.627807393166128,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004309216308472725,
      "loss": 5.738,
      "mean_token_accuracy": 0.17226550579071045,
      "num_tokens": 60943382.0,
      "step": 30070
    },
    {
      "entropy": 6.0251631259918215,
      "epoch": 2.6282443415188323,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004308991636967923,
      "loss": 5.6337,
      "mean_token_accuracy": 0.16963892430067062,
      "num_tokens": 60953505.0,
      "step": 30075
    },
    {
      "entropy": 6.061402177810669,
      "epoch": 2.6286812898715373,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004308766935560844,
      "loss": 5.6221,
      "mean_token_accuracy": 0.17309316098690034,
      "num_tokens": 60963851.0,
      "step": 30080
    },
    {
      "entropy": 5.97359528541565,
      "epoch": 2.6291182382242417,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043085422042557985,
      "loss": 5.6377,
      "mean_token_accuracy": 0.17425746470689774,
      "num_tokens": 60974919.0,
      "step": 30085
    },
    {
      "entropy": 6.002991628646851,
      "epoch": 2.6295551865769466,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043083174430570974,
      "loss": 5.6301,
      "mean_token_accuracy": 0.17399934828281402,
      "num_tokens": 60983866.0,
      "step": 30090
    },
    {
      "entropy": 5.917350482940674,
      "epoch": 2.629992134929651,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043080926519690517,
      "loss": 5.5635,
      "mean_token_accuracy": 0.1753796875476837,
      "num_tokens": 60994393.0,
      "step": 30095
    },
    {
      "entropy": 6.045995140075684,
      "epoch": 2.630429083282356,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043078678309959716,
      "loss": 5.6468,
      "mean_token_accuracy": 0.1804478347301483,
      "num_tokens": 61004634.0,
      "step": 30100
    },
    {
      "entropy": 5.920228576660156,
      "epoch": 2.630866031635061,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004307642980142171,
      "loss": 5.5752,
      "mean_token_accuracy": 0.17953124791383743,
      "num_tokens": 61014771.0,
      "step": 30105
    },
    {
      "entropy": 5.939391088485718,
      "epoch": 2.6313029799877654,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043074180994119604,
      "loss": 5.6275,
      "mean_token_accuracy": 0.1783164009451866,
      "num_tokens": 61025213.0,
      "step": 30110
    },
    {
      "entropy": 5.992785120010376,
      "epoch": 2.63173992834047,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004307193188809656,
      "loss": 5.5289,
      "mean_token_accuracy": 0.18132143765687941,
      "num_tokens": 61035601.0,
      "step": 30115
    },
    {
      "entropy": 5.863493728637695,
      "epoch": 2.632176876693175,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043069682483395694,
      "loss": 5.5721,
      "mean_token_accuracy": 0.17905950993299485,
      "num_tokens": 61045862.0,
      "step": 30120
    },
    {
      "entropy": 5.9854971408844,
      "epoch": 2.6326138250458797,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043067432780060163,
      "loss": 5.5855,
      "mean_token_accuracy": 0.1812186524271965,
      "num_tokens": 61056702.0,
      "step": 30125
    },
    {
      "entropy": 5.9612658500671385,
      "epoch": 2.633050773398584,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043065182778133116,
      "loss": 5.6377,
      "mean_token_accuracy": 0.17410535365343094,
      "num_tokens": 61066353.0,
      "step": 30130
    },
    {
      "entropy": 5.923985624313355,
      "epoch": 2.633487721751289,
      "grad_norm": 1.203125,
      "learning_rate": 0.000430629324776577,
      "loss": 5.5811,
      "mean_token_accuracy": 0.1779581770300865,
      "num_tokens": 61076534.0,
      "step": 30135
    },
    {
      "entropy": 5.928305435180664,
      "epoch": 2.6339246701039936,
      "grad_norm": 1.375,
      "learning_rate": 0.00043060681878677096,
      "loss": 5.4746,
      "mean_token_accuracy": 0.18550718873739241,
      "num_tokens": 61086635.0,
      "step": 30140
    },
    {
      "entropy": 5.925268125534058,
      "epoch": 2.6343616184566985,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043058430981234453,
      "loss": 5.6014,
      "mean_token_accuracy": 0.16952937245368957,
      "num_tokens": 61096605.0,
      "step": 30145
    },
    {
      "entropy": 5.9285862922668455,
      "epoch": 2.634798566809403,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004305617978537295,
      "loss": 5.5606,
      "mean_token_accuracy": 0.17875394821166993,
      "num_tokens": 61106465.0,
      "step": 30150
    },
    {
      "entropy": 5.965999412536621,
      "epoch": 2.635235515162108,
      "grad_norm": 1.375,
      "learning_rate": 0.0004305392829113577,
      "loss": 5.6503,
      "mean_token_accuracy": 0.18031575828790664,
      "num_tokens": 61116063.0,
      "step": 30155
    },
    {
      "entropy": 5.988164186477661,
      "epoch": 2.6356724635148128,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00043051676498566095,
      "loss": 5.6173,
      "mean_token_accuracy": 0.17239371836185455,
      "num_tokens": 61125794.0,
      "step": 30160
    },
    {
      "entropy": 5.954061508178711,
      "epoch": 2.6361094118675172,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004304942440770712,
      "loss": 5.6196,
      "mean_token_accuracy": 0.17145203202962875,
      "num_tokens": 61135734.0,
      "step": 30165
    },
    {
      "entropy": 5.950304317474365,
      "epoch": 2.6365463602202217,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004304717201860203,
      "loss": 5.5626,
      "mean_token_accuracy": 0.18341901153326035,
      "num_tokens": 61146071.0,
      "step": 30170
    },
    {
      "entropy": 5.892476177215576,
      "epoch": 2.6369833085729266,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004304491933129404,
      "loss": 5.5525,
      "mean_token_accuracy": 0.18119359463453294,
      "num_tokens": 61154931.0,
      "step": 30175
    },
    {
      "entropy": 5.93414077758789,
      "epoch": 2.6374202569256315,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004304266634582634,
      "loss": 5.637,
      "mean_token_accuracy": 0.17461657673120498,
      "num_tokens": 61164464.0,
      "step": 30180
    },
    {
      "entropy": 6.018046903610229,
      "epoch": 2.637857205278336,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004304041306224217,
      "loss": 5.6323,
      "mean_token_accuracy": 0.1753997728228569,
      "num_tokens": 61175379.0,
      "step": 30185
    },
    {
      "entropy": 5.935977840423584,
      "epoch": 2.638294153631041,
      "grad_norm": 1.953125,
      "learning_rate": 0.00043038159480584723,
      "loss": 5.538,
      "mean_token_accuracy": 0.18143924325704575,
      "num_tokens": 61186340.0,
      "step": 30190
    },
    {
      "entropy": 5.927707672119141,
      "epoch": 2.6387311019837454,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043035905600897237,
      "loss": 5.5363,
      "mean_token_accuracy": 0.1841174528002739,
      "num_tokens": 61196495.0,
      "step": 30195
    },
    {
      "entropy": 5.971861553192139,
      "epoch": 2.6391680503364503,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043033651423222947,
      "loss": 5.6511,
      "mean_token_accuracy": 0.17011522501707077,
      "num_tokens": 61206555.0,
      "step": 30200
    },
    {
      "entropy": 5.973488855361938,
      "epoch": 2.6396049986891548,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004303139694760508,
      "loss": 5.6493,
      "mean_token_accuracy": 0.17146132439374923,
      "num_tokens": 61215818.0,
      "step": 30205
    },
    {
      "entropy": 5.984538793563843,
      "epoch": 2.6400419470418597,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043029142174086875,
      "loss": 5.5902,
      "mean_token_accuracy": 0.17661833614110947,
      "num_tokens": 61226325.0,
      "step": 30210
    },
    {
      "entropy": 5.979688882827759,
      "epoch": 2.6404788953945646,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004302688710271159,
      "loss": 5.6012,
      "mean_token_accuracy": 0.1756158635020256,
      "num_tokens": 61236179.0,
      "step": 30215
    },
    {
      "entropy": 5.880737686157227,
      "epoch": 2.640915843747269,
      "grad_norm": 2.140625,
      "learning_rate": 0.00043024631733522484,
      "loss": 5.5497,
      "mean_token_accuracy": 0.18358119130134581,
      "num_tokens": 61245680.0,
      "step": 30220
    },
    {
      "entropy": 5.97234468460083,
      "epoch": 2.6413527920999735,
      "grad_norm": 1.53125,
      "learning_rate": 0.000430223760665628,
      "loss": 5.6356,
      "mean_token_accuracy": 0.17362118661403655,
      "num_tokens": 61256369.0,
      "step": 30225
    },
    {
      "entropy": 5.887058877944947,
      "epoch": 2.6417897404526784,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043020120101875807,
      "loss": 5.5124,
      "mean_token_accuracy": 0.1795685589313507,
      "num_tokens": 61266395.0,
      "step": 30230
    },
    {
      "entropy": 5.970242309570312,
      "epoch": 2.6422266888053834,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043017863839504784,
      "loss": 5.6421,
      "mean_token_accuracy": 0.17814180254936218,
      "num_tokens": 61276442.0,
      "step": 30235
    },
    {
      "entropy": 5.98960428237915,
      "epoch": 2.642663637158088,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043015607279493,
      "loss": 5.6694,
      "mean_token_accuracy": 0.16733269840478898,
      "num_tokens": 61286230.0,
      "step": 30240
    },
    {
      "entropy": 5.925936508178711,
      "epoch": 2.6431005855107927,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004301335042188373,
      "loss": 5.5636,
      "mean_token_accuracy": 0.1793666437268257,
      "num_tokens": 61297853.0,
      "step": 30245
    },
    {
      "entropy": 5.9232221126556395,
      "epoch": 2.643537533863497,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043011093266720273,
      "loss": 5.5093,
      "mean_token_accuracy": 0.18649578392505645,
      "num_tokens": 61307976.0,
      "step": 30250
    },
    {
      "entropy": 5.969130420684815,
      "epoch": 2.643974482216202,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043008835814045926,
      "loss": 5.6159,
      "mean_token_accuracy": 0.17565037310123444,
      "num_tokens": 61319101.0,
      "step": 30255
    },
    {
      "entropy": 5.990470886230469,
      "epoch": 2.6444114305689066,
      "grad_norm": 1.125,
      "learning_rate": 0.0004300657806390398,
      "loss": 5.5928,
      "mean_token_accuracy": 0.17908360213041305,
      "num_tokens": 61329492.0,
      "step": 30260
    },
    {
      "entropy": 5.974785661697387,
      "epoch": 2.6448483789216115,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043004320016337737,
      "loss": 5.6574,
      "mean_token_accuracy": 0.17148157805204392,
      "num_tokens": 61340480.0,
      "step": 30265
    },
    {
      "entropy": 5.942785024642944,
      "epoch": 2.6452853272743164,
      "grad_norm": 1.125,
      "learning_rate": 0.00043002061671390513,
      "loss": 5.5575,
      "mean_token_accuracy": 0.17422754764556886,
      "num_tokens": 61350572.0,
      "step": 30270
    },
    {
      "entropy": 5.9907409191131595,
      "epoch": 2.645722275627021,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004299980302910563,
      "loss": 5.6237,
      "mean_token_accuracy": 0.17612258046865464,
      "num_tokens": 61361329.0,
      "step": 30275
    },
    {
      "entropy": 6.046935176849365,
      "epoch": 2.6461592239797254,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004299754408952639,
      "loss": 5.6894,
      "mean_token_accuracy": 0.16902158707380294,
      "num_tokens": 61372618.0,
      "step": 30280
    },
    {
      "entropy": 6.039435005187988,
      "epoch": 2.6465961723324303,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004299528485269614,
      "loss": 5.7156,
      "mean_token_accuracy": 0.16352979093790054,
      "num_tokens": 61383831.0,
      "step": 30285
    },
    {
      "entropy": 5.900883436203003,
      "epoch": 2.647033120685135,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004299302531865821,
      "loss": 5.5713,
      "mean_token_accuracy": 0.17856715470552445,
      "num_tokens": 61394527.0,
      "step": 30290
    },
    {
      "entropy": 6.009081029891968,
      "epoch": 2.6474700690378397,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042990765487455935,
      "loss": 5.6837,
      "mean_token_accuracy": 0.17115160673856736,
      "num_tokens": 61404787.0,
      "step": 30295
    },
    {
      "entropy": 5.9336591243743895,
      "epoch": 2.6479070173905446,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004298850535913266,
      "loss": 5.4732,
      "mean_token_accuracy": 0.18721867650747298,
      "num_tokens": 61413922.0,
      "step": 30300
    },
    {
      "entropy": 5.98165602684021,
      "epoch": 2.648343965743249,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004298624493373174,
      "loss": 5.5755,
      "mean_token_accuracy": 0.17764462530612946,
      "num_tokens": 61424750.0,
      "step": 30305
    },
    {
      "entropy": 5.918058443069458,
      "epoch": 2.648780914095954,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004298398421129653,
      "loss": 5.5354,
      "mean_token_accuracy": 0.18129244893789292,
      "num_tokens": 61435222.0,
      "step": 30310
    },
    {
      "entropy": 5.966127967834472,
      "epoch": 2.6492178624486584,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004298172319187038,
      "loss": 5.5997,
      "mean_token_accuracy": 0.18717074990272523,
      "num_tokens": 61444230.0,
      "step": 30315
    },
    {
      "entropy": 5.96799602508545,
      "epoch": 2.6496548108013633,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004297946187549667,
      "loss": 5.5913,
      "mean_token_accuracy": 0.1690017580986023,
      "num_tokens": 61454975.0,
      "step": 30320
    },
    {
      "entropy": 5.978246974945068,
      "epoch": 2.6500917591540682,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042977200262218776,
      "loss": 5.6763,
      "mean_token_accuracy": 0.17065757066011428,
      "num_tokens": 61465563.0,
      "step": 30325
    },
    {
      "entropy": 5.984267044067383,
      "epoch": 2.6505287075067727,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004297493835208006,
      "loss": 5.6155,
      "mean_token_accuracy": 0.17437743693590163,
      "num_tokens": 61475330.0,
      "step": 30330
    },
    {
      "entropy": 5.961198997497559,
      "epoch": 2.650965655859477,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042972676145123926,
      "loss": 5.636,
      "mean_token_accuracy": 0.16830242425203323,
      "num_tokens": 61485577.0,
      "step": 30335
    },
    {
      "entropy": 6.015182113647461,
      "epoch": 2.651402604212182,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042970413641393754,
      "loss": 5.5652,
      "mean_token_accuracy": 0.1742838814854622,
      "num_tokens": 61495502.0,
      "step": 30340
    },
    {
      "entropy": 6.010659646987915,
      "epoch": 2.651839552564887,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004296815084093294,
      "loss": 5.6294,
      "mean_token_accuracy": 0.17167782187461852,
      "num_tokens": 61505383.0,
      "step": 30345
    },
    {
      "entropy": 5.915720415115357,
      "epoch": 2.6522765009175915,
      "grad_norm": 1.125,
      "learning_rate": 0.00042965887743784886,
      "loss": 5.5872,
      "mean_token_accuracy": 0.1827102929353714,
      "num_tokens": 61515052.0,
      "step": 30350
    },
    {
      "entropy": 5.966127634048462,
      "epoch": 2.6527134492702964,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004296362434999301,
      "loss": 5.5457,
      "mean_token_accuracy": 0.18219731748104095,
      "num_tokens": 61525139.0,
      "step": 30355
    },
    {
      "entropy": 5.905089855194092,
      "epoch": 2.653150397623001,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042961360659600706,
      "loss": 5.498,
      "mean_token_accuracy": 0.18247778117656707,
      "num_tokens": 61534901.0,
      "step": 30360
    },
    {
      "entropy": 5.930221891403198,
      "epoch": 2.653587345975706,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004295909667265141,
      "loss": 5.5946,
      "mean_token_accuracy": 0.17066411525011063,
      "num_tokens": 61543258.0,
      "step": 30365
    },
    {
      "entropy": 5.94978141784668,
      "epoch": 2.6540242943284102,
      "grad_norm": 1.25,
      "learning_rate": 0.00042956832389188534,
      "loss": 5.5681,
      "mean_token_accuracy": 0.17426714897155762,
      "num_tokens": 61553294.0,
      "step": 30370
    },
    {
      "entropy": 5.962167549133301,
      "epoch": 2.654461242681115,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004295456780925551,
      "loss": 5.5608,
      "mean_token_accuracy": 0.18456683158874512,
      "num_tokens": 61562986.0,
      "step": 30375
    },
    {
      "entropy": 5.904802417755127,
      "epoch": 2.65489819103382,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042952302932895784,
      "loss": 5.6239,
      "mean_token_accuracy": 0.18226825445890427,
      "num_tokens": 61573994.0,
      "step": 30380
    },
    {
      "entropy": 5.954628849029541,
      "epoch": 2.6553351393865245,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042950037760152793,
      "loss": 5.5555,
      "mean_token_accuracy": 0.18114988803863524,
      "num_tokens": 61584279.0,
      "step": 30385
    },
    {
      "entropy": 6.030858564376831,
      "epoch": 2.655772087739229,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004294777229106998,
      "loss": 5.6636,
      "mean_token_accuracy": 0.17198052257299423,
      "num_tokens": 61595188.0,
      "step": 30390
    },
    {
      "entropy": 5.951523542404175,
      "epoch": 2.656209036091934,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000429455065256908,
      "loss": 5.6747,
      "mean_token_accuracy": 0.17314059436321258,
      "num_tokens": 61605321.0,
      "step": 30395
    },
    {
      "entropy": 6.048943948745728,
      "epoch": 2.656645984444639,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004294324046405871,
      "loss": 5.6228,
      "mean_token_accuracy": 0.1715651735663414,
      "num_tokens": 61615137.0,
      "step": 30400
    },
    {
      "entropy": 5.929684162139893,
      "epoch": 2.6570829327973433,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042940974106217176,
      "loss": 5.5018,
      "mean_token_accuracy": 0.18061655461788179,
      "num_tokens": 61625912.0,
      "step": 30405
    },
    {
      "entropy": 5.931441688537598,
      "epoch": 2.6575198811500482,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004293870745220967,
      "loss": 5.535,
      "mean_token_accuracy": 0.17986656427383424,
      "num_tokens": 61635836.0,
      "step": 30410
    },
    {
      "entropy": 5.956988334655762,
      "epoch": 2.6579568295027527,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004293644050207966,
      "loss": 5.5802,
      "mean_token_accuracy": 0.17714661210775376,
      "num_tokens": 61645151.0,
      "step": 30415
    },
    {
      "entropy": 6.019709682464599,
      "epoch": 2.6583937778554576,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004293417325587064,
      "loss": 5.6244,
      "mean_token_accuracy": 0.17361649125814438,
      "num_tokens": 61654266.0,
      "step": 30420
    },
    {
      "entropy": 5.982089900970459,
      "epoch": 2.658830726208162,
      "grad_norm": 1.25,
      "learning_rate": 0.0004293190571362608,
      "loss": 5.7624,
      "mean_token_accuracy": 0.16313165724277495,
      "num_tokens": 61664355.0,
      "step": 30425
    },
    {
      "entropy": 6.0138609409332275,
      "epoch": 2.659267674560867,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00042929637875389486,
      "loss": 5.6624,
      "mean_token_accuracy": 0.1748173750936985,
      "num_tokens": 61674130.0,
      "step": 30430
    },
    {
      "entropy": 6.057560157775879,
      "epoch": 2.6597046229135715,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042927369741204357,
      "loss": 5.639,
      "mean_token_accuracy": 0.1718577638268471,
      "num_tokens": 61683612.0,
      "step": 30435
    },
    {
      "entropy": 5.92491397857666,
      "epoch": 2.6601415712662764,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004292510131111419,
      "loss": 5.6184,
      "mean_token_accuracy": 0.18457395136356353,
      "num_tokens": 61693156.0,
      "step": 30440
    },
    {
      "entropy": 5.883712720870972,
      "epoch": 2.660578519618981,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042922832585162496,
      "loss": 5.4812,
      "mean_token_accuracy": 0.18263590335845947,
      "num_tokens": 61701925.0,
      "step": 30445
    },
    {
      "entropy": 5.964308595657348,
      "epoch": 2.6610154679716858,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042920563563392803,
      "loss": 5.5828,
      "mean_token_accuracy": 0.18096308708190917,
      "num_tokens": 61712388.0,
      "step": 30450
    },
    {
      "entropy": 5.978031158447266,
      "epoch": 2.6614524163243907,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042918294245848605,
      "loss": 5.5767,
      "mean_token_accuracy": 0.1757027953863144,
      "num_tokens": 61722402.0,
      "step": 30455
    },
    {
      "entropy": 5.988435506820679,
      "epoch": 2.661889364677095,
      "grad_norm": 1.125,
      "learning_rate": 0.00042916024632573454,
      "loss": 5.6653,
      "mean_token_accuracy": 0.17371317446231843,
      "num_tokens": 61733160.0,
      "step": 30460
    },
    {
      "entropy": 5.962430477142334,
      "epoch": 2.6623263130297996,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042913754723610865,
      "loss": 5.5763,
      "mean_token_accuracy": 0.1816481053829193,
      "num_tokens": 61742536.0,
      "step": 30465
    },
    {
      "entropy": 5.960424137115479,
      "epoch": 2.6627632613825045,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004291148451900439,
      "loss": 5.613,
      "mean_token_accuracy": 0.17377733290195466,
      "num_tokens": 61753439.0,
      "step": 30470
    },
    {
      "entropy": 5.971997261047363,
      "epoch": 2.6632002097352094,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042909214018797565,
      "loss": 5.6006,
      "mean_token_accuracy": 0.18096148371696472,
      "num_tokens": 61763994.0,
      "step": 30475
    },
    {
      "entropy": 6.016799116134644,
      "epoch": 2.663637158087914,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004290694322303394,
      "loss": 5.6284,
      "mean_token_accuracy": 0.17475537210702896,
      "num_tokens": 61774082.0,
      "step": 30480
    },
    {
      "entropy": 5.963512802124024,
      "epoch": 2.664074106440619,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042904672131757065,
      "loss": 5.5624,
      "mean_token_accuracy": 0.1830036610364914,
      "num_tokens": 61784528.0,
      "step": 30485
    },
    {
      "entropy": 5.908842325210571,
      "epoch": 2.6645110547933233,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004290240074501051,
      "loss": 5.5786,
      "mean_token_accuracy": 0.17870448678731918,
      "num_tokens": 61794267.0,
      "step": 30490
    },
    {
      "entropy": 5.974799633026123,
      "epoch": 2.664948003146028,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004290012906283784,
      "loss": 5.5775,
      "mean_token_accuracy": 0.17906295508146286,
      "num_tokens": 61804842.0,
      "step": 30495
    },
    {
      "entropy": 5.966756153106689,
      "epoch": 2.6653849514987327,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042897857085282627,
      "loss": 5.5934,
      "mean_token_accuracy": 0.17286344468593598,
      "num_tokens": 61814316.0,
      "step": 30500
    },
    {
      "entropy": 5.926824140548706,
      "epoch": 2.6658218998514376,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042895584812388445,
      "loss": 5.5811,
      "mean_token_accuracy": 0.17756069600582122,
      "num_tokens": 61823828.0,
      "step": 30505
    },
    {
      "entropy": 5.938792085647583,
      "epoch": 2.6662588482041425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004289331224419888,
      "loss": 5.6134,
      "mean_token_accuracy": 0.17765748947858812,
      "num_tokens": 61833878.0,
      "step": 30510
    },
    {
      "entropy": 5.9140214920043945,
      "epoch": 2.666695796556847,
      "grad_norm": 1.578125,
      "learning_rate": 0.00042891039380757516,
      "loss": 5.5454,
      "mean_token_accuracy": 0.17951704412698746,
      "num_tokens": 61844591.0,
      "step": 30515
    },
    {
      "entropy": 6.009043025970459,
      "epoch": 2.6671327449095514,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042888766222107954,
      "loss": 5.6259,
      "mean_token_accuracy": 0.17707489430904388,
      "num_tokens": 61854791.0,
      "step": 30520
    },
    {
      "entropy": 5.969103765487671,
      "epoch": 2.6675696932622563,
      "grad_norm": 1.15625,
      "learning_rate": 0.000428864927682938,
      "loss": 5.6346,
      "mean_token_accuracy": 0.17631199657917024,
      "num_tokens": 61865194.0,
      "step": 30525
    },
    {
      "entropy": 5.911246204376221,
      "epoch": 2.6680066416149613,
      "grad_norm": 1.4375,
      "learning_rate": 0.00042884219019358644,
      "loss": 5.5491,
      "mean_token_accuracy": 0.17105466574430467,
      "num_tokens": 61874827.0,
      "step": 30530
    },
    {
      "entropy": 6.000005102157592,
      "epoch": 2.6684435899676657,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004288194497534611,
      "loss": 5.6255,
      "mean_token_accuracy": 0.17016498893499374,
      "num_tokens": 61885596.0,
      "step": 30535
    },
    {
      "entropy": 5.913864040374756,
      "epoch": 2.6688805383203706,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042879670636299805,
      "loss": 5.4566,
      "mean_token_accuracy": 0.1839867815375328,
      "num_tokens": 61895732.0,
      "step": 30540
    },
    {
      "entropy": 5.873208522796631,
      "epoch": 2.669317486673075,
      "grad_norm": 1.125,
      "learning_rate": 0.00042877396002263357,
      "loss": 5.5517,
      "mean_token_accuracy": 0.17559255957603453,
      "num_tokens": 61905581.0,
      "step": 30545
    },
    {
      "entropy": 5.891533422470093,
      "epoch": 2.66975443502578,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042875121073280404,
      "loss": 5.5241,
      "mean_token_accuracy": 0.18273208141326905,
      "num_tokens": 61914926.0,
      "step": 30550
    },
    {
      "entropy": 5.957119941711426,
      "epoch": 2.6701913833784845,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042872845849394563,
      "loss": 5.5979,
      "mean_token_accuracy": 0.17878122329711915,
      "num_tokens": 61926501.0,
      "step": 30555
    },
    {
      "entropy": 6.024358129501342,
      "epoch": 2.6706283317311894,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042870570330649495,
      "loss": 5.6525,
      "mean_token_accuracy": 0.17174504697322845,
      "num_tokens": 61937161.0,
      "step": 30560
    },
    {
      "entropy": 5.953457021713257,
      "epoch": 2.6710652800838943,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004286829451708883,
      "loss": 5.546,
      "mean_token_accuracy": 0.1782270759344101,
      "num_tokens": 61949125.0,
      "step": 30565
    },
    {
      "entropy": 5.901925325393677,
      "epoch": 2.671502228436599,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004286601840875622,
      "loss": 5.4591,
      "mean_token_accuracy": 0.18377489745616912,
      "num_tokens": 61958110.0,
      "step": 30570
    },
    {
      "entropy": 5.912994003295898,
      "epoch": 2.6719391767893033,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004286374200569533,
      "loss": 5.5708,
      "mean_token_accuracy": 0.17989026010036469,
      "num_tokens": 61967805.0,
      "step": 30575
    },
    {
      "entropy": 5.942583894729614,
      "epoch": 2.672376125142008,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042861465307949815,
      "loss": 5.5931,
      "mean_token_accuracy": 0.1803122267127037,
      "num_tokens": 61977189.0,
      "step": 30580
    },
    {
      "entropy": 6.019409799575806,
      "epoch": 2.672813073494713,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042859188315563345,
      "loss": 5.6877,
      "mean_token_accuracy": 0.17310856729745866,
      "num_tokens": 61988568.0,
      "step": 30585
    },
    {
      "entropy": 5.953616952896118,
      "epoch": 2.6732500218474176,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042856911028579597,
      "loss": 5.5391,
      "mean_token_accuracy": 0.17790040969848633,
      "num_tokens": 61999316.0,
      "step": 30590
    },
    {
      "entropy": 5.9339759826660154,
      "epoch": 2.6736869702001225,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042854633447042247,
      "loss": 5.6031,
      "mean_token_accuracy": 0.17756938934326172,
      "num_tokens": 62009345.0,
      "step": 30595
    },
    {
      "entropy": 5.976014804840088,
      "epoch": 2.674123918552827,
      "grad_norm": 2.140625,
      "learning_rate": 0.00042852355570994983,
      "loss": 5.5743,
      "mean_token_accuracy": 0.17612008452415467,
      "num_tokens": 62020847.0,
      "step": 30600
    },
    {
      "entropy": 5.917794036865234,
      "epoch": 2.674560866905532,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042850077400481495,
      "loss": 5.5442,
      "mean_token_accuracy": 0.18121481388807298,
      "num_tokens": 62030506.0,
      "step": 30605
    },
    {
      "entropy": 6.00868444442749,
      "epoch": 2.6749978152582363,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042847798935545484,
      "loss": 5.681,
      "mean_token_accuracy": 0.1694047436118126,
      "num_tokens": 62040238.0,
      "step": 30610
    },
    {
      "entropy": 5.898471355438232,
      "epoch": 2.6754347636109412,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004284552017623064,
      "loss": 5.5592,
      "mean_token_accuracy": 0.17741965651512145,
      "num_tokens": 62050192.0,
      "step": 30615
    },
    {
      "entropy": 5.930722951889038,
      "epoch": 2.675871711963646,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004284324112258068,
      "loss": 5.5858,
      "mean_token_accuracy": 0.1868114173412323,
      "num_tokens": 62059475.0,
      "step": 30620
    },
    {
      "entropy": 5.9717694282531735,
      "epoch": 2.6763086603163506,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042840961774639314,
      "loss": 5.559,
      "mean_token_accuracy": 0.17362243086099624,
      "num_tokens": 62068799.0,
      "step": 30625
    },
    {
      "entropy": 5.951978158950806,
      "epoch": 2.676745608669055,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004283868213245026,
      "loss": 5.6405,
      "mean_token_accuracy": 0.17252342104911805,
      "num_tokens": 62078205.0,
      "step": 30630
    },
    {
      "entropy": 5.9378174304962155,
      "epoch": 2.67718255702176,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042836402196057255,
      "loss": 5.4983,
      "mean_token_accuracy": 0.18331816792488098,
      "num_tokens": 62088084.0,
      "step": 30635
    },
    {
      "entropy": 5.944503498077393,
      "epoch": 2.677619505374465,
      "grad_norm": 1.625,
      "learning_rate": 0.00042834121965504014,
      "loss": 5.5336,
      "mean_token_accuracy": 0.18286419808864593,
      "num_tokens": 62098709.0,
      "step": 30640
    },
    {
      "entropy": 5.861937475204468,
      "epoch": 2.6780564537271694,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004283184144083428,
      "loss": 5.5142,
      "mean_token_accuracy": 0.17825721800327302,
      "num_tokens": 62108949.0,
      "step": 30645
    },
    {
      "entropy": 5.965588569641113,
      "epoch": 2.6784934020798743,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042829560622091795,
      "loss": 5.5743,
      "mean_token_accuracy": 0.1772512450814247,
      "num_tokens": 62118922.0,
      "step": 30650
    },
    {
      "entropy": 5.97500696182251,
      "epoch": 2.6789303504325788,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042827279509320297,
      "loss": 5.5709,
      "mean_token_accuracy": 0.17606416940689087,
      "num_tokens": 62128396.0,
      "step": 30655
    },
    {
      "entropy": 5.971624088287354,
      "epoch": 2.6793672987852837,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004282499810256356,
      "loss": 5.6332,
      "mean_token_accuracy": 0.16670106798410417,
      "num_tokens": 62139223.0,
      "step": 30660
    },
    {
      "entropy": 5.960240888595581,
      "epoch": 2.679804247137988,
      "grad_norm": 1.90625,
      "learning_rate": 0.00042822716401865314,
      "loss": 5.5222,
      "mean_token_accuracy": 0.17797500044107437,
      "num_tokens": 62149432.0,
      "step": 30665
    },
    {
      "entropy": 5.991175413131714,
      "epoch": 2.680241195490693,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042820434407269347,
      "loss": 5.5942,
      "mean_token_accuracy": 0.17290016561746596,
      "num_tokens": 62160945.0,
      "step": 30670
    },
    {
      "entropy": 5.986046886444091,
      "epoch": 2.680678143843398,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004281815211881942,
      "loss": 5.5979,
      "mean_token_accuracy": 0.16784563213586806,
      "num_tokens": 62171514.0,
      "step": 30675
    },
    {
      "entropy": 5.972820377349853,
      "epoch": 2.6811150921961024,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004281586953655931,
      "loss": 5.6159,
      "mean_token_accuracy": 0.16960854232311248,
      "num_tokens": 62181011.0,
      "step": 30680
    },
    {
      "entropy": 5.9202601432800295,
      "epoch": 2.681552040548807,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004281358666053279,
      "loss": 5.5657,
      "mean_token_accuracy": 0.1859351336956024,
      "num_tokens": 62190915.0,
      "step": 30685
    },
    {
      "entropy": 5.887601852416992,
      "epoch": 2.681988988901512,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004281130349078366,
      "loss": 5.5823,
      "mean_token_accuracy": 0.17297204285860063,
      "num_tokens": 62200660.0,
      "step": 30690
    },
    {
      "entropy": 5.893392610549927,
      "epoch": 2.6824259372542167,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000428090200273557,
      "loss": 5.5745,
      "mean_token_accuracy": 0.1777808204293251,
      "num_tokens": 62210425.0,
      "step": 30695
    },
    {
      "entropy": 6.006692790985108,
      "epoch": 2.682862885606921,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004280673627029272,
      "loss": 5.5441,
      "mean_token_accuracy": 0.17800808399915696,
      "num_tokens": 62220921.0,
      "step": 30700
    },
    {
      "entropy": 5.968232297897339,
      "epoch": 2.683299833959626,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004280445221963852,
      "loss": 5.5723,
      "mean_token_accuracy": 0.17674862444400788,
      "num_tokens": 62230561.0,
      "step": 30705
    },
    {
      "entropy": 5.994080114364624,
      "epoch": 2.6837367823123306,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000428021678754369,
      "loss": 5.6415,
      "mean_token_accuracy": 0.1753518983721733,
      "num_tokens": 62240744.0,
      "step": 30710
    },
    {
      "entropy": 5.893545055389405,
      "epoch": 2.6841737306650355,
      "grad_norm": 1.53125,
      "learning_rate": 0.00042799883237731683,
      "loss": 5.5058,
      "mean_token_accuracy": 0.18096552044153214,
      "num_tokens": 62250902.0,
      "step": 30715
    },
    {
      "entropy": 5.936309909820556,
      "epoch": 2.68461067901774,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042797598306566694,
      "loss": 5.6278,
      "mean_token_accuracy": 0.17816819995641708,
      "num_tokens": 62260459.0,
      "step": 30720
    },
    {
      "entropy": 5.962704658508301,
      "epoch": 2.685047627370445,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004279531308198575,
      "loss": 5.4923,
      "mean_token_accuracy": 0.17987795174121857,
      "num_tokens": 62269001.0,
      "step": 30725
    },
    {
      "entropy": 5.8662642478942875,
      "epoch": 2.6854845757231494,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042793027564032684,
      "loss": 5.5169,
      "mean_token_accuracy": 0.18019642680883408,
      "num_tokens": 62279197.0,
      "step": 30730
    },
    {
      "entropy": 5.873924779891968,
      "epoch": 2.6859215240758543,
      "grad_norm": 1.484375,
      "learning_rate": 0.00042790741752751337,
      "loss": 5.517,
      "mean_token_accuracy": 0.1767818585038185,
      "num_tokens": 62289229.0,
      "step": 30735
    },
    {
      "entropy": 5.894088554382324,
      "epoch": 2.6863584724285587,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042788455648185556,
      "loss": 5.5676,
      "mean_token_accuracy": 0.18683070242404937,
      "num_tokens": 62298819.0,
      "step": 30740
    },
    {
      "entropy": 5.957929706573486,
      "epoch": 2.6867954207812637,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042786169250379175,
      "loss": 5.5525,
      "mean_token_accuracy": 0.18023598790168763,
      "num_tokens": 62308567.0,
      "step": 30745
    },
    {
      "entropy": 6.006430768966675,
      "epoch": 2.6872323691339686,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042783882559376067,
      "loss": 5.61,
      "mean_token_accuracy": 0.17283107042312623,
      "num_tokens": 62318552.0,
      "step": 30750
    },
    {
      "entropy": 5.996453762054443,
      "epoch": 2.687669317486673,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042781595575220086,
      "loss": 5.5257,
      "mean_token_accuracy": 0.18637489825487136,
      "num_tokens": 62328430.0,
      "step": 30755
    },
    {
      "entropy": 5.937153339385986,
      "epoch": 2.688106265839378,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042779308297955084,
      "loss": 5.6415,
      "mean_token_accuracy": 0.17139867544174195,
      "num_tokens": 62338540.0,
      "step": 30760
    },
    {
      "entropy": 5.859754419326782,
      "epoch": 2.6885432141920824,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042777020727624945,
      "loss": 5.5806,
      "mean_token_accuracy": 0.17943557798862458,
      "num_tokens": 62348161.0,
      "step": 30765
    },
    {
      "entropy": 6.048112249374389,
      "epoch": 2.6889801625447873,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004277473286427354,
      "loss": 5.6696,
      "mean_token_accuracy": 0.16735382676124572,
      "num_tokens": 62358217.0,
      "step": 30770
    },
    {
      "entropy": 5.943034076690674,
      "epoch": 2.689417110897492,
      "grad_norm": 1.40625,
      "learning_rate": 0.00042772444707944767,
      "loss": 5.5898,
      "mean_token_accuracy": 0.1679429978132248,
      "num_tokens": 62368801.0,
      "step": 30775
    },
    {
      "entropy": 5.91063346862793,
      "epoch": 2.6898540592501967,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004277015625868249,
      "loss": 5.551,
      "mean_token_accuracy": 0.17689949721097947,
      "num_tokens": 62377632.0,
      "step": 30780
    },
    {
      "entropy": 5.979882287979126,
      "epoch": 2.690291007602901,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004276786751653062,
      "loss": 5.6173,
      "mean_token_accuracy": 0.17275638729333878,
      "num_tokens": 62388056.0,
      "step": 30785
    },
    {
      "entropy": 5.954830980300903,
      "epoch": 2.690727955955606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004276557848153305,
      "loss": 5.5692,
      "mean_token_accuracy": 0.17906465381383896,
      "num_tokens": 62398015.0,
      "step": 30790
    },
    {
      "entropy": 5.958869934082031,
      "epoch": 2.6911649043083106,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042763289153733685,
      "loss": 5.6367,
      "mean_token_accuracy": 0.1723467692732811,
      "num_tokens": 62408174.0,
      "step": 30795
    },
    {
      "entropy": 5.962563371658325,
      "epoch": 2.6916018526610155,
      "grad_norm": 1.46875,
      "learning_rate": 0.00042760999533176435,
      "loss": 5.573,
      "mean_token_accuracy": 0.1742850050330162,
      "num_tokens": 62417571.0,
      "step": 30800
    },
    {
      "entropy": 5.92403244972229,
      "epoch": 2.6920388010137204,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004275870961990523,
      "loss": 5.5903,
      "mean_token_accuracy": 0.18049822598695756,
      "num_tokens": 62427734.0,
      "step": 30805
    },
    {
      "entropy": 5.988879156112671,
      "epoch": 2.692475749366425,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042756419413963965,
      "loss": 5.5805,
      "mean_token_accuracy": 0.17520487904548646,
      "num_tokens": 62437109.0,
      "step": 30810
    },
    {
      "entropy": 5.881449413299561,
      "epoch": 2.6929126977191293,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042754128915396583,
      "loss": 5.5354,
      "mean_token_accuracy": 0.17733808755874633,
      "num_tokens": 62448163.0,
      "step": 30815
    },
    {
      "entropy": 5.954152154922485,
      "epoch": 2.6933496460718342,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042751838124247015,
      "loss": 5.5757,
      "mean_token_accuracy": 0.17791568040847777,
      "num_tokens": 62457993.0,
      "step": 30820
    },
    {
      "entropy": 6.00210337638855,
      "epoch": 2.693786594424539,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042749547040559204,
      "loss": 5.6644,
      "mean_token_accuracy": 0.17390268594026564,
      "num_tokens": 62468554.0,
      "step": 30825
    },
    {
      "entropy": 5.931888103485107,
      "epoch": 2.6942235427772436,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004274725566437709,
      "loss": 5.5424,
      "mean_token_accuracy": 0.1791839838027954,
      "num_tokens": 62479204.0,
      "step": 30830
    },
    {
      "entropy": 5.938584899902343,
      "epoch": 2.6946604911299485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004274496399574463,
      "loss": 5.5781,
      "mean_token_accuracy": 0.1762434720993042,
      "num_tokens": 62491391.0,
      "step": 30835
    },
    {
      "entropy": 5.91251802444458,
      "epoch": 2.695097439482653,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004274267203470576,
      "loss": 5.5029,
      "mean_token_accuracy": 0.1847705438733101,
      "num_tokens": 62500796.0,
      "step": 30840
    },
    {
      "entropy": 5.916656112670898,
      "epoch": 2.695534387835358,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042740379781304455,
      "loss": 5.5579,
      "mean_token_accuracy": 0.1820398285984993,
      "num_tokens": 62509996.0,
      "step": 30845
    },
    {
      "entropy": 5.962804126739502,
      "epoch": 2.6959713361880624,
      "grad_norm": 1.625,
      "learning_rate": 0.00042738087235584684,
      "loss": 5.5439,
      "mean_token_accuracy": 0.18224650472402573,
      "num_tokens": 62519676.0,
      "step": 30850
    },
    {
      "entropy": 5.977188014984131,
      "epoch": 2.6964082845407673,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042735794397590407,
      "loss": 5.5845,
      "mean_token_accuracy": 0.17544095367193221,
      "num_tokens": 62530775.0,
      "step": 30855
    },
    {
      "entropy": 5.879706811904907,
      "epoch": 2.696845232893472,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004273350126736562,
      "loss": 5.5601,
      "mean_token_accuracy": 0.17527805864810944,
      "num_tokens": 62540643.0,
      "step": 30860
    },
    {
      "entropy": 5.959574127197266,
      "epoch": 2.6972821812461767,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004273120784495429,
      "loss": 5.5178,
      "mean_token_accuracy": 0.1866527408361435,
      "num_tokens": 62549766.0,
      "step": 30865
    },
    {
      "entropy": 5.914683055877686,
      "epoch": 2.697719129598881,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004272891413040042,
      "loss": 5.5019,
      "mean_token_accuracy": 0.18329695165157317,
      "num_tokens": 62560130.0,
      "step": 30870
    },
    {
      "entropy": 5.907237529754639,
      "epoch": 2.698156077951586,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004272662012374799,
      "loss": 5.6037,
      "mean_token_accuracy": 0.17755764722824097,
      "num_tokens": 62571256.0,
      "step": 30875
    },
    {
      "entropy": 5.924151277542114,
      "epoch": 2.698593026304291,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004272432582504101,
      "loss": 5.4975,
      "mean_token_accuracy": 0.1882933109998703,
      "num_tokens": 62582269.0,
      "step": 30880
    },
    {
      "entropy": 5.919210815429688,
      "epoch": 2.6990299746569955,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042722031234323485,
      "loss": 5.5448,
      "mean_token_accuracy": 0.18133321553468704,
      "num_tokens": 62591875.0,
      "step": 30885
    },
    {
      "entropy": 5.965442371368408,
      "epoch": 2.6994669230097004,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042719736351639416,
      "loss": 5.6446,
      "mean_token_accuracy": 0.16923145353794097,
      "num_tokens": 62602467.0,
      "step": 30890
    },
    {
      "entropy": 5.880702209472656,
      "epoch": 2.699903871362405,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004271744117703284,
      "loss": 5.4964,
      "mean_token_accuracy": 0.18012176901102067,
      "num_tokens": 62611191.0,
      "step": 30895
    },
    {
      "entropy": 5.913506078720093,
      "epoch": 2.7003408197151098,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00042715145710547767,
      "loss": 5.5965,
      "mean_token_accuracy": 0.18616634458303452,
      "num_tokens": 62620961.0,
      "step": 30900
    },
    {
      "entropy": 5.997739934921265,
      "epoch": 2.700777768067814,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004271284995222822,
      "loss": 5.5691,
      "mean_token_accuracy": 0.179379378259182,
      "num_tokens": 62631972.0,
      "step": 30905
    },
    {
      "entropy": 5.961874103546142,
      "epoch": 2.701214716420519,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004271055390211824,
      "loss": 5.5278,
      "mean_token_accuracy": 0.18730137944221498,
      "num_tokens": 62641073.0,
      "step": 30910
    },
    {
      "entropy": 5.899742269515992,
      "epoch": 2.701651664773224,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004270825756026187,
      "loss": 5.5251,
      "mean_token_accuracy": 0.17596804797649385,
      "num_tokens": 62650364.0,
      "step": 30915
    },
    {
      "entropy": 5.947104501724243,
      "epoch": 2.7020886131259285,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004270596092670315,
      "loss": 5.6029,
      "mean_token_accuracy": 0.1798390880227089,
      "num_tokens": 62659846.0,
      "step": 30920
    },
    {
      "entropy": 5.976685237884522,
      "epoch": 2.702525561478633,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004270366400148613,
      "loss": 5.5983,
      "mean_token_accuracy": 0.17809589803218842,
      "num_tokens": 62670466.0,
      "step": 30925
    },
    {
      "entropy": 5.875787401199341,
      "epoch": 2.702962509831338,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004270136678465487,
      "loss": 5.5088,
      "mean_token_accuracy": 0.18112876117229462,
      "num_tokens": 62680330.0,
      "step": 30930
    },
    {
      "entropy": 5.9967944622039795,
      "epoch": 2.703399458184043,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042699069276253427,
      "loss": 5.6236,
      "mean_token_accuracy": 0.17020920366048814,
      "num_tokens": 62690177.0,
      "step": 30935
    },
    {
      "entropy": 5.957319068908691,
      "epoch": 2.7038364065367473,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004269677147632587,
      "loss": 5.5556,
      "mean_token_accuracy": 0.17780180275440216,
      "num_tokens": 62699365.0,
      "step": 30940
    },
    {
      "entropy": 5.810771894454956,
      "epoch": 2.704273354889452,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004269447338491628,
      "loss": 5.515,
      "mean_token_accuracy": 0.1900184705853462,
      "num_tokens": 62709825.0,
      "step": 30945
    },
    {
      "entropy": 5.989329433441162,
      "epoch": 2.7047103032421567,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004269217500206873,
      "loss": 5.6139,
      "mean_token_accuracy": 0.17935628592967987,
      "num_tokens": 62719848.0,
      "step": 30950
    },
    {
      "entropy": 5.994155645370483,
      "epoch": 2.7051472515948616,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042689876327827293,
      "loss": 5.6766,
      "mean_token_accuracy": 0.16820597797632217,
      "num_tokens": 62730405.0,
      "step": 30955
    },
    {
      "entropy": 5.961483764648437,
      "epoch": 2.705584199947566,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042687577362236073,
      "loss": 5.6009,
      "mean_token_accuracy": 0.17900968641042708,
      "num_tokens": 62740275.0,
      "step": 30960
    },
    {
      "entropy": 6.000072479248047,
      "epoch": 2.706021148300271,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004268527810533917,
      "loss": 5.5783,
      "mean_token_accuracy": 0.18403709530830384,
      "num_tokens": 62750847.0,
      "step": 30965
    },
    {
      "entropy": 5.945344638824463,
      "epoch": 2.706458096652976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042682978557180665,
      "loss": 5.6272,
      "mean_token_accuracy": 0.17266781032085418,
      "num_tokens": 62761688.0,
      "step": 30970
    },
    {
      "entropy": 5.990989732742309,
      "epoch": 2.7068950450056803,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042680678717804676,
      "loss": 5.6439,
      "mean_token_accuracy": 0.17649539709091186,
      "num_tokens": 62771168.0,
      "step": 30975
    },
    {
      "entropy": 5.91522479057312,
      "epoch": 2.707331993358385,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042678378587255315,
      "loss": 5.4977,
      "mean_token_accuracy": 0.18766643553972245,
      "num_tokens": 62780412.0,
      "step": 30980
    },
    {
      "entropy": 5.941982412338257,
      "epoch": 2.7077689417110897,
      "grad_norm": 1.453125,
      "learning_rate": 0.00042676078165576706,
      "loss": 5.6092,
      "mean_token_accuracy": 0.1801733195781708,
      "num_tokens": 62791673.0,
      "step": 30985
    },
    {
      "entropy": 5.984760379791259,
      "epoch": 2.7082058900637946,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004267377745281296,
      "loss": 5.5782,
      "mean_token_accuracy": 0.17939413487911224,
      "num_tokens": 62800980.0,
      "step": 30990
    },
    {
      "entropy": 5.94252381324768,
      "epoch": 2.708642838416499,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00042671476449008205,
      "loss": 5.5903,
      "mean_token_accuracy": 0.1739756390452385,
      "num_tokens": 62813183.0,
      "step": 30995
    },
    {
      "entropy": 5.964067077636718,
      "epoch": 2.709079786769204,
      "grad_norm": 1.09375,
      "learning_rate": 0.000426691751542066,
      "loss": 5.5855,
      "mean_token_accuracy": 0.17433235198259353,
      "num_tokens": 62823170.0,
      "step": 31000
    },
    {
      "entropy": 5.94468126296997,
      "epoch": 2.7095167351219085,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004266687356845225,
      "loss": 5.5781,
      "mean_token_accuracy": 0.17548059076070785,
      "num_tokens": 62833024.0,
      "step": 31005
    },
    {
      "entropy": 5.907732725143433,
      "epoch": 2.7099536834746134,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042664571691789323,
      "loss": 5.5816,
      "mean_token_accuracy": 0.17778460681438446,
      "num_tokens": 62842533.0,
      "step": 31010
    },
    {
      "entropy": 5.955104970932007,
      "epoch": 2.710390631827318,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004266226952426197,
      "loss": 5.6083,
      "mean_token_accuracy": 0.17516210824251174,
      "num_tokens": 62852984.0,
      "step": 31015
    },
    {
      "entropy": 5.955763053894043,
      "epoch": 2.710827580180023,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004265996706591434,
      "loss": 5.5986,
      "mean_token_accuracy": 0.17240016162395477,
      "num_tokens": 62861926.0,
      "step": 31020
    },
    {
      "entropy": 6.044569635391236,
      "epoch": 2.7112645285327277,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042657664316790594,
      "loss": 5.4949,
      "mean_token_accuracy": 0.1856416195631027,
      "num_tokens": 62872194.0,
      "step": 31025
    },
    {
      "entropy": 5.914891862869263,
      "epoch": 2.711701476885432,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000426553612769349,
      "loss": 5.5138,
      "mean_token_accuracy": 0.18000964671373368,
      "num_tokens": 62882887.0,
      "step": 31030
    },
    {
      "entropy": 5.8494795799255375,
      "epoch": 2.7121384252381366,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042653057946391446,
      "loss": 5.5676,
      "mean_token_accuracy": 0.17678734958171843,
      "num_tokens": 62893563.0,
      "step": 31035
    },
    {
      "entropy": 5.990234375,
      "epoch": 2.7125753735908416,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042650754325204395,
      "loss": 5.5544,
      "mean_token_accuracy": 0.18074440956115723,
      "num_tokens": 62902691.0,
      "step": 31040
    },
    {
      "entropy": 5.950732612609864,
      "epoch": 2.7130123219435465,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042648450413417937,
      "loss": 5.5156,
      "mean_token_accuracy": 0.1855270966887474,
      "num_tokens": 62912261.0,
      "step": 31045
    },
    {
      "entropy": 5.993121337890625,
      "epoch": 2.713449270296251,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042646146211076263,
      "loss": 5.6572,
      "mean_token_accuracy": 0.171201853454113,
      "num_tokens": 62922438.0,
      "step": 31050
    },
    {
      "entropy": 5.922832918167114,
      "epoch": 2.713886218648956,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004264384171822357,
      "loss": 5.5326,
      "mean_token_accuracy": 0.1825630098581314,
      "num_tokens": 62932378.0,
      "step": 31055
    },
    {
      "entropy": 5.968349504470825,
      "epoch": 2.7143231670016603,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00042641536934904064,
      "loss": 5.6952,
      "mean_token_accuracy": 0.1674965113401413,
      "num_tokens": 62943590.0,
      "step": 31060
    },
    {
      "entropy": 5.953985357284546,
      "epoch": 2.7147601153543652,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004263923186116193,
      "loss": 5.5556,
      "mean_token_accuracy": 0.17993401736021042,
      "num_tokens": 62953393.0,
      "step": 31065
    },
    {
      "entropy": 5.9873944282531735,
      "epoch": 2.7151970637070697,
      "grad_norm": 1.515625,
      "learning_rate": 0.00042636926497041406,
      "loss": 5.5913,
      "mean_token_accuracy": 0.17495439052581788,
      "num_tokens": 62964407.0,
      "step": 31070
    },
    {
      "entropy": 5.959587430953979,
      "epoch": 2.7156340120597746,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000426346208425867,
      "loss": 5.6191,
      "mean_token_accuracy": 0.17030072659254075,
      "num_tokens": 62974968.0,
      "step": 31075
    },
    {
      "entropy": 5.982704019546508,
      "epoch": 2.716070960412479,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004263231489784203,
      "loss": 5.5202,
      "mean_token_accuracy": 0.18033975213766099,
      "num_tokens": 62984978.0,
      "step": 31080
    },
    {
      "entropy": 5.927773284912109,
      "epoch": 2.716507908765184,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004263000866285164,
      "loss": 5.6338,
      "mean_token_accuracy": 0.17328521013259887,
      "num_tokens": 62995139.0,
      "step": 31085
    },
    {
      "entropy": 5.901360654830933,
      "epoch": 2.7169448571178885,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042627702137659746,
      "loss": 5.6238,
      "mean_token_accuracy": 0.1750980868935585,
      "num_tokens": 63006127.0,
      "step": 31090
    },
    {
      "entropy": 6.067293739318847,
      "epoch": 2.7173818054705934,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042625395322310605,
      "loss": 5.6266,
      "mean_token_accuracy": 0.17489292472600937,
      "num_tokens": 63015844.0,
      "step": 31095
    },
    {
      "entropy": 6.027743482589722,
      "epoch": 2.7178187538232983,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042623088216848446,
      "loss": 5.5839,
      "mean_token_accuracy": 0.1828131765127182,
      "num_tokens": 63025815.0,
      "step": 31100
    },
    {
      "entropy": 5.906958675384521,
      "epoch": 2.7182557021760028,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00042620780821317543,
      "loss": 5.641,
      "mean_token_accuracy": 0.17412914484739303,
      "num_tokens": 63036108.0,
      "step": 31105
    },
    {
      "entropy": 5.987539625167846,
      "epoch": 2.7186926505287072,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004261847313576214,
      "loss": 5.5301,
      "mean_token_accuracy": 0.1796095296740532,
      "num_tokens": 63046305.0,
      "step": 31110
    },
    {
      "entropy": 6.020885372161866,
      "epoch": 2.719129598881412,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000426161651602265,
      "loss": 5.6936,
      "mean_token_accuracy": 0.16404161900281905,
      "num_tokens": 63057115.0,
      "step": 31115
    },
    {
      "entropy": 5.977636051177979,
      "epoch": 2.719566547234117,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042613856894754886,
      "loss": 5.5695,
      "mean_token_accuracy": 0.1828301280736923,
      "num_tokens": 63067730.0,
      "step": 31120
    },
    {
      "entropy": 5.937607145309448,
      "epoch": 2.7200034955868215,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042611548339391574,
      "loss": 5.5569,
      "mean_token_accuracy": 0.17984631955623626,
      "num_tokens": 63077510.0,
      "step": 31125
    },
    {
      "entropy": 5.881743860244751,
      "epoch": 2.7204404439395264,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042609239494180857,
      "loss": 5.4968,
      "mean_token_accuracy": 0.18282256722450257,
      "num_tokens": 63087160.0,
      "step": 31130
    },
    {
      "entropy": 5.990683650970459,
      "epoch": 2.720877392292231,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004260693035916701,
      "loss": 5.6322,
      "mean_token_accuracy": 0.1744640961289406,
      "num_tokens": 63097284.0,
      "step": 31135
    },
    {
      "entropy": 5.976884937286377,
      "epoch": 2.721314340644936,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042604620934394313,
      "loss": 5.5842,
      "mean_token_accuracy": 0.17199771851301193,
      "num_tokens": 63106910.0,
      "step": 31140
    },
    {
      "entropy": 5.913565826416016,
      "epoch": 2.7217512889976403,
      "grad_norm": 1.453125,
      "learning_rate": 0.00042602311219907076,
      "loss": 5.4983,
      "mean_token_accuracy": 0.1777167335152626,
      "num_tokens": 63116129.0,
      "step": 31145
    },
    {
      "entropy": 5.914252519607544,
      "epoch": 2.722188237350345,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042600001215749593,
      "loss": 5.5804,
      "mean_token_accuracy": 0.17411267906427383,
      "num_tokens": 63125602.0,
      "step": 31150
    },
    {
      "entropy": 5.963639545440674,
      "epoch": 2.72262518570305,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004259769092196618,
      "loss": 5.5669,
      "mean_token_accuracy": 0.17491503357887267,
      "num_tokens": 63136592.0,
      "step": 31155
    },
    {
      "entropy": 6.011972808837891,
      "epoch": 2.7230621340557546,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042595380338601144,
      "loss": 5.5579,
      "mean_token_accuracy": 0.18191962391138078,
      "num_tokens": 63147040.0,
      "step": 31160
    },
    {
      "entropy": 5.893363046646118,
      "epoch": 2.723499082408459,
      "grad_norm": 1.34375,
      "learning_rate": 0.000425930694656988,
      "loss": 5.5621,
      "mean_token_accuracy": 0.17154232114553453,
      "num_tokens": 63156766.0,
      "step": 31165
    },
    {
      "entropy": 5.927434062957763,
      "epoch": 2.723936030761164,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004259075830330348,
      "loss": 5.5765,
      "mean_token_accuracy": 0.176756589114666,
      "num_tokens": 63166253.0,
      "step": 31170
    },
    {
      "entropy": 6.025254154205323,
      "epoch": 2.724372979113869,
      "grad_norm": 2.578125,
      "learning_rate": 0.000425884468514595,
      "loss": 5.6458,
      "mean_token_accuracy": 0.1696021780371666,
      "num_tokens": 63177352.0,
      "step": 31175
    },
    {
      "entropy": 6.035243368148803,
      "epoch": 2.7248099274665734,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004258613511021121,
      "loss": 5.6525,
      "mean_token_accuracy": 0.17032257616519927,
      "num_tokens": 63188092.0,
      "step": 31180
    },
    {
      "entropy": 5.943561983108521,
      "epoch": 2.7252468758192783,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004258382307960294,
      "loss": 5.4935,
      "mean_token_accuracy": 0.1865013837814331,
      "num_tokens": 63197677.0,
      "step": 31185
    },
    {
      "entropy": 5.98064079284668,
      "epoch": 2.7256838241719827,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004258151075967904,
      "loss": 5.6776,
      "mean_token_accuracy": 0.1703468978404999,
      "num_tokens": 63208536.0,
      "step": 31190
    },
    {
      "entropy": 5.934188556671143,
      "epoch": 2.7261207725246877,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004257919815048387,
      "loss": 5.5411,
      "mean_token_accuracy": 0.18396392464637756,
      "num_tokens": 63217966.0,
      "step": 31195
    },
    {
      "entropy": 5.9571404457092285,
      "epoch": 2.726557720877392,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004257688525206176,
      "loss": 5.5414,
      "mean_token_accuracy": 0.17578188180923462,
      "num_tokens": 63227420.0,
      "step": 31200
    },
    {
      "entropy": 5.992783451080323,
      "epoch": 2.726994669230097,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004257457206445711,
      "loss": 5.6313,
      "mean_token_accuracy": 0.1729407325387001,
      "num_tokens": 63237887.0,
      "step": 31205
    },
    {
      "entropy": 5.886266708374023,
      "epoch": 2.727431617582802,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042572258587714255,
      "loss": 5.5387,
      "mean_token_accuracy": 0.18627834767103196,
      "num_tokens": 63247539.0,
      "step": 31210
    },
    {
      "entropy": 5.93590350151062,
      "epoch": 2.7278685659355064,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004256994482187759,
      "loss": 5.5004,
      "mean_token_accuracy": 0.18074309527873994,
      "num_tokens": 63257839.0,
      "step": 31215
    },
    {
      "entropy": 5.970898008346557,
      "epoch": 2.728305514288211,
      "grad_norm": 2.421875,
      "learning_rate": 0.00042567630766991477,
      "loss": 5.6349,
      "mean_token_accuracy": 0.17219175696372985,
      "num_tokens": 63268667.0,
      "step": 31220
    },
    {
      "entropy": 6.007015132904053,
      "epoch": 2.728742462640916,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004256531642310032,
      "loss": 5.6085,
      "mean_token_accuracy": 0.18093670010566712,
      "num_tokens": 63277413.0,
      "step": 31225
    },
    {
      "entropy": 6.012926912307739,
      "epoch": 2.7291794109936207,
      "grad_norm": 1.578125,
      "learning_rate": 0.000425630017902485,
      "loss": 5.6052,
      "mean_token_accuracy": 0.1817570224404335,
      "num_tokens": 63287225.0,
      "step": 31230
    },
    {
      "entropy": 5.930014705657959,
      "epoch": 2.729616359346325,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004256068686848041,
      "loss": 5.5572,
      "mean_token_accuracy": 0.1772441953420639,
      "num_tokens": 63297211.0,
      "step": 31235
    },
    {
      "entropy": 5.9080588817596436,
      "epoch": 2.73005330769903,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004255837165784045,
      "loss": 5.5326,
      "mean_token_accuracy": 0.18671616017818451,
      "num_tokens": 63307443.0,
      "step": 31240
    },
    {
      "entropy": 5.926528644561768,
      "epoch": 2.7304902560517346,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004255605615837304,
      "loss": 5.6289,
      "mean_token_accuracy": 0.17707698047161102,
      "num_tokens": 63317951.0,
      "step": 31245
    },
    {
      "entropy": 5.971393346786499,
      "epoch": 2.7309272044044395,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042553740370122574,
      "loss": 5.5449,
      "mean_token_accuracy": 0.18433847278356552,
      "num_tokens": 63328138.0,
      "step": 31250
    },
    {
      "entropy": 5.9469287395477295,
      "epoch": 2.731364152757144,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004255142429313348,
      "loss": 5.4597,
      "mean_token_accuracy": 0.18763009905815126,
      "num_tokens": 63338229.0,
      "step": 31255
    },
    {
      "entropy": 5.942113733291626,
      "epoch": 2.731801101109849,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004254910792745019,
      "loss": 5.6244,
      "mean_token_accuracy": 0.1770287424325943,
      "num_tokens": 63348454.0,
      "step": 31260
    },
    {
      "entropy": 5.927615356445313,
      "epoch": 2.7322380494625538,
      "grad_norm": 1.375,
      "learning_rate": 0.00042546791273117125,
      "loss": 5.4496,
      "mean_token_accuracy": 0.18945411294698716,
      "num_tokens": 63357346.0,
      "step": 31265
    },
    {
      "entropy": 5.946689224243164,
      "epoch": 2.7326749978152582,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004254447433017871,
      "loss": 5.6351,
      "mean_token_accuracy": 0.17325370237231255,
      "num_tokens": 63368329.0,
      "step": 31270
    },
    {
      "entropy": 5.998834180831909,
      "epoch": 2.7331119461679627,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042542157098679397,
      "loss": 5.6509,
      "mean_token_accuracy": 0.17127542942762375,
      "num_tokens": 63378815.0,
      "step": 31275
    },
    {
      "entropy": 5.956162309646606,
      "epoch": 2.7335488945206676,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004253983957866364,
      "loss": 5.5545,
      "mean_token_accuracy": 0.18172529339790344,
      "num_tokens": 63390099.0,
      "step": 31280
    },
    {
      "entropy": 5.964241409301758,
      "epoch": 2.7339858428733725,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042537521770175856,
      "loss": 5.6274,
      "mean_token_accuracy": 0.17063786089420319,
      "num_tokens": 63401244.0,
      "step": 31285
    },
    {
      "entropy": 5.916257667541504,
      "epoch": 2.734422791226077,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042535203673260544,
      "loss": 5.5585,
      "mean_token_accuracy": 0.18170331716537474,
      "num_tokens": 63410750.0,
      "step": 31290
    },
    {
      "entropy": 5.89330358505249,
      "epoch": 2.734859739578782,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042532885287962146,
      "loss": 5.5385,
      "mean_token_accuracy": 0.1804671436548233,
      "num_tokens": 63420936.0,
      "step": 31295
    },
    {
      "entropy": 5.926695442199707,
      "epoch": 2.7352966879314864,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042530566614325125,
      "loss": 5.5988,
      "mean_token_accuracy": 0.17416592687368393,
      "num_tokens": 63431614.0,
      "step": 31300
    },
    {
      "entropy": 5.918217611312866,
      "epoch": 2.7357336362841913,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042528247652393964,
      "loss": 5.5355,
      "mean_token_accuracy": 0.17843178063631057,
      "num_tokens": 63441298.0,
      "step": 31305
    },
    {
      "entropy": 5.938329744338989,
      "epoch": 2.7361705846368958,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042525928402213127,
      "loss": 5.5489,
      "mean_token_accuracy": 0.17389279156923293,
      "num_tokens": 63451542.0,
      "step": 31310
    },
    {
      "entropy": 5.981653642654419,
      "epoch": 2.7366075329896007,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004252360886382712,
      "loss": 5.5991,
      "mean_token_accuracy": 0.17363885492086412,
      "num_tokens": 63460863.0,
      "step": 31315
    },
    {
      "entropy": 5.919799089431763,
      "epoch": 2.7370444813423056,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042521289037280425,
      "loss": 5.6077,
      "mean_token_accuracy": 0.1709859699010849,
      "num_tokens": 63470499.0,
      "step": 31320
    },
    {
      "entropy": 6.016832780838013,
      "epoch": 2.73748142969501,
      "grad_norm": 1.390625,
      "learning_rate": 0.00042518968922617534,
      "loss": 5.6165,
      "mean_token_accuracy": 0.17744412869215012,
      "num_tokens": 63480803.0,
      "step": 31325
    },
    {
      "entropy": 5.968651151657104,
      "epoch": 2.7379183780477145,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004251664851988295,
      "loss": 5.5639,
      "mean_token_accuracy": 0.1789831832051277,
      "num_tokens": 63491544.0,
      "step": 31330
    },
    {
      "entropy": 5.88733811378479,
      "epoch": 2.7383553264004195,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042514327829121175,
      "loss": 5.4393,
      "mean_token_accuracy": 0.18386521935462952,
      "num_tokens": 63501086.0,
      "step": 31335
    },
    {
      "entropy": 5.9214826107025145,
      "epoch": 2.7387922747531244,
      "grad_norm": 2.34375,
      "learning_rate": 0.00042512006850376724,
      "loss": 5.5815,
      "mean_token_accuracy": 0.1803106486797333,
      "num_tokens": 63512408.0,
      "step": 31340
    },
    {
      "entropy": 6.003998184204102,
      "epoch": 2.739229223105829,
      "grad_norm": 1.125,
      "learning_rate": 0.00042509685583694116,
      "loss": 5.5566,
      "mean_token_accuracy": 0.17927024215459825,
      "num_tokens": 63521735.0,
      "step": 31345
    },
    {
      "entropy": 5.940384149551392,
      "epoch": 2.7396661714585337,
      "grad_norm": 1.25,
      "learning_rate": 0.0004250736402911788,
      "loss": 5.5192,
      "mean_token_accuracy": 0.18681200444698334,
      "num_tokens": 63532045.0,
      "step": 31350
    },
    {
      "entropy": 5.858130359649659,
      "epoch": 2.740103119811238,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004250504218669253,
      "loss": 5.5213,
      "mean_token_accuracy": 0.18325980007648468,
      "num_tokens": 63541286.0,
      "step": 31355
    },
    {
      "entropy": 5.892609643936157,
      "epoch": 2.740540068163943,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004250272005646261,
      "loss": 5.5085,
      "mean_token_accuracy": 0.1853812262415886,
      "num_tokens": 63552195.0,
      "step": 31360
    },
    {
      "entropy": 5.950520992279053,
      "epoch": 2.7409770165166476,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004250039763847266,
      "loss": 5.5125,
      "mean_token_accuracy": 0.1843504086136818,
      "num_tokens": 63561318.0,
      "step": 31365
    },
    {
      "entropy": 5.881163167953491,
      "epoch": 2.7414139648693525,
      "grad_norm": 1.421875,
      "learning_rate": 0.00042498074932767227,
      "loss": 5.5688,
      "mean_token_accuracy": 0.18151180148124696,
      "num_tokens": 63571637.0,
      "step": 31370
    },
    {
      "entropy": 5.959279775619507,
      "epoch": 2.7418509132220574,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042495751939390853,
      "loss": 5.6251,
      "mean_token_accuracy": 0.1721947282552719,
      "num_tokens": 63581618.0,
      "step": 31375
    },
    {
      "entropy": 5.881095933914184,
      "epoch": 2.742287861574762,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042493428658388094,
      "loss": 5.4498,
      "mean_token_accuracy": 0.19464221298694612,
      "num_tokens": 63592129.0,
      "step": 31380
    },
    {
      "entropy": 5.967544460296631,
      "epoch": 2.7427248099274664,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004249110508980353,
      "loss": 5.6476,
      "mean_token_accuracy": 0.17777210772037505,
      "num_tokens": 63602861.0,
      "step": 31385
    },
    {
      "entropy": 5.976181125640869,
      "epoch": 2.7431617582801713,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042488781233681705,
      "loss": 5.5532,
      "mean_token_accuracy": 0.1788635104894638,
      "num_tokens": 63612423.0,
      "step": 31390
    },
    {
      "entropy": 5.909365892410278,
      "epoch": 2.743598706632876,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042486457090067206,
      "loss": 5.5627,
      "mean_token_accuracy": 0.18025910407304763,
      "num_tokens": 63622351.0,
      "step": 31395
    },
    {
      "entropy": 5.9420722961425785,
      "epoch": 2.7440356549855807,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00042484132659004604,
      "loss": 5.5124,
      "mean_token_accuracy": 0.1871438130736351,
      "num_tokens": 63632024.0,
      "step": 31400
    },
    {
      "entropy": 5.919685745239258,
      "epoch": 2.7444726033382856,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004248180794053849,
      "loss": 5.5228,
      "mean_token_accuracy": 0.18235511630773543,
      "num_tokens": 63641962.0,
      "step": 31405
    },
    {
      "entropy": 5.851445579528809,
      "epoch": 2.74490955169099,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042479482934713446,
      "loss": 5.5747,
      "mean_token_accuracy": 0.1782868355512619,
      "num_tokens": 63652095.0,
      "step": 31410
    },
    {
      "entropy": 6.035810947418213,
      "epoch": 2.745346500043695,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004247715764157407,
      "loss": 5.6484,
      "mean_token_accuracy": 0.17433976978063584,
      "num_tokens": 63661392.0,
      "step": 31415
    },
    {
      "entropy": 5.961418294906617,
      "epoch": 2.7457834483963994,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042474832061164966,
      "loss": 5.5585,
      "mean_token_accuracy": 0.17526283413171767,
      "num_tokens": 63672383.0,
      "step": 31420
    },
    {
      "entropy": 5.928472328186035,
      "epoch": 2.7462203967491043,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042472506193530734,
      "loss": 5.582,
      "mean_token_accuracy": 0.17487904727458953,
      "num_tokens": 63682522.0,
      "step": 31425
    },
    {
      "entropy": 5.9473024845123295,
      "epoch": 2.746657345101809,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004247018003871599,
      "loss": 5.5231,
      "mean_token_accuracy": 0.18407471626996993,
      "num_tokens": 63692443.0,
      "step": 31430
    },
    {
      "entropy": 5.878679990768433,
      "epoch": 2.7470942934545137,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042467853596765345,
      "loss": 5.5167,
      "mean_token_accuracy": 0.18411050289869307,
      "num_tokens": 63702268.0,
      "step": 31435
    },
    {
      "entropy": 5.8189427852630615,
      "epoch": 2.747531241807218,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042465526867723423,
      "loss": 5.4407,
      "mean_token_accuracy": 0.19282397776842117,
      "num_tokens": 63712621.0,
      "step": 31440
    },
    {
      "entropy": 6.001898908615113,
      "epoch": 2.747968190159923,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004246319985163487,
      "loss": 5.6081,
      "mean_token_accuracy": 0.17649172991514206,
      "num_tokens": 63722319.0,
      "step": 31445
    },
    {
      "entropy": 5.996595573425293,
      "epoch": 2.748405138512628,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004246087254854428,
      "loss": 5.6148,
      "mean_token_accuracy": 0.1794084995985031,
      "num_tokens": 63732726.0,
      "step": 31450
    },
    {
      "entropy": 5.8905925273895265,
      "epoch": 2.7488420868653325,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042458544958496333,
      "loss": 5.5077,
      "mean_token_accuracy": 0.1880096361041069,
      "num_tokens": 63741197.0,
      "step": 31455
    },
    {
      "entropy": 5.908514165878296,
      "epoch": 2.749279035218037,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042456217081535645,
      "loss": 5.5285,
      "mean_token_accuracy": 0.18377624899148942,
      "num_tokens": 63751316.0,
      "step": 31460
    },
    {
      "entropy": 5.98678503036499,
      "epoch": 2.749715983570742,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042453888917706885,
      "loss": 5.6128,
      "mean_token_accuracy": 0.17524250745773315,
      "num_tokens": 63761246.0,
      "step": 31465
    },
    {
      "entropy": 5.964221239089966,
      "epoch": 2.750152931923447,
      "grad_norm": 1.40625,
      "learning_rate": 0.00042451560467054696,
      "loss": 5.6133,
      "mean_token_accuracy": 0.17376265078783035,
      "num_tokens": 63771442.0,
      "step": 31470
    },
    {
      "entropy": 5.978780364990234,
      "epoch": 2.7505898802761513,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042449231729623735,
      "loss": 5.648,
      "mean_token_accuracy": 0.1775224268436432,
      "num_tokens": 63782840.0,
      "step": 31475
    },
    {
      "entropy": 5.926541805267334,
      "epoch": 2.751026828628856,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042446902705458683,
      "loss": 5.5281,
      "mean_token_accuracy": 0.17843584716320038,
      "num_tokens": 63793089.0,
      "step": 31480
    },
    {
      "entropy": 5.935405254364014,
      "epoch": 2.7514637769815606,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004244457339460421,
      "loss": 5.6046,
      "mean_token_accuracy": 0.16768838614225387,
      "num_tokens": 63804125.0,
      "step": 31485
    },
    {
      "entropy": 5.9167070388793945,
      "epoch": 2.7519007253342656,
      "grad_norm": 1.25,
      "learning_rate": 0.00042442243797104976,
      "loss": 5.5887,
      "mean_token_accuracy": 0.18113145232200623,
      "num_tokens": 63814713.0,
      "step": 31490
    },
    {
      "entropy": 6.025100231170654,
      "epoch": 2.75233767368697,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004243991391300569,
      "loss": 5.6217,
      "mean_token_accuracy": 0.1698010802268982,
      "num_tokens": 63824403.0,
      "step": 31495
    },
    {
      "entropy": 5.956995487213135,
      "epoch": 2.752774622039675,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004243758374235102,
      "loss": 5.6021,
      "mean_token_accuracy": 0.17759883254766465,
      "num_tokens": 63834182.0,
      "step": 31500
    },
    {
      "entropy": 5.921602582931518,
      "epoch": 2.75321157039238,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004243525328518567,
      "loss": 5.5909,
      "mean_token_accuracy": 0.17712990045547486,
      "num_tokens": 63844282.0,
      "step": 31505
    },
    {
      "entropy": 5.965380096435547,
      "epoch": 2.7536485187450843,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004243292254155433,
      "loss": 5.5436,
      "mean_token_accuracy": 0.18052661567926406,
      "num_tokens": 63854898.0,
      "step": 31510
    },
    {
      "entropy": 5.886848735809326,
      "epoch": 2.754085467097789,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042430591511501717,
      "loss": 5.4755,
      "mean_token_accuracy": 0.19033523947000502,
      "num_tokens": 63864501.0,
      "step": 31515
    },
    {
      "entropy": 5.878275394439697,
      "epoch": 2.7545224154504937,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042428260195072527,
      "loss": 5.5792,
      "mean_token_accuracy": 0.1808053195476532,
      "num_tokens": 63874812.0,
      "step": 31520
    },
    {
      "entropy": 6.009347438812256,
      "epoch": 2.7549593638031986,
      "grad_norm": 1.25,
      "learning_rate": 0.0004242592859231149,
      "loss": 5.5523,
      "mean_token_accuracy": 0.17812125533819198,
      "num_tokens": 63884028.0,
      "step": 31525
    },
    {
      "entropy": 5.931916999816894,
      "epoch": 2.755396312155903,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042423596703263324,
      "loss": 5.5238,
      "mean_token_accuracy": 0.18504435569047928,
      "num_tokens": 63893991.0,
      "step": 31530
    },
    {
      "entropy": 5.9078511714935305,
      "epoch": 2.755833260508608,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004242126452797275,
      "loss": 5.6152,
      "mean_token_accuracy": 0.1753977820277214,
      "num_tokens": 63904168.0,
      "step": 31535
    },
    {
      "entropy": 5.9674615383148195,
      "epoch": 2.7562702088613125,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042418932066484505,
      "loss": 5.5939,
      "mean_token_accuracy": 0.17299525290727616,
      "num_tokens": 63913881.0,
      "step": 31540
    },
    {
      "entropy": 5.984518909454346,
      "epoch": 2.7567071572140174,
      "grad_norm": 1.125,
      "learning_rate": 0.00042416599318843324,
      "loss": 5.6601,
      "mean_token_accuracy": 0.16942810416221618,
      "num_tokens": 63924521.0,
      "step": 31545
    },
    {
      "entropy": 6.0105016231536865,
      "epoch": 2.757144105566722,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004241426628509395,
      "loss": 5.6061,
      "mean_token_accuracy": 0.16774597018957138,
      "num_tokens": 63935046.0,
      "step": 31550
    },
    {
      "entropy": 5.925435733795166,
      "epoch": 2.7575810539194268,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042411932965281143,
      "loss": 5.5751,
      "mean_token_accuracy": 0.16988580524921418,
      "num_tokens": 63944048.0,
      "step": 31555
    },
    {
      "entropy": 5.960390424728393,
      "epoch": 2.7580180022721317,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004240959935944964,
      "loss": 5.542,
      "mean_token_accuracy": 0.18564472496509551,
      "num_tokens": 63954943.0,
      "step": 31560
    },
    {
      "entropy": 5.876499891281128,
      "epoch": 2.758454950624836,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004240726546764421,
      "loss": 5.5347,
      "mean_token_accuracy": 0.18252288401126862,
      "num_tokens": 63965929.0,
      "step": 31565
    },
    {
      "entropy": 5.933775091171265,
      "epoch": 2.7588918989775406,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042404931289909614,
      "loss": 5.5814,
      "mean_token_accuracy": 0.17156823426485063,
      "num_tokens": 63976353.0,
      "step": 31570
    },
    {
      "entropy": 6.035974502563477,
      "epoch": 2.7593288473302455,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042402596826290634,
      "loss": 5.5967,
      "mean_token_accuracy": 0.1787727043032646,
      "num_tokens": 63987073.0,
      "step": 31575
    },
    {
      "entropy": 5.911772727966309,
      "epoch": 2.7597657956829504,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004240026207683203,
      "loss": 5.5698,
      "mean_token_accuracy": 0.18916374444961548,
      "num_tokens": 63996631.0,
      "step": 31580
    },
    {
      "entropy": 5.924098062515259,
      "epoch": 2.760202744035655,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042397927041578596,
      "loss": 5.5929,
      "mean_token_accuracy": 0.17874462008476258,
      "num_tokens": 64007662.0,
      "step": 31585
    },
    {
      "entropy": 5.928346967697143,
      "epoch": 2.76063969238836,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042395591720575113,
      "loss": 5.5126,
      "mean_token_accuracy": 0.182171268761158,
      "num_tokens": 64017868.0,
      "step": 31590
    },
    {
      "entropy": 5.973548507690429,
      "epoch": 2.7610766407410643,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042393256113866377,
      "loss": 5.6068,
      "mean_token_accuracy": 0.16618584394454955,
      "num_tokens": 64027464.0,
      "step": 31595
    },
    {
      "entropy": 5.920731782913208,
      "epoch": 2.761513589093769,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004239092022149718,
      "loss": 5.5969,
      "mean_token_accuracy": 0.17416347116231917,
      "num_tokens": 64037795.0,
      "step": 31600
    },
    {
      "entropy": 6.032570791244507,
      "epoch": 2.7619505374464737,
      "grad_norm": 1.8125,
      "learning_rate": 0.00042388584043512333,
      "loss": 5.595,
      "mean_token_accuracy": 0.1740223228931427,
      "num_tokens": 64048055.0,
      "step": 31605
    },
    {
      "entropy": 5.911257266998291,
      "epoch": 2.7623874857991786,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004238624757995664,
      "loss": 5.47,
      "mean_token_accuracy": 0.18500899970531465,
      "num_tokens": 64057531.0,
      "step": 31610
    },
    {
      "entropy": 5.917443466186524,
      "epoch": 2.7628244341518835,
      "grad_norm": 1.125,
      "learning_rate": 0.0004238391083087493,
      "loss": 5.5234,
      "mean_token_accuracy": 0.17824902087450029,
      "num_tokens": 64066476.0,
      "step": 31615
    },
    {
      "entropy": 5.919487047195434,
      "epoch": 2.763261382504588,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042381573796312,
      "loss": 5.5288,
      "mean_token_accuracy": 0.18991990387439728,
      "num_tokens": 64077069.0,
      "step": 31620
    },
    {
      "entropy": 5.909550428390503,
      "epoch": 2.7636983308572924,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004237923647631269,
      "loss": 5.5414,
      "mean_token_accuracy": 0.18369584679603576,
      "num_tokens": 64087325.0,
      "step": 31625
    },
    {
      "entropy": 5.94566068649292,
      "epoch": 2.7641352792099974,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004237689887092183,
      "loss": 5.59,
      "mean_token_accuracy": 0.1764730766415596,
      "num_tokens": 64097040.0,
      "step": 31630
    },
    {
      "entropy": 5.9409264087677,
      "epoch": 2.7645722275627023,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042374560980184254,
      "loss": 5.5749,
      "mean_token_accuracy": 0.18315970301628112,
      "num_tokens": 64107798.0,
      "step": 31635
    },
    {
      "entropy": 6.016220140457153,
      "epoch": 2.7650091759154067,
      "grad_norm": 1.03125,
      "learning_rate": 0.000423722228041448,
      "loss": 5.6206,
      "mean_token_accuracy": 0.17301631718873978,
      "num_tokens": 64118395.0,
      "step": 31640
    },
    {
      "entropy": 5.967028522491455,
      "epoch": 2.7654461242681116,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042369884342848327,
      "loss": 5.6188,
      "mean_token_accuracy": 0.1725737124681473,
      "num_tokens": 64128040.0,
      "step": 31645
    },
    {
      "entropy": 5.898792886734009,
      "epoch": 2.765883072620816,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042367545596339674,
      "loss": 5.5524,
      "mean_token_accuracy": 0.18205137997865678,
      "num_tokens": 64137660.0,
      "step": 31650
    },
    {
      "entropy": 5.906246948242187,
      "epoch": 2.766320020973521,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042365206564663705,
      "loss": 5.5526,
      "mean_token_accuracy": 0.18114017695188522,
      "num_tokens": 64147632.0,
      "step": 31655
    },
    {
      "entropy": 5.9295135021209715,
      "epoch": 2.7667569693262255,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004236286724786529,
      "loss": 5.5559,
      "mean_token_accuracy": 0.17970579862594604,
      "num_tokens": 64157604.0,
      "step": 31660
    },
    {
      "entropy": 6.003112888336181,
      "epoch": 2.7671939176789304,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042360527645989287,
      "loss": 5.6304,
      "mean_token_accuracy": 0.1702803775668144,
      "num_tokens": 64167626.0,
      "step": 31665
    },
    {
      "entropy": 6.028597974777222,
      "epoch": 2.7676308660316353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004235818775908059,
      "loss": 5.6274,
      "mean_token_accuracy": 0.17649204432964324,
      "num_tokens": 64178436.0,
      "step": 31670
    },
    {
      "entropy": 5.840271472930908,
      "epoch": 2.76806781438434,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042355847587184056,
      "loss": 5.4859,
      "mean_token_accuracy": 0.19066119343042373,
      "num_tokens": 64188911.0,
      "step": 31675
    },
    {
      "entropy": 5.914220237731934,
      "epoch": 2.7685047627370443,
      "grad_norm": 1.390625,
      "learning_rate": 0.00042353507130344584,
      "loss": 5.4814,
      "mean_token_accuracy": 0.19015003442764283,
      "num_tokens": 64198539.0,
      "step": 31680
    },
    {
      "entropy": 5.971554756164551,
      "epoch": 2.768941711089749,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004235116638860705,
      "loss": 5.6074,
      "mean_token_accuracy": 0.1775437206029892,
      "num_tokens": 64209121.0,
      "step": 31685
    },
    {
      "entropy": 5.9412376403808596,
      "epoch": 2.769378659442454,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004234882536201638,
      "loss": 5.5522,
      "mean_token_accuracy": 0.17438496202230452,
      "num_tokens": 64218610.0,
      "step": 31690
    },
    {
      "entropy": 5.880580759048462,
      "epoch": 2.7698156077951586,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004234648405061745,
      "loss": 5.5732,
      "mean_token_accuracy": 0.177228382229805,
      "num_tokens": 64228626.0,
      "step": 31695
    },
    {
      "entropy": 5.955782413482666,
      "epoch": 2.7702525561478635,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042344142454455174,
      "loss": 5.5356,
      "mean_token_accuracy": 0.18314214646816254,
      "num_tokens": 64238575.0,
      "step": 31700
    },
    {
      "entropy": 5.959626960754394,
      "epoch": 2.770689504500568,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004234180057357448,
      "loss": 5.5594,
      "mean_token_accuracy": 0.17795795798301697,
      "num_tokens": 64248551.0,
      "step": 31705
    },
    {
      "entropy": 5.985945415496826,
      "epoch": 2.771126452853273,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042339458408020267,
      "loss": 5.5747,
      "mean_token_accuracy": 0.17724396884441376,
      "num_tokens": 64259315.0,
      "step": 31710
    },
    {
      "entropy": 5.956926345825195,
      "epoch": 2.7715634012059773,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004233711595783746,
      "loss": 5.5837,
      "mean_token_accuracy": 0.17814336568117142,
      "num_tokens": 64269742.0,
      "step": 31715
    },
    {
      "entropy": 5.947073268890381,
      "epoch": 2.7720003495586822,
      "grad_norm": 1.453125,
      "learning_rate": 0.00042334773223071,
      "loss": 5.5763,
      "mean_token_accuracy": 0.18246948421001435,
      "num_tokens": 64279266.0,
      "step": 31720
    },
    {
      "entropy": 5.9581708908081055,
      "epoch": 2.7724372979113867,
      "grad_norm": 1.125,
      "learning_rate": 0.00042332430203765816,
      "loss": 5.5289,
      "mean_token_accuracy": 0.1790711134672165,
      "num_tokens": 64289567.0,
      "step": 31725
    },
    {
      "entropy": 5.901415205001831,
      "epoch": 2.7728742462640916,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004233008689996685,
      "loss": 5.6313,
      "mean_token_accuracy": 0.1757206693291664,
      "num_tokens": 64299126.0,
      "step": 31730
    },
    {
      "entropy": 5.945808362960816,
      "epoch": 2.773311194616796,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004232774331171904,
      "loss": 5.4919,
      "mean_token_accuracy": 0.1879425659775734,
      "num_tokens": 64308964.0,
      "step": 31735
    },
    {
      "entropy": 5.991978883743286,
      "epoch": 2.773748142969501,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042325399439067354,
      "loss": 5.5869,
      "mean_token_accuracy": 0.17443047612905502,
      "num_tokens": 64318463.0,
      "step": 31740
    },
    {
      "entropy": 6.025221490859986,
      "epoch": 2.774185091322206,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004232305528205673,
      "loss": 5.7247,
      "mean_token_accuracy": 0.17016735821962356,
      "num_tokens": 64328841.0,
      "step": 31745
    },
    {
      "entropy": 5.949898195266724,
      "epoch": 2.7746220396749104,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004232071084073214,
      "loss": 5.5609,
      "mean_token_accuracy": 0.1821172922849655,
      "num_tokens": 64339037.0,
      "step": 31750
    },
    {
      "entropy": 5.976774120330811,
      "epoch": 2.775058988027615,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042318366115138553,
      "loss": 5.7031,
      "mean_token_accuracy": 0.17225360572338105,
      "num_tokens": 64350048.0,
      "step": 31755
    },
    {
      "entropy": 6.015827941894531,
      "epoch": 2.7754959363803198,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042316021105320937,
      "loss": 5.6674,
      "mean_token_accuracy": 0.1723084881901741,
      "num_tokens": 64361066.0,
      "step": 31760
    },
    {
      "entropy": 6.006498861312866,
      "epoch": 2.7759328847330247,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042313675811324276,
      "loss": 5.6188,
      "mean_token_accuracy": 0.17552109211683273,
      "num_tokens": 64371064.0,
      "step": 31765
    },
    {
      "entropy": 5.92082405090332,
      "epoch": 2.776369833085729,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042311330233193545,
      "loss": 5.5033,
      "mean_token_accuracy": 0.19033185690641402,
      "num_tokens": 64380820.0,
      "step": 31770
    },
    {
      "entropy": 5.908173704147339,
      "epoch": 2.776806781438434,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004230898437097374,
      "loss": 5.5183,
      "mean_token_accuracy": 0.18234141319990158,
      "num_tokens": 64389777.0,
      "step": 31775
    },
    {
      "entropy": 5.991281938552857,
      "epoch": 2.7772437297911385,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004230663822470986,
      "loss": 5.6728,
      "mean_token_accuracy": 0.17189572602510453,
      "num_tokens": 64399542.0,
      "step": 31780
    },
    {
      "entropy": 5.968345975875854,
      "epoch": 2.7776806781438435,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004230429179444689,
      "loss": 5.5585,
      "mean_token_accuracy": 0.17394347190856935,
      "num_tokens": 64410254.0,
      "step": 31785
    },
    {
      "entropy": 5.907100534439087,
      "epoch": 2.778117626496548,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004230194508022986,
      "loss": 5.5345,
      "mean_token_accuracy": 0.1744745686650276,
      "num_tokens": 64420487.0,
      "step": 31790
    },
    {
      "entropy": 5.942773628234863,
      "epoch": 2.778554574849253,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042299598082103757,
      "loss": 5.5593,
      "mean_token_accuracy": 0.18317256718873978,
      "num_tokens": 64430171.0,
      "step": 31795
    },
    {
      "entropy": 5.977210760116577,
      "epoch": 2.7789915232019577,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042297250800113614,
      "loss": 5.6363,
      "mean_token_accuracy": 0.17015163898468016,
      "num_tokens": 64439960.0,
      "step": 31800
    },
    {
      "entropy": 6.06213412284851,
      "epoch": 2.779428471554662,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004229490323430444,
      "loss": 5.5867,
      "mean_token_accuracy": 0.1765194594860077,
      "num_tokens": 64450238.0,
      "step": 31805
    },
    {
      "entropy": 5.914744424819946,
      "epoch": 2.7798654199073667,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042292555384721277,
      "loss": 5.5818,
      "mean_token_accuracy": 0.17595392018556594,
      "num_tokens": 64460137.0,
      "step": 31810
    },
    {
      "entropy": 5.941164159774781,
      "epoch": 2.7803023682600716,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004229020725140915,
      "loss": 5.5389,
      "mean_token_accuracy": 0.18392397761344909,
      "num_tokens": 64470138.0,
      "step": 31815
    },
    {
      "entropy": 5.98099422454834,
      "epoch": 2.7807393166127765,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004228785883441309,
      "loss": 5.6178,
      "mean_token_accuracy": 0.1718726187944412,
      "num_tokens": 64480536.0,
      "step": 31820
    },
    {
      "entropy": 5.982625246047974,
      "epoch": 2.781176264965481,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004228551013377816,
      "loss": 5.5321,
      "mean_token_accuracy": 0.18112628310918807,
      "num_tokens": 64489899.0,
      "step": 31825
    },
    {
      "entropy": 5.888697290420533,
      "epoch": 2.781613213318186,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042283161149549387,
      "loss": 5.521,
      "mean_token_accuracy": 0.18421332091093062,
      "num_tokens": 64500272.0,
      "step": 31830
    },
    {
      "entropy": 5.978967237472534,
      "epoch": 2.7820501616708904,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042280811881771846,
      "loss": 5.6021,
      "mean_token_accuracy": 0.1768765404820442,
      "num_tokens": 64510507.0,
      "step": 31835
    },
    {
      "entropy": 5.911552429199219,
      "epoch": 2.7824871100235953,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042278462330490584,
      "loss": 5.4974,
      "mean_token_accuracy": 0.19096709340810775,
      "num_tokens": 64521105.0,
      "step": 31840
    },
    {
      "entropy": 5.856142234802246,
      "epoch": 2.7829240583762997,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042276112495750676,
      "loss": 5.6039,
      "mean_token_accuracy": 0.1763867050409317,
      "num_tokens": 64531828.0,
      "step": 31845
    },
    {
      "entropy": 5.963413715362549,
      "epoch": 2.7833610067290047,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004227376237759718,
      "loss": 5.4509,
      "mean_token_accuracy": 0.18333663046360016,
      "num_tokens": 64540783.0,
      "step": 31850
    },
    {
      "entropy": 5.896183586120605,
      "epoch": 2.7837979550817096,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004227141197607518,
      "loss": 5.5502,
      "mean_token_accuracy": 0.17802507281303406,
      "num_tokens": 64550958.0,
      "step": 31855
    },
    {
      "entropy": 5.855904626846313,
      "epoch": 2.784234903434414,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042269061291229766,
      "loss": 5.5188,
      "mean_token_accuracy": 0.18659676164388656,
      "num_tokens": 64561666.0,
      "step": 31860
    },
    {
      "entropy": 5.976672077178955,
      "epoch": 2.7846718517871185,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042266710323106004,
      "loss": 5.6029,
      "mean_token_accuracy": 0.1689373165369034,
      "num_tokens": 64570908.0,
      "step": 31865
    },
    {
      "entropy": 5.894699907302856,
      "epoch": 2.7851088001398234,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00042264359071749017,
      "loss": 5.4807,
      "mean_token_accuracy": 0.18617452681064606,
      "num_tokens": 64580981.0,
      "step": 31870
    },
    {
      "entropy": 5.946484804153442,
      "epoch": 2.7855457484925283,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042262007537203876,
      "loss": 5.6032,
      "mean_token_accuracy": 0.17617638111114503,
      "num_tokens": 64591776.0,
      "step": 31875
    },
    {
      "entropy": 5.97418212890625,
      "epoch": 2.785982696845233,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042259655719515706,
      "loss": 5.5053,
      "mean_token_accuracy": 0.18024947494268417,
      "num_tokens": 64601201.0,
      "step": 31880
    },
    {
      "entropy": 5.9154270648956295,
      "epoch": 2.7864196451979377,
      "grad_norm": 1.125,
      "learning_rate": 0.0004225730361872959,
      "loss": 5.5582,
      "mean_token_accuracy": 0.1784817695617676,
      "num_tokens": 64610998.0,
      "step": 31885
    },
    {
      "entropy": 5.987641954421997,
      "epoch": 2.786856593550642,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004225495123489067,
      "loss": 5.6588,
      "mean_token_accuracy": 0.17298590391874313,
      "num_tokens": 64620503.0,
      "step": 31890
    },
    {
      "entropy": 5.955937099456787,
      "epoch": 2.787293541903347,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004225259856804405,
      "loss": 5.497,
      "mean_token_accuracy": 0.1878872275352478,
      "num_tokens": 64630172.0,
      "step": 31895
    },
    {
      "entropy": 5.896091270446777,
      "epoch": 2.7877304902560516,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004225024561823486,
      "loss": 5.5234,
      "mean_token_accuracy": 0.1847564160823822,
      "num_tokens": 64639737.0,
      "step": 31900
    },
    {
      "entropy": 5.935596132278443,
      "epoch": 2.7881674386087565,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004224789238550823,
      "loss": 5.463,
      "mean_token_accuracy": 0.18450547158718109,
      "num_tokens": 64650324.0,
      "step": 31905
    },
    {
      "entropy": 6.001823759078979,
      "epoch": 2.7886043869614614,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004224553886990929,
      "loss": 5.668,
      "mean_token_accuracy": 0.17348091602325438,
      "num_tokens": 64661938.0,
      "step": 31910
    },
    {
      "entropy": 5.94419846534729,
      "epoch": 2.789041335314166,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042243185071483193,
      "loss": 5.4834,
      "mean_token_accuracy": 0.1852865293622017,
      "num_tokens": 64672042.0,
      "step": 31915
    },
    {
      "entropy": 5.899397277832032,
      "epoch": 2.7894782836668703,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042240830990275085,
      "loss": 5.5615,
      "mean_token_accuracy": 0.17735703140497208,
      "num_tokens": 64681798.0,
      "step": 31920
    },
    {
      "entropy": 5.944286775588989,
      "epoch": 2.7899152320195753,
      "grad_norm": 1.390625,
      "learning_rate": 0.000422384766263301,
      "loss": 5.5634,
      "mean_token_accuracy": 0.175331811606884,
      "num_tokens": 64692329.0,
      "step": 31925
    },
    {
      "entropy": 5.950680112838745,
      "epoch": 2.79035218037228,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00042236121979693424,
      "loss": 5.5536,
      "mean_token_accuracy": 0.18077067732810975,
      "num_tokens": 64701605.0,
      "step": 31930
    },
    {
      "entropy": 5.928223323822022,
      "epoch": 2.7907891287249846,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042233767050410206,
      "loss": 5.6264,
      "mean_token_accuracy": 0.17308334410190582,
      "num_tokens": 64711649.0,
      "step": 31935
    },
    {
      "entropy": 5.95442385673523,
      "epoch": 2.7912260770776895,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004223141183852562,
      "loss": 5.5549,
      "mean_token_accuracy": 0.17548940777778627,
      "num_tokens": 64721451.0,
      "step": 31940
    },
    {
      "entropy": 5.94903154373169,
      "epoch": 2.791663025430394,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004222905634408482,
      "loss": 5.5347,
      "mean_token_accuracy": 0.18354329764842986,
      "num_tokens": 64731725.0,
      "step": 31945
    },
    {
      "entropy": 6.0069890975952145,
      "epoch": 2.792099973783099,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042226700567133015,
      "loss": 5.5474,
      "mean_token_accuracy": 0.17372677773237227,
      "num_tokens": 64741731.0,
      "step": 31950
    },
    {
      "entropy": 5.872357797622681,
      "epoch": 2.7925369221358034,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004222434450771537,
      "loss": 5.5967,
      "mean_token_accuracy": 0.17489796578884126,
      "num_tokens": 64753324.0,
      "step": 31955
    },
    {
      "entropy": 5.89802885055542,
      "epoch": 2.7929738704885083,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042221988165877083,
      "loss": 5.4979,
      "mean_token_accuracy": 0.18844634890556336,
      "num_tokens": 64763209.0,
      "step": 31960
    },
    {
      "entropy": 5.910222721099854,
      "epoch": 2.7934108188412132,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004221963154166335,
      "loss": 5.5448,
      "mean_token_accuracy": 0.1806204706430435,
      "num_tokens": 64773867.0,
      "step": 31965
    },
    {
      "entropy": 5.9212987422943115,
      "epoch": 2.7938477671939177,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004221727463511937,
      "loss": 5.5447,
      "mean_token_accuracy": 0.17249602377414702,
      "num_tokens": 64785242.0,
      "step": 31970
    },
    {
      "entropy": 5.884181785583496,
      "epoch": 2.794284715546622,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004221491744629035,
      "loss": 5.5219,
      "mean_token_accuracy": 0.18155558705329894,
      "num_tokens": 64795852.0,
      "step": 31975
    },
    {
      "entropy": 5.94847002029419,
      "epoch": 2.794721663899327,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042212559975221504,
      "loss": 5.4564,
      "mean_token_accuracy": 0.19058388024568557,
      "num_tokens": 64805982.0,
      "step": 31980
    },
    {
      "entropy": 5.910938262939453,
      "epoch": 2.795158612252032,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042210202221958055,
      "loss": 5.5812,
      "mean_token_accuracy": 0.17921919524669647,
      "num_tokens": 64816997.0,
      "step": 31985
    },
    {
      "entropy": 5.938722896575928,
      "epoch": 2.7955955606047365,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004220784418654521,
      "loss": 5.5265,
      "mean_token_accuracy": 0.18121230453252793,
      "num_tokens": 64826597.0,
      "step": 31990
    },
    {
      "entropy": 5.868039321899414,
      "epoch": 2.7960325089574414,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042205485869028207,
      "loss": 5.4923,
      "mean_token_accuracy": 0.183644862473011,
      "num_tokens": 64835917.0,
      "step": 31995
    },
    {
      "entropy": 5.96658353805542,
      "epoch": 2.796469457310146,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042203127269452284,
      "loss": 5.5327,
      "mean_token_accuracy": 0.17724790573120117,
      "num_tokens": 64846889.0,
      "step": 32000
    },
    {
      "entropy": 5.9552511215209964,
      "epoch": 2.7969064056628508,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042200768387862676,
      "loss": 5.6134,
      "mean_token_accuracy": 0.18088554739952087,
      "num_tokens": 64857321.0,
      "step": 32005
    },
    {
      "entropy": 5.868195056915283,
      "epoch": 2.7973433540155552,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042198409224304624,
      "loss": 5.5678,
      "mean_token_accuracy": 0.1809096708893776,
      "num_tokens": 64867255.0,
      "step": 32010
    },
    {
      "entropy": 5.940811634063721,
      "epoch": 2.79778030236826,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004219604977882338,
      "loss": 5.5289,
      "mean_token_accuracy": 0.17906056940555573,
      "num_tokens": 64876179.0,
      "step": 32015
    },
    {
      "entropy": 5.969575214385986,
      "epoch": 2.798217250720965,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042193690051464206,
      "loss": 5.6514,
      "mean_token_accuracy": 0.1744050547480583,
      "num_tokens": 64886387.0,
      "step": 32020
    },
    {
      "entropy": 5.951184463500977,
      "epoch": 2.7986541990736695,
      "grad_norm": 1.765625,
      "learning_rate": 0.00042191330042272353,
      "loss": 5.6051,
      "mean_token_accuracy": 0.1774564191699028,
      "num_tokens": 64896816.0,
      "step": 32025
    },
    {
      "entropy": 5.898892164230347,
      "epoch": 2.799091147426374,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042188969751293093,
      "loss": 5.5356,
      "mean_token_accuracy": 0.18803796023130417,
      "num_tokens": 64907747.0,
      "step": 32030
    },
    {
      "entropy": 5.9352028369903564,
      "epoch": 2.799528095779079,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042186609178571696,
      "loss": 5.5786,
      "mean_token_accuracy": 0.17101622223854065,
      "num_tokens": 64918066.0,
      "step": 32035
    },
    {
      "entropy": 5.975682735443115,
      "epoch": 2.799965044131784,
      "grad_norm": 1.25,
      "learning_rate": 0.0004218424832415344,
      "loss": 5.6114,
      "mean_token_accuracy": 0.174569571018219,
      "num_tokens": 64927021.0,
      "step": 32040
    },
    {
      "entropy": 5.901786279678345,
      "epoch": 2.8004019924844883,
      "grad_norm": 2.890625,
      "learning_rate": 0.00042181887188083605,
      "loss": 5.4318,
      "mean_token_accuracy": 0.1894183114171028,
      "num_tokens": 64937477.0,
      "step": 32045
    },
    {
      "entropy": 5.852967309951782,
      "epoch": 2.800838940837193,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042179525770407486,
      "loss": 5.5354,
      "mean_token_accuracy": 0.18216910660266877,
      "num_tokens": 64947178.0,
      "step": 32050
    },
    {
      "entropy": 5.906250095367431,
      "epoch": 2.8012758891898977,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042177164071170373,
      "loss": 5.6063,
      "mean_token_accuracy": 0.1730998247861862,
      "num_tokens": 64956921.0,
      "step": 32055
    },
    {
      "entropy": 5.964391231536865,
      "epoch": 2.8017128375426026,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004217480209041755,
      "loss": 5.4909,
      "mean_token_accuracy": 0.185419961810112,
      "num_tokens": 64966688.0,
      "step": 32060
    },
    {
      "entropy": 5.951850461959839,
      "epoch": 2.802149785895307,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004217243982819434,
      "loss": 5.4904,
      "mean_token_accuracy": 0.18017064779996872,
      "num_tokens": 64975577.0,
      "step": 32065
    },
    {
      "entropy": 5.915057897567749,
      "epoch": 2.802586734248012,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004217007728454605,
      "loss": 5.5768,
      "mean_token_accuracy": 0.17700579017400742,
      "num_tokens": 64986121.0,
      "step": 32070
    },
    {
      "entropy": 5.982910299301148,
      "epoch": 2.8030236826007164,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004216771445951798,
      "loss": 5.6081,
      "mean_token_accuracy": 0.17316919863224028,
      "num_tokens": 64996623.0,
      "step": 32075
    },
    {
      "entropy": 5.919320583343506,
      "epoch": 2.8034606309534214,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00042165351353155476,
      "loss": 5.4993,
      "mean_token_accuracy": 0.17811266779899598,
      "num_tokens": 65006756.0,
      "step": 32080
    },
    {
      "entropy": 5.939432859420776,
      "epoch": 2.803897579306126,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042162987965503834,
      "loss": 5.5869,
      "mean_token_accuracy": 0.17810448706150056,
      "num_tokens": 65017453.0,
      "step": 32085
    },
    {
      "entropy": 5.962155818939209,
      "epoch": 2.8043345276588307,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042160624296608403,
      "loss": 5.5917,
      "mean_token_accuracy": 0.17412515878677368,
      "num_tokens": 65028090.0,
      "step": 32090
    },
    {
      "entropy": 5.97168869972229,
      "epoch": 2.8047714760115356,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004215826034651452,
      "loss": 5.65,
      "mean_token_accuracy": 0.16973262429237365,
      "num_tokens": 65038422.0,
      "step": 32095
    },
    {
      "entropy": 5.973032855987549,
      "epoch": 2.80520842436424,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004215589611526752,
      "loss": 5.4651,
      "mean_token_accuracy": 0.18626467734575272,
      "num_tokens": 65048244.0,
      "step": 32100
    },
    {
      "entropy": 5.974317264556885,
      "epoch": 2.8056453727169446,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004215353160291275,
      "loss": 5.5791,
      "mean_token_accuracy": 0.17876603454351425,
      "num_tokens": 65058411.0,
      "step": 32105
    },
    {
      "entropy": 5.854944467544556,
      "epoch": 2.8060823210696495,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004215116680949558,
      "loss": 5.5185,
      "mean_token_accuracy": 0.18096914440393447,
      "num_tokens": 65068148.0,
      "step": 32110
    },
    {
      "entropy": 5.898129844665528,
      "epoch": 2.8065192694223544,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004214880173506133,
      "loss": 5.5622,
      "mean_token_accuracy": 0.1798992335796356,
      "num_tokens": 65078212.0,
      "step": 32115
    },
    {
      "entropy": 5.980316638946533,
      "epoch": 2.806956217775059,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042146436379655405,
      "loss": 5.5714,
      "mean_token_accuracy": 0.1758424997329712,
      "num_tokens": 65088133.0,
      "step": 32120
    },
    {
      "entropy": 5.905688762664795,
      "epoch": 2.807393166127764,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042144070743323155,
      "loss": 5.5052,
      "mean_token_accuracy": 0.17837952971458435,
      "num_tokens": 65097671.0,
      "step": 32125
    },
    {
      "entropy": 5.983434534072876,
      "epoch": 2.8078301144804683,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004214170482610995,
      "loss": 5.6568,
      "mean_token_accuracy": 0.17082528918981552,
      "num_tokens": 65107833.0,
      "step": 32130
    },
    {
      "entropy": 5.950030660629272,
      "epoch": 2.808267062833173,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042139338628061177,
      "loss": 5.5261,
      "mean_token_accuracy": 0.18014440089464187,
      "num_tokens": 65118414.0,
      "step": 32135
    },
    {
      "entropy": 5.958419370651245,
      "epoch": 2.8087040111858776,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042136972149222203,
      "loss": 5.6563,
      "mean_token_accuracy": 0.16968477368354798,
      "num_tokens": 65128920.0,
      "step": 32140
    },
    {
      "entropy": 5.873667144775391,
      "epoch": 2.8091409595385826,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004213460538963845,
      "loss": 5.4934,
      "mean_token_accuracy": 0.18296280652284622,
      "num_tokens": 65138869.0,
      "step": 32145
    },
    {
      "entropy": 5.93813157081604,
      "epoch": 2.8095779078912875,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004213223834935529,
      "loss": 5.5535,
      "mean_token_accuracy": 0.18645016402006148,
      "num_tokens": 65148737.0,
      "step": 32150
    },
    {
      "entropy": 5.979941701889038,
      "epoch": 2.810014856243992,
      "grad_norm": 1.25,
      "learning_rate": 0.00042129871028418137,
      "loss": 5.6045,
      "mean_token_accuracy": 0.17370517402887345,
      "num_tokens": 65158075.0,
      "step": 32155
    },
    {
      "entropy": 6.01328444480896,
      "epoch": 2.8104518045966964,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004212750342687239,
      "loss": 5.6118,
      "mean_token_accuracy": 0.17398294657468796,
      "num_tokens": 65167788.0,
      "step": 32160
    },
    {
      "entropy": 5.901038122177124,
      "epoch": 2.8108887529494013,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004212513554476346,
      "loss": 5.5792,
      "mean_token_accuracy": 0.18087365180253984,
      "num_tokens": 65177581.0,
      "step": 32165
    },
    {
      "entropy": 5.923845052719116,
      "epoch": 2.8113257013021062,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042122767382136766,
      "loss": 5.5377,
      "mean_token_accuracy": 0.18772168904542924,
      "num_tokens": 65186546.0,
      "step": 32170
    },
    {
      "entropy": 5.830459022521973,
      "epoch": 2.8117626496548107,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004212039893903774,
      "loss": 5.4388,
      "mean_token_accuracy": 0.19395436048507692,
      "num_tokens": 65195717.0,
      "step": 32175
    },
    {
      "entropy": 5.942040967941284,
      "epoch": 2.8121995980075156,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042118030215511793,
      "loss": 5.5972,
      "mean_token_accuracy": 0.17202506214380264,
      "num_tokens": 65206090.0,
      "step": 32180
    },
    {
      "entropy": 5.915534591674804,
      "epoch": 2.81263654636022,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042115661211604373,
      "loss": 5.5883,
      "mean_token_accuracy": 0.18069159537553786,
      "num_tokens": 65217194.0,
      "step": 32185
    },
    {
      "entropy": 6.028683137893677,
      "epoch": 2.813073494712925,
      "grad_norm": 1.453125,
      "learning_rate": 0.000421132919273609,
      "loss": 5.6558,
      "mean_token_accuracy": 0.17484935373067856,
      "num_tokens": 65228906.0,
      "step": 32190
    },
    {
      "entropy": 5.967750597000122,
      "epoch": 2.8135104430656295,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004211092236282685,
      "loss": 5.6016,
      "mean_token_accuracy": 0.17460673600435256,
      "num_tokens": 65238606.0,
      "step": 32195
    },
    {
      "entropy": 5.89074559211731,
      "epoch": 2.8139473914183344,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042108552518047644,
      "loss": 5.521,
      "mean_token_accuracy": 0.1823913723230362,
      "num_tokens": 65248259.0,
      "step": 32200
    },
    {
      "entropy": 5.953576469421387,
      "epoch": 2.8143843397710393,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004210618239306874,
      "loss": 5.6098,
      "mean_token_accuracy": 0.1756802961230278,
      "num_tokens": 65259294.0,
      "step": 32205
    },
    {
      "entropy": 6.002966117858887,
      "epoch": 2.8148212881237438,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004210381198793561,
      "loss": 5.6149,
      "mean_token_accuracy": 0.17718038707971573,
      "num_tokens": 65269498.0,
      "step": 32210
    },
    {
      "entropy": 5.99600076675415,
      "epoch": 2.8152582364764482,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004210144130269371,
      "loss": 5.6695,
      "mean_token_accuracy": 0.16377051770687104,
      "num_tokens": 65280027.0,
      "step": 32215
    },
    {
      "entropy": 5.913276290893554,
      "epoch": 2.815695184829153,
      "grad_norm": 1.546875,
      "learning_rate": 0.00042099070337388517,
      "loss": 5.5463,
      "mean_token_accuracy": 0.1779095083475113,
      "num_tokens": 65290451.0,
      "step": 32220
    },
    {
      "entropy": 5.905139636993408,
      "epoch": 2.816132133181858,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042096699092065504,
      "loss": 5.528,
      "mean_token_accuracy": 0.18553666770458221,
      "num_tokens": 65299906.0,
      "step": 32225
    },
    {
      "entropy": 5.930331230163574,
      "epoch": 2.8165690815345625,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004209432756677015,
      "loss": 5.535,
      "mean_token_accuracy": 0.17985738664865494,
      "num_tokens": 65310197.0,
      "step": 32230
    },
    {
      "entropy": 5.944227313995361,
      "epoch": 2.8170060298872674,
      "grad_norm": 1.125,
      "learning_rate": 0.00042091955761547963,
      "loss": 5.6119,
      "mean_token_accuracy": 0.1757593959569931,
      "num_tokens": 65320779.0,
      "step": 32235
    },
    {
      "entropy": 5.871133947372437,
      "epoch": 2.817442978239972,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000420895836764444,
      "loss": 5.4579,
      "mean_token_accuracy": 0.18542921543121338,
      "num_tokens": 65330578.0,
      "step": 32240
    },
    {
      "entropy": 5.889280652999878,
      "epoch": 2.817879926592677,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042087211311504986,
      "loss": 5.5646,
      "mean_token_accuracy": 0.17728192061185838,
      "num_tokens": 65341202.0,
      "step": 32245
    },
    {
      "entropy": 5.985068225860596,
      "epoch": 2.8183168749453813,
      "grad_norm": 1.25,
      "learning_rate": 0.0004208483866677521,
      "loss": 5.554,
      "mean_token_accuracy": 0.18414463549852372,
      "num_tokens": 65352584.0,
      "step": 32250
    },
    {
      "entropy": 5.962665319442749,
      "epoch": 2.818753823298086,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004208246574230059,
      "loss": 5.5151,
      "mean_token_accuracy": 0.17811396718025208,
      "num_tokens": 65363191.0,
      "step": 32255
    },
    {
      "entropy": 5.873468351364136,
      "epoch": 2.819190771650791,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042080092538126633,
      "loss": 5.5274,
      "mean_token_accuracy": 0.1801239475607872,
      "num_tokens": 65373880.0,
      "step": 32260
    },
    {
      "entropy": 5.914931440353394,
      "epoch": 2.8196277200034956,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042077719054298867,
      "loss": 5.6074,
      "mean_token_accuracy": 0.1799386441707611,
      "num_tokens": 65383149.0,
      "step": 32265
    },
    {
      "entropy": 5.9398791790008545,
      "epoch": 2.8200646683562,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004207534529086281,
      "loss": 5.5073,
      "mean_token_accuracy": 0.18963757902383804,
      "num_tokens": 65393005.0,
      "step": 32270
    },
    {
      "entropy": 5.955580377578736,
      "epoch": 2.820501616708905,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042072971247863984,
      "loss": 5.5763,
      "mean_token_accuracy": 0.17050206810235977,
      "num_tokens": 65402640.0,
      "step": 32275
    },
    {
      "entropy": 5.97556848526001,
      "epoch": 2.82093856506161,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004207059692534794,
      "loss": 5.6152,
      "mean_token_accuracy": 0.17087069898843765,
      "num_tokens": 65412863.0,
      "step": 32280
    },
    {
      "entropy": 5.88806791305542,
      "epoch": 2.8213755134143144,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042068222323360214,
      "loss": 5.5949,
      "mean_token_accuracy": 0.18099595457315446,
      "num_tokens": 65422581.0,
      "step": 32285
    },
    {
      "entropy": 5.93459119796753,
      "epoch": 2.8218124617670193,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042065847441946354,
      "loss": 5.6011,
      "mean_token_accuracy": 0.17884751707315444,
      "num_tokens": 65433604.0,
      "step": 32290
    },
    {
      "entropy": 5.894299507141113,
      "epoch": 2.8222494101197237,
      "grad_norm": 1.171875,
      "learning_rate": 0.000420634722811519,
      "loss": 5.4769,
      "mean_token_accuracy": 0.1829648047685623,
      "num_tokens": 65442956.0,
      "step": 32295
    },
    {
      "entropy": 5.935194301605224,
      "epoch": 2.8226863584724287,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042061096841022416,
      "loss": 5.5602,
      "mean_token_accuracy": 0.17626478224992753,
      "num_tokens": 65452441.0,
      "step": 32300
    },
    {
      "entropy": 5.961315727233886,
      "epoch": 2.823123306825133,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042058721121603467,
      "loss": 5.6243,
      "mean_token_accuracy": 0.17851441651582717,
      "num_tokens": 65463628.0,
      "step": 32305
    },
    {
      "entropy": 5.930991792678833,
      "epoch": 2.823560255177838,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004205634512294062,
      "loss": 5.5283,
      "mean_token_accuracy": 0.18270598948001862,
      "num_tokens": 65472771.0,
      "step": 32310
    },
    {
      "entropy": 5.943559980392456,
      "epoch": 2.823997203530543,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042053968845079436,
      "loss": 5.571,
      "mean_token_accuracy": 0.17722005099058152,
      "num_tokens": 65482678.0,
      "step": 32315
    },
    {
      "entropy": 6.070168590545654,
      "epoch": 2.8244341518832474,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042051592288065514,
      "loss": 5.6004,
      "mean_token_accuracy": 0.1740674525499344,
      "num_tokens": 65493127.0,
      "step": 32320
    },
    {
      "entropy": 5.930430173873901,
      "epoch": 2.824871100235952,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004204921545194443,
      "loss": 5.605,
      "mean_token_accuracy": 0.17845335751771926,
      "num_tokens": 65502508.0,
      "step": 32325
    },
    {
      "entropy": 5.889770412445069,
      "epoch": 2.825308048588657,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004204683833676176,
      "loss": 5.6047,
      "mean_token_accuracy": 0.17764317691326142,
      "num_tokens": 65512336.0,
      "step": 32330
    },
    {
      "entropy": 5.9418824195861815,
      "epoch": 2.8257449969413617,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004204446094256312,
      "loss": 5.5406,
      "mean_token_accuracy": 0.17391349822282792,
      "num_tokens": 65523139.0,
      "step": 32335
    },
    {
      "entropy": 5.935797071456909,
      "epoch": 2.826181945294066,
      "grad_norm": 1.125,
      "learning_rate": 0.00042042083269394083,
      "loss": 5.5213,
      "mean_token_accuracy": 0.1826691821217537,
      "num_tokens": 65532846.0,
      "step": 32340
    },
    {
      "entropy": 5.950242185592652,
      "epoch": 2.826618893646771,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004203970531730028,
      "loss": 5.5366,
      "mean_token_accuracy": 0.17420467287302016,
      "num_tokens": 65544493.0,
      "step": 32345
    },
    {
      "entropy": 5.935913610458374,
      "epoch": 2.8270558419994756,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042037327086327304,
      "loss": 5.6162,
      "mean_token_accuracy": 0.18004437088966369,
      "num_tokens": 65554183.0,
      "step": 32350
    },
    {
      "entropy": 5.968365335464478,
      "epoch": 2.8274927903521805,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004203494857652078,
      "loss": 5.5386,
      "mean_token_accuracy": 0.1847367227077484,
      "num_tokens": 65564068.0,
      "step": 32355
    },
    {
      "entropy": 5.847999143600464,
      "epoch": 2.827929738704885,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042032569787926326,
      "loss": 5.4829,
      "mean_token_accuracy": 0.1880132347345352,
      "num_tokens": 65574272.0,
      "step": 32360
    },
    {
      "entropy": 5.933985185623169,
      "epoch": 2.82836668705759,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042030190720589573,
      "loss": 5.5715,
      "mean_token_accuracy": 0.17721689194440843,
      "num_tokens": 65584890.0,
      "step": 32365
    },
    {
      "entropy": 5.987453269958496,
      "epoch": 2.8288036354102943,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004202781137455614,
      "loss": 5.5706,
      "mean_token_accuracy": 0.17705088555812837,
      "num_tokens": 65595776.0,
      "step": 32370
    },
    {
      "entropy": 5.888703727722168,
      "epoch": 2.8292405837629993,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004202543174987168,
      "loss": 5.5322,
      "mean_token_accuracy": 0.18410244286060334,
      "num_tokens": 65606234.0,
      "step": 32375
    },
    {
      "entropy": 5.947072553634643,
      "epoch": 2.8296775321157037,
      "grad_norm": 1.125,
      "learning_rate": 0.00042023051846581824,
      "loss": 5.6772,
      "mean_token_accuracy": 0.17404940575361252,
      "num_tokens": 65616576.0,
      "step": 32380
    },
    {
      "entropy": 5.908352279663086,
      "epoch": 2.8301144804684086,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004202067166473222,
      "loss": 5.4946,
      "mean_token_accuracy": 0.18892699778079985,
      "num_tokens": 65626792.0,
      "step": 32385
    },
    {
      "entropy": 5.9301519870758055,
      "epoch": 2.8305514288211135,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004201829120436854,
      "loss": 5.5354,
      "mean_token_accuracy": 0.18013690561056137,
      "num_tokens": 65637041.0,
      "step": 32390
    },
    {
      "entropy": 5.882362985610962,
      "epoch": 2.830988377173818,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004201591046553641,
      "loss": 5.6084,
      "mean_token_accuracy": 0.17649733126163483,
      "num_tokens": 65647132.0,
      "step": 32395
    },
    {
      "entropy": 5.985000705718994,
      "epoch": 2.831425325526523,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042013529448281517,
      "loss": 5.6036,
      "mean_token_accuracy": 0.17512834668159485,
      "num_tokens": 65657372.0,
      "step": 32400
    },
    {
      "entropy": 5.928832149505615,
      "epoch": 2.8318622738792274,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004201114815264952,
      "loss": 5.5638,
      "mean_token_accuracy": 0.1799599528312683,
      "num_tokens": 65667118.0,
      "step": 32405
    },
    {
      "entropy": 5.969529438018799,
      "epoch": 2.8322992222319323,
      "grad_norm": 1.03125,
      "learning_rate": 0.000420087665786861,
      "loss": 5.6085,
      "mean_token_accuracy": 0.17785904854536055,
      "num_tokens": 65677830.0,
      "step": 32410
    },
    {
      "entropy": 5.848909473419189,
      "epoch": 2.832736170584637,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042006384726436943,
      "loss": 5.5495,
      "mean_token_accuracy": 0.1834823176264763,
      "num_tokens": 65687238.0,
      "step": 32415
    },
    {
      "entropy": 5.907644748687744,
      "epoch": 2.8331731189373417,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042004002595947713,
      "loss": 5.4986,
      "mean_token_accuracy": 0.18842406421899796,
      "num_tokens": 65697547.0,
      "step": 32420
    },
    {
      "entropy": 5.952865028381348,
      "epoch": 2.833610067290046,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004200162018726412,
      "loss": 5.665,
      "mean_token_accuracy": 0.17439982891082764,
      "num_tokens": 65709272.0,
      "step": 32425
    },
    {
      "entropy": 5.907271242141723,
      "epoch": 2.834047015642751,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004199923750043185,
      "loss": 5.6055,
      "mean_token_accuracy": 0.1766653686761856,
      "num_tokens": 65720986.0,
      "step": 32430
    },
    {
      "entropy": 5.989446353912354,
      "epoch": 2.8344839639954555,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041996854535496616,
      "loss": 5.5853,
      "mean_token_accuracy": 0.1784168526530266,
      "num_tokens": 65731339.0,
      "step": 32435
    },
    {
      "entropy": 5.968107652664185,
      "epoch": 2.8349209123481605,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004199447129250411,
      "loss": 5.5799,
      "mean_token_accuracy": 0.18125290423631668,
      "num_tokens": 65741826.0,
      "step": 32440
    },
    {
      "entropy": 5.91627311706543,
      "epoch": 2.8353578607008654,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004199208777150004,
      "loss": 5.6115,
      "mean_token_accuracy": 0.17972599416971208,
      "num_tokens": 65752925.0,
      "step": 32445
    },
    {
      "entropy": 5.983340930938721,
      "epoch": 2.83579480905357,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004198970397253015,
      "loss": 5.5832,
      "mean_token_accuracy": 0.17540715783834457,
      "num_tokens": 65763948.0,
      "step": 32450
    },
    {
      "entropy": 5.961351442337036,
      "epoch": 2.8362317574062743,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004198731989564014,
      "loss": 5.5451,
      "mean_token_accuracy": 0.1812466025352478,
      "num_tokens": 65774086.0,
      "step": 32455
    },
    {
      "entropy": 5.841054391860962,
      "epoch": 2.8366687057589792,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004198493554087574,
      "loss": 5.4716,
      "mean_token_accuracy": 0.1869415447115898,
      "num_tokens": 65783630.0,
      "step": 32460
    },
    {
      "entropy": 5.922359609603882,
      "epoch": 2.837105654111684,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041982550908282686,
      "loss": 5.5725,
      "mean_token_accuracy": 0.18009048253297805,
      "num_tokens": 65793903.0,
      "step": 32465
    },
    {
      "entropy": 5.974708938598633,
      "epoch": 2.8375426024643886,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041980165997906714,
      "loss": 5.4616,
      "mean_token_accuracy": 0.19027011692523957,
      "num_tokens": 65803483.0,
      "step": 32470
    },
    {
      "entropy": 5.900651836395264,
      "epoch": 2.8379795508170935,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004197778080979358,
      "loss": 5.5768,
      "mean_token_accuracy": 0.17978862076997756,
      "num_tokens": 65814336.0,
      "step": 32475
    },
    {
      "entropy": 5.9074944972991945,
      "epoch": 2.838416499169798,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041975395343989017,
      "loss": 5.6218,
      "mean_token_accuracy": 0.17569705545902253,
      "num_tokens": 65825525.0,
      "step": 32480
    },
    {
      "entropy": 5.945803546905518,
      "epoch": 2.838853447522503,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004197300960053879,
      "loss": 5.5618,
      "mean_token_accuracy": 0.18648894727230073,
      "num_tokens": 65836276.0,
      "step": 32485
    },
    {
      "entropy": 5.961492919921875,
      "epoch": 2.8392903958752074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004197062357948865,
      "loss": 5.5781,
      "mean_token_accuracy": 0.17231428474187852,
      "num_tokens": 65846164.0,
      "step": 32490
    },
    {
      "entropy": 5.939288854598999,
      "epoch": 2.8397273442279123,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041968237280884364,
      "loss": 5.5615,
      "mean_token_accuracy": 0.17670219987630845,
      "num_tokens": 65855759.0,
      "step": 32495
    },
    {
      "entropy": 5.871018123626709,
      "epoch": 2.840164292580617,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004196585070477172,
      "loss": 5.4125,
      "mean_token_accuracy": 0.19554568231105804,
      "num_tokens": 65866009.0,
      "step": 32500
    },
    {
      "entropy": 5.962742137908935,
      "epoch": 2.8406012409333217,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041963463851196466,
      "loss": 5.5207,
      "mean_token_accuracy": 0.1818176820874214,
      "num_tokens": 65875707.0,
      "step": 32505
    },
    {
      "entropy": 5.936881494522095,
      "epoch": 2.841038189286026,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041961076720204403,
      "loss": 5.5394,
      "mean_token_accuracy": 0.18623351752758027,
      "num_tokens": 65884396.0,
      "step": 32510
    },
    {
      "entropy": 5.9080846309661865,
      "epoch": 2.841475137638731,
      "grad_norm": 1.125,
      "learning_rate": 0.0004195868931184131,
      "loss": 5.571,
      "mean_token_accuracy": 0.1748806431889534,
      "num_tokens": 65894278.0,
      "step": 32515
    },
    {
      "entropy": 5.941784954071045,
      "epoch": 2.841912085991436,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041956301626152974,
      "loss": 5.5736,
      "mean_token_accuracy": 0.18087685853242874,
      "num_tokens": 65903863.0,
      "step": 32520
    },
    {
      "entropy": 5.865962743759155,
      "epoch": 2.8423490343441404,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041953913663185207,
      "loss": 5.4215,
      "mean_token_accuracy": 0.18679831475019454,
      "num_tokens": 65913996.0,
      "step": 32525
    },
    {
      "entropy": 5.819665002822876,
      "epoch": 2.8427859826968453,
      "grad_norm": 1.28125,
      "learning_rate": 0.000419515254229838,
      "loss": 5.483,
      "mean_token_accuracy": 0.19248989075422288,
      "num_tokens": 65923457.0,
      "step": 32530
    },
    {
      "entropy": 5.961307048797607,
      "epoch": 2.84322293104955,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041949136905594555,
      "loss": 5.5552,
      "mean_token_accuracy": 0.1786558672785759,
      "num_tokens": 65933132.0,
      "step": 32535
    },
    {
      "entropy": 5.9466088771820065,
      "epoch": 2.8436598794022547,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000419467481110633,
      "loss": 5.5156,
      "mean_token_accuracy": 0.18385612666606904,
      "num_tokens": 65942784.0,
      "step": 32540
    },
    {
      "entropy": 5.895177745819092,
      "epoch": 2.844096827754959,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004194435903943585,
      "loss": 5.4901,
      "mean_token_accuracy": 0.18873875439167023,
      "num_tokens": 65952734.0,
      "step": 32545
    },
    {
      "entropy": 5.8363536357879635,
      "epoch": 2.844533776107664,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041941969690758023,
      "loss": 5.4035,
      "mean_token_accuracy": 0.1963243916630745,
      "num_tokens": 65962732.0,
      "step": 32550
    },
    {
      "entropy": 5.91572117805481,
      "epoch": 2.844970724460369,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041939580065075645,
      "loss": 5.6212,
      "mean_token_accuracy": 0.17660484760999678,
      "num_tokens": 65973287.0,
      "step": 32555
    },
    {
      "entropy": 5.988425827026367,
      "epoch": 2.8454076728130735,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004193719016243456,
      "loss": 5.5087,
      "mean_token_accuracy": 0.18227145671844483,
      "num_tokens": 65982034.0,
      "step": 32560
    },
    {
      "entropy": 5.901849699020386,
      "epoch": 2.845844621165778,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041934799982880594,
      "loss": 5.5464,
      "mean_token_accuracy": 0.18210566341876983,
      "num_tokens": 65992612.0,
      "step": 32565
    },
    {
      "entropy": 5.903510189056396,
      "epoch": 2.846281569518483,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041932409526459607,
      "loss": 5.5172,
      "mean_token_accuracy": 0.18377561420202254,
      "num_tokens": 66002788.0,
      "step": 32570
    },
    {
      "entropy": 5.90720043182373,
      "epoch": 2.846718517871188,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041930018793217443,
      "loss": 5.5808,
      "mean_token_accuracy": 0.1801111102104187,
      "num_tokens": 66012559.0,
      "step": 32575
    },
    {
      "entropy": 5.929148101806641,
      "epoch": 2.8471554662238923,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004192762778319996,
      "loss": 5.5836,
      "mean_token_accuracy": 0.17272269874811172,
      "num_tokens": 66023230.0,
      "step": 32580
    },
    {
      "entropy": 5.946524381637573,
      "epoch": 2.847592414576597,
      "grad_norm": 1.78125,
      "learning_rate": 0.00041925236496453006,
      "loss": 5.6311,
      "mean_token_accuracy": 0.17971227318048477,
      "num_tokens": 66033556.0,
      "step": 32585
    },
    {
      "entropy": 6.006966590881348,
      "epoch": 2.8480293629293016,
      "grad_norm": 1.125,
      "learning_rate": 0.00041922844933022467,
      "loss": 5.5905,
      "mean_token_accuracy": 0.17659637778997422,
      "num_tokens": 66042495.0,
      "step": 32590
    },
    {
      "entropy": 5.909358358383178,
      "epoch": 2.8484663112820066,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041920453092954204,
      "loss": 5.5318,
      "mean_token_accuracy": 0.18184587210416794,
      "num_tokens": 66051933.0,
      "step": 32595
    },
    {
      "entropy": 5.9206029891967775,
      "epoch": 2.848903259634711,
      "grad_norm": 1.046875,
      "learning_rate": 0.000419180609762941,
      "loss": 5.4688,
      "mean_token_accuracy": 0.18769942969083786,
      "num_tokens": 66060964.0,
      "step": 32600
    },
    {
      "entropy": 5.952478742599487,
      "epoch": 2.849340207987416,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041915668583088016,
      "loss": 5.5343,
      "mean_token_accuracy": 0.17890715301036836,
      "num_tokens": 66070210.0,
      "step": 32605
    },
    {
      "entropy": 5.939025974273681,
      "epoch": 2.849777156340121,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041913275913381864,
      "loss": 5.6234,
      "mean_token_accuracy": 0.17953532487154006,
      "num_tokens": 66081010.0,
      "step": 32610
    },
    {
      "entropy": 5.89336199760437,
      "epoch": 2.8502141046928253,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004191088296722153,
      "loss": 5.4864,
      "mean_token_accuracy": 0.18468460142612458,
      "num_tokens": 66090448.0,
      "step": 32615
    },
    {
      "entropy": 5.93741774559021,
      "epoch": 2.85065105304553,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000419084897446529,
      "loss": 5.5999,
      "mean_token_accuracy": 0.17762184590101243,
      "num_tokens": 66100357.0,
      "step": 32620
    },
    {
      "entropy": 5.914103364944458,
      "epoch": 2.8510880013982347,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000419060962457219,
      "loss": 5.5328,
      "mean_token_accuracy": 0.18073560148477555,
      "num_tokens": 66111068.0,
      "step": 32625
    },
    {
      "entropy": 5.939388847351074,
      "epoch": 2.8515249497509396,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041903702470474414,
      "loss": 5.5705,
      "mean_token_accuracy": 0.17801453918218613,
      "num_tokens": 66122185.0,
      "step": 32630
    },
    {
      "entropy": 5.977344799041748,
      "epoch": 2.851961898103644,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004190130841895637,
      "loss": 5.5443,
      "mean_token_accuracy": 0.17902077436447145,
      "num_tokens": 66133192.0,
      "step": 32635
    },
    {
      "entropy": 5.924424648284912,
      "epoch": 2.852398846456349,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004189891409121369,
      "loss": 5.6132,
      "mean_token_accuracy": 0.1755354106426239,
      "num_tokens": 66143381.0,
      "step": 32640
    },
    {
      "entropy": 5.948329162597656,
      "epoch": 2.8528357948090535,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041896519487292285,
      "loss": 5.5282,
      "mean_token_accuracy": 0.18157321363687515,
      "num_tokens": 66152883.0,
      "step": 32645
    },
    {
      "entropy": 5.973902797698974,
      "epoch": 2.8532727431617584,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041894124607238104,
      "loss": 5.4715,
      "mean_token_accuracy": 0.18273821771144866,
      "num_tokens": 66162725.0,
      "step": 32650
    },
    {
      "entropy": 5.940306186676025,
      "epoch": 2.853709691514463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004189172945109706,
      "loss": 5.5866,
      "mean_token_accuracy": 0.1775839775800705,
      "num_tokens": 66171478.0,
      "step": 32655
    },
    {
      "entropy": 5.784211444854736,
      "epoch": 2.8541466398671678,
      "grad_norm": 1.875,
      "learning_rate": 0.0004188933401891511,
      "loss": 5.4306,
      "mean_token_accuracy": 0.18428168594837188,
      "num_tokens": 66181044.0,
      "step": 32660
    },
    {
      "entropy": 5.916531610488891,
      "epoch": 2.8545835882198727,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004188693831073818,
      "loss": 5.5543,
      "mean_token_accuracy": 0.1835540145635605,
      "num_tokens": 66190848.0,
      "step": 32665
    },
    {
      "entropy": 5.974076461791992,
      "epoch": 2.855020536572577,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041884542326612245,
      "loss": 5.5829,
      "mean_token_accuracy": 0.1731831043958664,
      "num_tokens": 66200756.0,
      "step": 32670
    },
    {
      "entropy": 5.953938293457031,
      "epoch": 2.8554574849252816,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041882146066583257,
      "loss": 5.6023,
      "mean_token_accuracy": 0.17488162815570832,
      "num_tokens": 66211463.0,
      "step": 32675
    },
    {
      "entropy": 5.972238683700562,
      "epoch": 2.8558944332779865,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041879749530697165,
      "loss": 5.576,
      "mean_token_accuracy": 0.1765894591808319,
      "num_tokens": 66222522.0,
      "step": 32680
    },
    {
      "entropy": 5.922454118728638,
      "epoch": 2.8563313816306914,
      "grad_norm": 1.0,
      "learning_rate": 0.0004187735271899994,
      "loss": 5.5874,
      "mean_token_accuracy": 0.1806205004453659,
      "num_tokens": 66232741.0,
      "step": 32685
    },
    {
      "entropy": 5.92623987197876,
      "epoch": 2.856768329983396,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041874955631537555,
      "loss": 5.4779,
      "mean_token_accuracy": 0.18182586282491683,
      "num_tokens": 66242845.0,
      "step": 32690
    },
    {
      "entropy": 5.8934589385986325,
      "epoch": 2.857205278336101,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041872558268355993,
      "loss": 5.4916,
      "mean_token_accuracy": 0.18497335016727448,
      "num_tokens": 66252697.0,
      "step": 32695
    },
    {
      "entropy": 5.8680603981018065,
      "epoch": 2.8576422266888053,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041870160629501224,
      "loss": 5.572,
      "mean_token_accuracy": 0.18003729581832886,
      "num_tokens": 66263607.0,
      "step": 32700
    },
    {
      "entropy": 5.8932453155517575,
      "epoch": 2.85807917504151,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041867762715019257,
      "loss": 5.4964,
      "mean_token_accuracy": 0.18339649587869644,
      "num_tokens": 66274352.0,
      "step": 32705
    },
    {
      "entropy": 5.928488779067993,
      "epoch": 2.8585161233942147,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041865364524956066,
      "loss": 5.5425,
      "mean_token_accuracy": 0.18720127493143082,
      "num_tokens": 66283972.0,
      "step": 32710
    },
    {
      "entropy": 5.920738649368286,
      "epoch": 2.8589530717469196,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041862966059357665,
      "loss": 5.6001,
      "mean_token_accuracy": 0.17647165954113006,
      "num_tokens": 66295179.0,
      "step": 32715
    },
    {
      "entropy": 5.990260457992553,
      "epoch": 2.859390020099624,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004186056731827003,
      "loss": 5.5892,
      "mean_token_accuracy": 0.17500533312559127,
      "num_tokens": 66306065.0,
      "step": 32720
    },
    {
      "entropy": 5.886679553985596,
      "epoch": 2.859826968452329,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000418581683017392,
      "loss": 5.5024,
      "mean_token_accuracy": 0.1836540773510933,
      "num_tokens": 66315491.0,
      "step": 32725
    },
    {
      "entropy": 5.888741493225098,
      "epoch": 2.8602639168050334,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004185576900981119,
      "loss": 5.4986,
      "mean_token_accuracy": 0.18328256607055665,
      "num_tokens": 66324327.0,
      "step": 32730
    },
    {
      "entropy": 5.940319776535034,
      "epoch": 2.8607008651577384,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004185336944253199,
      "loss": 5.5171,
      "mean_token_accuracy": 0.17987994402647017,
      "num_tokens": 66334201.0,
      "step": 32735
    },
    {
      "entropy": 5.966088342666626,
      "epoch": 2.8611378135104433,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041850969599947655,
      "loss": 5.5235,
      "mean_token_accuracy": 0.18443182855844498,
      "num_tokens": 66344745.0,
      "step": 32740
    },
    {
      "entropy": 5.8206277847290036,
      "epoch": 2.8615747618631477,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041848569482104197,
      "loss": 5.515,
      "mean_token_accuracy": 0.18771580159664153,
      "num_tokens": 66354815.0,
      "step": 32745
    },
    {
      "entropy": 5.9325034618377686,
      "epoch": 2.862011710215852,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004184616908904765,
      "loss": 5.5453,
      "mean_token_accuracy": 0.18223462551832198,
      "num_tokens": 66364744.0,
      "step": 32750
    },
    {
      "entropy": 5.979539680480957,
      "epoch": 2.862448658568557,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004184376842082408,
      "loss": 5.6099,
      "mean_token_accuracy": 0.17535383254289627,
      "num_tokens": 66376363.0,
      "step": 32755
    },
    {
      "entropy": 5.919520902633667,
      "epoch": 2.862885606921262,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041841367477479503,
      "loss": 5.5801,
      "mean_token_accuracy": 0.17458575069904328,
      "num_tokens": 66386149.0,
      "step": 32760
    },
    {
      "entropy": 5.960975408554077,
      "epoch": 2.8633225552739665,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004183896625905999,
      "loss": 5.4806,
      "mean_token_accuracy": 0.1861794710159302,
      "num_tokens": 66396874.0,
      "step": 32765
    },
    {
      "entropy": 5.859244441986084,
      "epoch": 2.8637595036266714,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041836564765611583,
      "loss": 5.5052,
      "mean_token_accuracy": 0.17949940413236617,
      "num_tokens": 66406761.0,
      "step": 32770
    },
    {
      "entropy": 5.958828544616699,
      "epoch": 2.864196451979376,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041834162997180364,
      "loss": 5.6391,
      "mean_token_accuracy": 0.175751955807209,
      "num_tokens": 66416947.0,
      "step": 32775
    },
    {
      "entropy": 5.930038785934448,
      "epoch": 2.864633400332081,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004183176095381238,
      "loss": 5.5183,
      "mean_token_accuracy": 0.18514966517686843,
      "num_tokens": 66427980.0,
      "step": 32780
    },
    {
      "entropy": 5.8127106666564945,
      "epoch": 2.8650703486847853,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041829358635553707,
      "loss": 5.4279,
      "mean_token_accuracy": 0.1891499176621437,
      "num_tokens": 66437226.0,
      "step": 32785
    },
    {
      "entropy": 5.884727907180786,
      "epoch": 2.86550729703749,
      "grad_norm": 1.515625,
      "learning_rate": 0.00041826956042450436,
      "loss": 5.5995,
      "mean_token_accuracy": 0.17685236185789108,
      "num_tokens": 66447820.0,
      "step": 32790
    },
    {
      "entropy": 5.963935089111328,
      "epoch": 2.865944245390195,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004182455317454863,
      "loss": 5.5634,
      "mean_token_accuracy": 0.17424123287200927,
      "num_tokens": 66459727.0,
      "step": 32795
    },
    {
      "entropy": 5.910506105422973,
      "epoch": 2.8663811937428996,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041822150031894397,
      "loss": 5.5132,
      "mean_token_accuracy": 0.18196303248405457,
      "num_tokens": 66469743.0,
      "step": 32800
    },
    {
      "entropy": 5.961207342147827,
      "epoch": 2.866818142095604,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041819746614533824,
      "loss": 5.6139,
      "mean_token_accuracy": 0.177800914645195,
      "num_tokens": 66479974.0,
      "step": 32805
    },
    {
      "entropy": 5.894542121887207,
      "epoch": 2.867255090448309,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041817342922513,
      "loss": 5.4583,
      "mean_token_accuracy": 0.18756259828805924,
      "num_tokens": 66489605.0,
      "step": 32810
    },
    {
      "entropy": 5.907950353622437,
      "epoch": 2.867692038801014,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041814938955878035,
      "loss": 5.5736,
      "mean_token_accuracy": 0.1831655666232109,
      "num_tokens": 66499454.0,
      "step": 32815
    },
    {
      "entropy": 5.962640714645386,
      "epoch": 2.8681289871537183,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041812534714675046,
      "loss": 5.5906,
      "mean_token_accuracy": 0.1791380375623703,
      "num_tokens": 66510076.0,
      "step": 32820
    },
    {
      "entropy": 5.980338191986084,
      "epoch": 2.8685659355064232,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041810130198950137,
      "loss": 5.5639,
      "mean_token_accuracy": 0.1790060207247734,
      "num_tokens": 66520659.0,
      "step": 32825
    },
    {
      "entropy": 5.969989252090454,
      "epoch": 2.8690028838591277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004180772540874943,
      "loss": 5.6148,
      "mean_token_accuracy": 0.17664211988449097,
      "num_tokens": 66531387.0,
      "step": 32830
    },
    {
      "entropy": 5.958529996871948,
      "epoch": 2.8694398322118326,
      "grad_norm": 2.40625,
      "learning_rate": 0.00041805320344119064,
      "loss": 5.5286,
      "mean_token_accuracy": 0.1866590052843094,
      "num_tokens": 66540654.0,
      "step": 32835
    },
    {
      "entropy": 5.899622774124145,
      "epoch": 2.869876780564537,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004180291500510515,
      "loss": 5.544,
      "mean_token_accuracy": 0.18019194453954696,
      "num_tokens": 66550966.0,
      "step": 32840
    },
    {
      "entropy": 5.960386753082275,
      "epoch": 2.870313728917242,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041800509391753834,
      "loss": 5.6293,
      "mean_token_accuracy": 0.17268354147672654,
      "num_tokens": 66561550.0,
      "step": 32845
    },
    {
      "entropy": 5.930650758743286,
      "epoch": 2.870750677269947,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004179810350411125,
      "loss": 5.5115,
      "mean_token_accuracy": 0.18609691858291627,
      "num_tokens": 66571477.0,
      "step": 32850
    },
    {
      "entropy": 5.918006467819214,
      "epoch": 2.8711876256226514,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004179569734222355,
      "loss": 5.5755,
      "mean_token_accuracy": 0.1762360081076622,
      "num_tokens": 66581976.0,
      "step": 32855
    },
    {
      "entropy": 5.955997085571289,
      "epoch": 2.871624573975356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041793290906136886,
      "loss": 5.5211,
      "mean_token_accuracy": 0.17971102893352509,
      "num_tokens": 66591747.0,
      "step": 32860
    },
    {
      "entropy": 5.935006046295166,
      "epoch": 2.872061522328061,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004179088419589741,
      "loss": 5.5311,
      "mean_token_accuracy": 0.1800038307905197,
      "num_tokens": 66602681.0,
      "step": 32865
    },
    {
      "entropy": 5.900189542770386,
      "epoch": 2.8724984706807657,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041788477211551295,
      "loss": 5.5664,
      "mean_token_accuracy": 0.17922881841659546,
      "num_tokens": 66612887.0,
      "step": 32870
    },
    {
      "entropy": 5.900208806991577,
      "epoch": 2.87293541903347,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041786069953144697,
      "loss": 5.4997,
      "mean_token_accuracy": 0.18372149467468263,
      "num_tokens": 66622418.0,
      "step": 32875
    },
    {
      "entropy": 5.937957811355591,
      "epoch": 2.873372367386175,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041783662420723786,
      "loss": 5.6178,
      "mean_token_accuracy": 0.17625166922807695,
      "num_tokens": 66633457.0,
      "step": 32880
    },
    {
      "entropy": 5.947567987442016,
      "epoch": 2.8738093157388795,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004178125461433475,
      "loss": 5.5031,
      "mean_token_accuracy": 0.18339373022317887,
      "num_tokens": 66643691.0,
      "step": 32885
    },
    {
      "entropy": 5.873569679260254,
      "epoch": 2.8742462640915845,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041778846534023763,
      "loss": 5.5434,
      "mean_token_accuracy": 0.17964337021112442,
      "num_tokens": 66653745.0,
      "step": 32890
    },
    {
      "entropy": 5.898121547698975,
      "epoch": 2.874683212444289,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041776438179837026,
      "loss": 5.5308,
      "mean_token_accuracy": 0.18008908927440642,
      "num_tokens": 66664093.0,
      "step": 32895
    },
    {
      "entropy": 5.909351015090943,
      "epoch": 2.875120160796994,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004177402955182071,
      "loss": 5.5905,
      "mean_token_accuracy": 0.1806681901216507,
      "num_tokens": 66675204.0,
      "step": 32900
    },
    {
      "entropy": 5.969059848785401,
      "epoch": 2.8755571091496988,
      "grad_norm": 1.125,
      "learning_rate": 0.00041771620650021043,
      "loss": 5.5733,
      "mean_token_accuracy": 0.18174973130226135,
      "num_tokens": 66684789.0,
      "step": 32905
    },
    {
      "entropy": 5.964711761474609,
      "epoch": 2.8759940575024032,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041769211474484204,
      "loss": 5.6048,
      "mean_token_accuracy": 0.17559404373168946,
      "num_tokens": 66694734.0,
      "step": 32910
    },
    {
      "entropy": 5.945205545425415,
      "epoch": 2.8764310058551077,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004176680202525641,
      "loss": 5.5665,
      "mean_token_accuracy": 0.17275780886411668,
      "num_tokens": 66705231.0,
      "step": 32915
    },
    {
      "entropy": 5.903293228149414,
      "epoch": 2.8768679542078126,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041764392302383893,
      "loss": 5.5134,
      "mean_token_accuracy": 0.18076230883598327,
      "num_tokens": 66715627.0,
      "step": 32920
    },
    {
      "entropy": 5.893376731872559,
      "epoch": 2.8773049025605175,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004176198230591284,
      "loss": 5.5071,
      "mean_token_accuracy": 0.18181143552064896,
      "num_tokens": 66725525.0,
      "step": 32925
    },
    {
      "entropy": 5.889356088638306,
      "epoch": 2.877741850913222,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000417595720358895,
      "loss": 5.5491,
      "mean_token_accuracy": 0.1737162470817566,
      "num_tokens": 66736352.0,
      "step": 32930
    },
    {
      "entropy": 5.896110582351684,
      "epoch": 2.878178799265927,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00041757161492360096,
      "loss": 5.6055,
      "mean_token_accuracy": 0.1716202661395073,
      "num_tokens": 66745961.0,
      "step": 32935
    },
    {
      "entropy": 5.932370185852051,
      "epoch": 2.8786157476186314,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041754750675370865,
      "loss": 5.4725,
      "mean_token_accuracy": 0.18887826353311538,
      "num_tokens": 66755295.0,
      "step": 32940
    },
    {
      "entropy": 5.884123849868774,
      "epoch": 2.8790526959713363,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004175233958496804,
      "loss": 5.4974,
      "mean_token_accuracy": 0.1811988338828087,
      "num_tokens": 66765089.0,
      "step": 32945
    },
    {
      "entropy": 5.880622816085816,
      "epoch": 2.8794896443240408,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004174992822119788,
      "loss": 5.5403,
      "mean_token_accuracy": 0.1796663820743561,
      "num_tokens": 66774995.0,
      "step": 32950
    },
    {
      "entropy": 5.882524204254151,
      "epoch": 2.8799265926767457,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004174751658410663,
      "loss": 5.5405,
      "mean_token_accuracy": 0.18695630580186845,
      "num_tokens": 66784584.0,
      "step": 32955
    },
    {
      "entropy": 5.946340274810791,
      "epoch": 2.8803635410294506,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041745104673740545,
      "loss": 5.5427,
      "mean_token_accuracy": 0.18130811899900437,
      "num_tokens": 66794747.0,
      "step": 32960
    },
    {
      "entropy": 5.93714451789856,
      "epoch": 2.880800489382155,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041742692490145895,
      "loss": 5.527,
      "mean_token_accuracy": 0.18680267184972763,
      "num_tokens": 66804632.0,
      "step": 32965
    },
    {
      "entropy": 5.817714595794678,
      "epoch": 2.8812374377348595,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004174028003336893,
      "loss": 5.4801,
      "mean_token_accuracy": 0.18438446521759033,
      "num_tokens": 66814534.0,
      "step": 32970
    },
    {
      "entropy": 5.924748659133911,
      "epoch": 2.8816743860875644,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004173786730345594,
      "loss": 5.6077,
      "mean_token_accuracy": 0.17368056029081344,
      "num_tokens": 66824629.0,
      "step": 32975
    },
    {
      "entropy": 5.977549505233765,
      "epoch": 2.8821113344402693,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004173545430045319,
      "loss": 5.5814,
      "mean_token_accuracy": 0.1798764169216156,
      "num_tokens": 66834345.0,
      "step": 32980
    },
    {
      "entropy": 5.915374517440796,
      "epoch": 2.882548282792974,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041733041024406966,
      "loss": 5.5246,
      "mean_token_accuracy": 0.18292677253484727,
      "num_tokens": 66845272.0,
      "step": 32985
    },
    {
      "entropy": 5.957058763504028,
      "epoch": 2.8829852311456787,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004173062747536356,
      "loss": 5.5692,
      "mean_token_accuracy": 0.1797536477446556,
      "num_tokens": 66855022.0,
      "step": 32990
    },
    {
      "entropy": 5.939364576339722,
      "epoch": 2.883422179498383,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004172821365336926,
      "loss": 5.4837,
      "mean_token_accuracy": 0.18227959871292115,
      "num_tokens": 66864895.0,
      "step": 32995
    },
    {
      "entropy": 5.902257347106934,
      "epoch": 2.883859127851088,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004172579955847037,
      "loss": 5.5017,
      "mean_token_accuracy": 0.1856248587369919,
      "num_tokens": 66876134.0,
      "step": 33000
    },
    {
      "epoch": 2.883859127851088,
      "eval_entropy": 5.64924422226587,
      "eval_loss": 5.591622829437256,
      "eval_mean_token_accuracy": 0.1858150579779213,
      "eval_num_tokens": 66876134.0,
      "eval_runtime": 28.3367,
      "eval_samples_per_second": 1306.223,
      "eval_steps_per_second": 163.287,
      "step": 33000
    },
    {
      "entropy": 5.86556715965271,
      "epoch": 2.8842960762037926,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004172338519071318,
      "loss": 5.4974,
      "mean_token_accuracy": 0.1798661783337593,
      "num_tokens": 66886089.0,
      "step": 33005
    },
    {
      "entropy": 5.785141658782959,
      "epoch": 2.8847330245564975,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041720970550144024,
      "loss": 5.4147,
      "mean_token_accuracy": 0.198044154047966,
      "num_tokens": 66896072.0,
      "step": 33010
    },
    {
      "entropy": 5.953512477874756,
      "epoch": 2.8851699729092024,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041718555636809194,
      "loss": 5.6237,
      "mean_token_accuracy": 0.1761544018983841,
      "num_tokens": 66906030.0,
      "step": 33015
    },
    {
      "entropy": 5.936711215972901,
      "epoch": 2.885606921261907,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004171614045075501,
      "loss": 5.5065,
      "mean_token_accuracy": 0.17707391679286957,
      "num_tokens": 66916714.0,
      "step": 33020
    },
    {
      "entropy": 5.863673114776612,
      "epoch": 2.8860438696146113,
      "grad_norm": 1.375,
      "learning_rate": 0.0004171372499202781,
      "loss": 5.5567,
      "mean_token_accuracy": 0.17704673260450363,
      "num_tokens": 66926570.0,
      "step": 33025
    },
    {
      "entropy": 5.965664291381836,
      "epoch": 2.8864808179673163,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004171130926067391,
      "loss": 5.6606,
      "mean_token_accuracy": 0.1745353952050209,
      "num_tokens": 66937565.0,
      "step": 33030
    },
    {
      "entropy": 5.954148674011231,
      "epoch": 2.886917766320021,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004170889325673967,
      "loss": 5.6006,
      "mean_token_accuracy": 0.17932167053222656,
      "num_tokens": 66948363.0,
      "step": 33035
    },
    {
      "entropy": 5.872537565231323,
      "epoch": 2.8873547146727256,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041706476980271395,
      "loss": 5.5051,
      "mean_token_accuracy": 0.1866131067276001,
      "num_tokens": 66958312.0,
      "step": 33040
    },
    {
      "entropy": 5.91400785446167,
      "epoch": 2.8877916630254306,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041704060431315456,
      "loss": 5.4671,
      "mean_token_accuracy": 0.18696114420890808,
      "num_tokens": 66967880.0,
      "step": 33045
    },
    {
      "entropy": 5.91360650062561,
      "epoch": 2.888228611378135,
      "grad_norm": 1.125,
      "learning_rate": 0.00041701643609918187,
      "loss": 5.4448,
      "mean_token_accuracy": 0.18790665864944459,
      "num_tokens": 66977930.0,
      "step": 33050
    },
    {
      "entropy": 5.873953628540039,
      "epoch": 2.88866555973084,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004169922651612596,
      "loss": 5.6069,
      "mean_token_accuracy": 0.17336364984512329,
      "num_tokens": 66987780.0,
      "step": 33055
    },
    {
      "entropy": 5.955817794799804,
      "epoch": 2.8891025080835444,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041696809149985124,
      "loss": 5.5589,
      "mean_token_accuracy": 0.17685231566429138,
      "num_tokens": 66997910.0,
      "step": 33060
    },
    {
      "entropy": 5.929818916320801,
      "epoch": 2.8895394564362493,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041694391511542053,
      "loss": 5.4659,
      "mean_token_accuracy": 0.1900287836790085,
      "num_tokens": 67007365.0,
      "step": 33065
    },
    {
      "entropy": 5.776772546768188,
      "epoch": 2.889976404788954,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004169197360084311,
      "loss": 5.4381,
      "mean_token_accuracy": 0.1899375945329666,
      "num_tokens": 67016438.0,
      "step": 33070
    },
    {
      "entropy": 5.814446926116943,
      "epoch": 2.8904133531416587,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041689555417934675,
      "loss": 5.3631,
      "mean_token_accuracy": 0.20011244118213653,
      "num_tokens": 67026842.0,
      "step": 33075
    },
    {
      "entropy": 5.897689247131348,
      "epoch": 2.890850301494363,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041687136962863144,
      "loss": 5.584,
      "mean_token_accuracy": 0.17385654151439667,
      "num_tokens": 67037343.0,
      "step": 33080
    },
    {
      "entropy": 5.932361745834351,
      "epoch": 2.891287249847068,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041684718235674887,
      "loss": 5.515,
      "mean_token_accuracy": 0.18200369030237198,
      "num_tokens": 67047352.0,
      "step": 33085
    },
    {
      "entropy": 5.935082912445068,
      "epoch": 2.891724198199773,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004168229923641629,
      "loss": 5.5913,
      "mean_token_accuracy": 0.17052954137325288,
      "num_tokens": 67058717.0,
      "step": 33090
    },
    {
      "entropy": 5.9101542949676515,
      "epoch": 2.8921611465524775,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004167987996513378,
      "loss": 5.5208,
      "mean_token_accuracy": 0.18273056894540787,
      "num_tokens": 67068124.0,
      "step": 33095
    },
    {
      "entropy": 5.902303266525268,
      "epoch": 2.892598094905182,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004167746042187373,
      "loss": 5.5673,
      "mean_token_accuracy": 0.18057213872671127,
      "num_tokens": 67078906.0,
      "step": 33100
    },
    {
      "entropy": 5.793193531036377,
      "epoch": 2.893035043257887,
      "grad_norm": 1.40625,
      "learning_rate": 0.00041675040606682564,
      "loss": 5.513,
      "mean_token_accuracy": 0.1852891743183136,
      "num_tokens": 67090424.0,
      "step": 33105
    },
    {
      "entropy": 5.945244169235229,
      "epoch": 2.8934719916105918,
      "grad_norm": 1.0,
      "learning_rate": 0.00041672620519606687,
      "loss": 5.5297,
      "mean_token_accuracy": 0.18312589824199677,
      "num_tokens": 67100311.0,
      "step": 33110
    },
    {
      "entropy": 5.87488865852356,
      "epoch": 2.8939089399632962,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004167020016069253,
      "loss": 5.5453,
      "mean_token_accuracy": 0.17672574520111084,
      "num_tokens": 67109337.0,
      "step": 33115
    },
    {
      "entropy": 5.908036947250366,
      "epoch": 2.894345888316001,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041667779529986504,
      "loss": 5.5272,
      "mean_token_accuracy": 0.17569477260112762,
      "num_tokens": 67119795.0,
      "step": 33120
    },
    {
      "entropy": 5.904347705841064,
      "epoch": 2.8947828366687056,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041665358627535043,
      "loss": 5.5642,
      "mean_token_accuracy": 0.17552576959133148,
      "num_tokens": 67129444.0,
      "step": 33125
    },
    {
      "entropy": 5.890871143341064,
      "epoch": 2.8952197850214105,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004166293745338458,
      "loss": 5.5811,
      "mean_token_accuracy": 0.17987482249736786,
      "num_tokens": 67138868.0,
      "step": 33130
    },
    {
      "entropy": 5.953214025497436,
      "epoch": 2.895656733374115,
      "grad_norm": 1.25,
      "learning_rate": 0.0004166051600758156,
      "loss": 5.6071,
      "mean_token_accuracy": 0.18071295619010924,
      "num_tokens": 67149615.0,
      "step": 33135
    },
    {
      "entropy": 5.952824831008911,
      "epoch": 2.89609368172682,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041658094290172415,
      "loss": 5.5491,
      "mean_token_accuracy": 0.1833210438489914,
      "num_tokens": 67159745.0,
      "step": 33140
    },
    {
      "entropy": 5.928142976760864,
      "epoch": 2.896530630079525,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004165567230120362,
      "loss": 5.4825,
      "mean_token_accuracy": 0.18327043801546097,
      "num_tokens": 67169900.0,
      "step": 33145
    },
    {
      "entropy": 5.8530285358428955,
      "epoch": 2.8969675784322293,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004165325004072159,
      "loss": 5.5181,
      "mean_token_accuracy": 0.18827421814203263,
      "num_tokens": 67180752.0,
      "step": 33150
    },
    {
      "entropy": 5.945817661285401,
      "epoch": 2.8974045267849338,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041650827508772826,
      "loss": 5.6006,
      "mean_token_accuracy": 0.1749882757663727,
      "num_tokens": 67191466.0,
      "step": 33155
    },
    {
      "entropy": 5.91028733253479,
      "epoch": 2.8978414751376387,
      "grad_norm": 1.40625,
      "learning_rate": 0.00041648404705403765,
      "loss": 5.5213,
      "mean_token_accuracy": 0.17952819019556046,
      "num_tokens": 67202813.0,
      "step": 33160
    },
    {
      "entropy": 5.917185306549072,
      "epoch": 2.8982784234903436,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041645981630660894,
      "loss": 5.451,
      "mean_token_accuracy": 0.1857671096920967,
      "num_tokens": 67213686.0,
      "step": 33165
    },
    {
      "entropy": 5.904818344116211,
      "epoch": 2.898715371843048,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004164355828459068,
      "loss": 5.5402,
      "mean_token_accuracy": 0.17968382686376572,
      "num_tokens": 67224955.0,
      "step": 33170
    },
    {
      "entropy": 5.876313543319702,
      "epoch": 2.899152320195753,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000416411346672396,
      "loss": 5.4948,
      "mean_token_accuracy": 0.18565340638160704,
      "num_tokens": 67234438.0,
      "step": 33175
    },
    {
      "entropy": 5.935800933837891,
      "epoch": 2.8995892685484574,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041638710778654157,
      "loss": 5.5727,
      "mean_token_accuracy": 0.17665264755487442,
      "num_tokens": 67245151.0,
      "step": 33180
    },
    {
      "entropy": 5.910060739517212,
      "epoch": 2.9000262169011624,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004163628661888083,
      "loss": 5.4993,
      "mean_token_accuracy": 0.18599077612161635,
      "num_tokens": 67254942.0,
      "step": 33185
    },
    {
      "entropy": 5.950627851486206,
      "epoch": 2.900463165253867,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041633862187966125,
      "loss": 5.5407,
      "mean_token_accuracy": 0.18091760128736495,
      "num_tokens": 67264435.0,
      "step": 33190
    },
    {
      "entropy": 5.866166591644287,
      "epoch": 2.9009001136065717,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004163143748595653,
      "loss": 5.4836,
      "mean_token_accuracy": 0.18276394307613372,
      "num_tokens": 67273304.0,
      "step": 33195
    },
    {
      "entropy": 5.862940073013306,
      "epoch": 2.9013370619592767,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041629012512898556,
      "loss": 5.5933,
      "mean_token_accuracy": 0.1765421912074089,
      "num_tokens": 67283713.0,
      "step": 33200
    },
    {
      "entropy": 5.878285837173462,
      "epoch": 2.901774010311981,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004162658726883873,
      "loss": 5.4146,
      "mean_token_accuracy": 0.19351580142974853,
      "num_tokens": 67292420.0,
      "step": 33205
    },
    {
      "entropy": 5.887427234649659,
      "epoch": 2.9022109586646856,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004162416175382354,
      "loss": 5.4983,
      "mean_token_accuracy": 0.18642055839300156,
      "num_tokens": 67302138.0,
      "step": 33210
    },
    {
      "entropy": 5.854867458343506,
      "epoch": 2.9026479070173905,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004162173596789954,
      "loss": 5.5104,
      "mean_token_accuracy": 0.185054512321949,
      "num_tokens": 67311776.0,
      "step": 33215
    },
    {
      "entropy": 5.9270768642425535,
      "epoch": 2.9030848553700954,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004161930991111325,
      "loss": 5.5616,
      "mean_token_accuracy": 0.17697660326957704,
      "num_tokens": 67322296.0,
      "step": 33220
    },
    {
      "entropy": 5.905867671966552,
      "epoch": 2.9035218037228,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004161688358351118,
      "loss": 5.4941,
      "mean_token_accuracy": 0.18128438144922257,
      "num_tokens": 67332173.0,
      "step": 33225
    },
    {
      "entropy": 5.869507265090943,
      "epoch": 2.903958752075505,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000416144569851399,
      "loss": 5.5197,
      "mean_token_accuracy": 0.18188222348690034,
      "num_tokens": 67342156.0,
      "step": 33230
    },
    {
      "entropy": 5.892373085021973,
      "epoch": 2.9043957004282093,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041612030116045933,
      "loss": 5.5575,
      "mean_token_accuracy": 0.18255189806222916,
      "num_tokens": 67353208.0,
      "step": 33235
    },
    {
      "entropy": 5.903980159759522,
      "epoch": 2.904832648780914,
      "grad_norm": 1.125,
      "learning_rate": 0.00041609602976275836,
      "loss": 5.4701,
      "mean_token_accuracy": 0.18946123123168945,
      "num_tokens": 67362577.0,
      "step": 33240
    },
    {
      "entropy": 5.861398792266845,
      "epoch": 2.9052695971336187,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004160717556587616,
      "loss": 5.4756,
      "mean_token_accuracy": 0.1896316960453987,
      "num_tokens": 67371996.0,
      "step": 33245
    },
    {
      "entropy": 5.849746513366699,
      "epoch": 2.9057065454863236,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004160474788489346,
      "loss": 5.4608,
      "mean_token_accuracy": 0.19010675251483916,
      "num_tokens": 67381714.0,
      "step": 33250
    },
    {
      "entropy": 5.8894720554351805,
      "epoch": 2.9061434938390285,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041602319933374305,
      "loss": 5.4971,
      "mean_token_accuracy": 0.18560765385627748,
      "num_tokens": 67391551.0,
      "step": 33255
    },
    {
      "entropy": 5.895506381988525,
      "epoch": 2.906580442191733,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004159989171136527,
      "loss": 5.5138,
      "mean_token_accuracy": 0.17657217085361482,
      "num_tokens": 67400816.0,
      "step": 33260
    },
    {
      "entropy": 5.88080096244812,
      "epoch": 2.9070173905444374,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004159746321891292,
      "loss": 5.5102,
      "mean_token_accuracy": 0.18109063059091568,
      "num_tokens": 67410633.0,
      "step": 33265
    },
    {
      "entropy": 5.898853492736817,
      "epoch": 2.9074543388971423,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004159503445606385,
      "loss": 5.5182,
      "mean_token_accuracy": 0.1876624420285225,
      "num_tokens": 67420588.0,
      "step": 33270
    },
    {
      "entropy": 5.874871921539307,
      "epoch": 2.9078912872498472,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041592605422864615,
      "loss": 5.537,
      "mean_token_accuracy": 0.18647703677415847,
      "num_tokens": 67430494.0,
      "step": 33275
    },
    {
      "entropy": 5.844671154022217,
      "epoch": 2.9083282356025517,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004159017611936183,
      "loss": 5.4362,
      "mean_token_accuracy": 0.19249300360679628,
      "num_tokens": 67440692.0,
      "step": 33280
    },
    {
      "entropy": 5.901129198074341,
      "epoch": 2.9087651839552566,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041587746545602094,
      "loss": 5.514,
      "mean_token_accuracy": 0.18333183377981185,
      "num_tokens": 67450936.0,
      "step": 33285
    },
    {
      "entropy": 5.889043521881104,
      "epoch": 2.909202132307961,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004158531670163199,
      "loss": 5.5118,
      "mean_token_accuracy": 0.18692963421344758,
      "num_tokens": 67460997.0,
      "step": 33290
    },
    {
      "entropy": 5.87371187210083,
      "epoch": 2.909639080660666,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004158288658749813,
      "loss": 5.5501,
      "mean_token_accuracy": 0.18647727370262146,
      "num_tokens": 67470685.0,
      "step": 33295
    },
    {
      "entropy": 5.889357614517212,
      "epoch": 2.9100760290133705,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041580456203247127,
      "loss": 5.5087,
      "mean_token_accuracy": 0.1835425615310669,
      "num_tokens": 67481734.0,
      "step": 33300
    },
    {
      "entropy": 5.766331338882447,
      "epoch": 2.9105129773660754,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000415780255489256,
      "loss": 5.3036,
      "mean_token_accuracy": 0.20895959436893463,
      "num_tokens": 67491290.0,
      "step": 33305
    },
    {
      "entropy": 5.877029466629028,
      "epoch": 2.9109499257187803,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004157559462458016,
      "loss": 5.5772,
      "mean_token_accuracy": 0.18043473213911057,
      "num_tokens": 67501758.0,
      "step": 33310
    },
    {
      "entropy": 5.928523778915405,
      "epoch": 2.9113868740714848,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041573163430257435,
      "loss": 5.5351,
      "mean_token_accuracy": 0.18340069502592088,
      "num_tokens": 67511435.0,
      "step": 33315
    },
    {
      "entropy": 5.923321962356567,
      "epoch": 2.9118238224241892,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041570731966004065,
      "loss": 5.5661,
      "mean_token_accuracy": 0.1762152835726738,
      "num_tokens": 67521408.0,
      "step": 33320
    },
    {
      "entropy": 5.932860565185547,
      "epoch": 2.912260770776894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004156830023186669,
      "loss": 5.5893,
      "mean_token_accuracy": 0.17863551378250123,
      "num_tokens": 67531631.0,
      "step": 33325
    },
    {
      "entropy": 5.964561939239502,
      "epoch": 2.912697719129599,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004156586822789194,
      "loss": 5.564,
      "mean_token_accuracy": 0.16883697658777236,
      "num_tokens": 67542265.0,
      "step": 33330
    },
    {
      "entropy": 5.918488264083862,
      "epoch": 2.9131346674823035,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041563435954126467,
      "loss": 5.5571,
      "mean_token_accuracy": 0.1775955319404602,
      "num_tokens": 67553664.0,
      "step": 33335
    },
    {
      "entropy": 5.874403095245361,
      "epoch": 2.9135716158350085,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004156100341061693,
      "loss": 5.4792,
      "mean_token_accuracy": 0.18215263038873672,
      "num_tokens": 67564500.0,
      "step": 33340
    },
    {
      "entropy": 5.8947680473327635,
      "epoch": 2.914008564187713,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004155857059740997,
      "loss": 5.4535,
      "mean_token_accuracy": 0.19325597882270812,
      "num_tokens": 67573611.0,
      "step": 33345
    },
    {
      "entropy": 5.855216121673584,
      "epoch": 2.914445512540418,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004155613751455226,
      "loss": 5.5007,
      "mean_token_accuracy": 0.18821567595005034,
      "num_tokens": 67583721.0,
      "step": 33350
    },
    {
      "entropy": 5.896247386932373,
      "epoch": 2.9148824608931223,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041553704162090475,
      "loss": 5.5535,
      "mean_token_accuracy": 0.18302929252386094,
      "num_tokens": 67593381.0,
      "step": 33355
    },
    {
      "entropy": 5.913576650619507,
      "epoch": 2.915319409245827,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004155127054007127,
      "loss": 5.558,
      "mean_token_accuracy": 0.17788611352443695,
      "num_tokens": 67603799.0,
      "step": 33360
    },
    {
      "entropy": 5.929439353942871,
      "epoch": 2.9157563575985317,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041548836648541346,
      "loss": 5.5468,
      "mean_token_accuracy": 0.1799096941947937,
      "num_tokens": 67613543.0,
      "step": 33365
    },
    {
      "entropy": 5.922424173355102,
      "epoch": 2.9161933059512366,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041546402487547363,
      "loss": 5.5558,
      "mean_token_accuracy": 0.18155300468206406,
      "num_tokens": 67623302.0,
      "step": 33370
    },
    {
      "entropy": 5.9173903465271,
      "epoch": 2.916630254303941,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004154396805713602,
      "loss": 5.4416,
      "mean_token_accuracy": 0.1925421640276909,
      "num_tokens": 67633655.0,
      "step": 33375
    },
    {
      "entropy": 5.864323282241822,
      "epoch": 2.917067202656646,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041541533357354013,
      "loss": 5.5641,
      "mean_token_accuracy": 0.18241755813360214,
      "num_tokens": 67644832.0,
      "step": 33380
    },
    {
      "entropy": 5.932661008834839,
      "epoch": 2.917504151009351,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004153909838824804,
      "loss": 5.5201,
      "mean_token_accuracy": 0.18276529312133788,
      "num_tokens": 67655290.0,
      "step": 33385
    },
    {
      "entropy": 5.893868017196655,
      "epoch": 2.9179410993620554,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000415366631498648,
      "loss": 5.4619,
      "mean_token_accuracy": 0.19023314714431763,
      "num_tokens": 67664252.0,
      "step": 33390
    },
    {
      "entropy": 5.875927591323853,
      "epoch": 2.9183780477147603,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004153422764225101,
      "loss": 5.5206,
      "mean_token_accuracy": 0.18378852009773256,
      "num_tokens": 67674655.0,
      "step": 33395
    },
    {
      "entropy": 5.887672281265258,
      "epoch": 2.9188149960674648,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004153179186545338,
      "loss": 5.4485,
      "mean_token_accuracy": 0.18613499253988267,
      "num_tokens": 67685126.0,
      "step": 33400
    },
    {
      "entropy": 5.77366681098938,
      "epoch": 2.9192519444201697,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041529355819518626,
      "loss": 5.4439,
      "mean_token_accuracy": 0.1868752583861351,
      "num_tokens": 67694111.0,
      "step": 33405
    },
    {
      "entropy": 5.912567472457885,
      "epoch": 2.919688892772874,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041526919504493475,
      "loss": 5.5391,
      "mean_token_accuracy": 0.1817198008298874,
      "num_tokens": 67704229.0,
      "step": 33410
    },
    {
      "entropy": 5.9228941917419435,
      "epoch": 2.920125841125579,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041524482920424666,
      "loss": 5.5425,
      "mean_token_accuracy": 0.18537815809249877,
      "num_tokens": 67714625.0,
      "step": 33415
    },
    {
      "entropy": 5.870808696746826,
      "epoch": 2.9205627894782835,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041522046067358917,
      "loss": 5.5214,
      "mean_token_accuracy": 0.17564816176891326,
      "num_tokens": 67724094.0,
      "step": 33420
    },
    {
      "entropy": 5.860282564163208,
      "epoch": 2.9209997378309884,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041519608945342985,
      "loss": 5.5168,
      "mean_token_accuracy": 0.183514603972435,
      "num_tokens": 67733233.0,
      "step": 33425
    },
    {
      "entropy": 5.868614530563354,
      "epoch": 2.921436686183693,
      "grad_norm": 1.078125,
      "learning_rate": 0.000415171715544236,
      "loss": 5.4612,
      "mean_token_accuracy": 0.18860098272562026,
      "num_tokens": 67742781.0,
      "step": 33430
    },
    {
      "entropy": 5.892099475860595,
      "epoch": 2.921873634536398,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041514733894647526,
      "loss": 5.4978,
      "mean_token_accuracy": 0.183912792801857,
      "num_tokens": 67753166.0,
      "step": 33435
    },
    {
      "entropy": 5.900177049636841,
      "epoch": 2.9223105828891027,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004151229596606151,
      "loss": 5.614,
      "mean_token_accuracy": 0.18304163813591004,
      "num_tokens": 67762599.0,
      "step": 33440
    },
    {
      "entropy": 5.859924602508545,
      "epoch": 2.922747531241807,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004150985776871231,
      "loss": 5.511,
      "mean_token_accuracy": 0.18551765382289886,
      "num_tokens": 67772620.0,
      "step": 33445
    },
    {
      "entropy": 5.987096548080444,
      "epoch": 2.9231844795945117,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000415074193026467,
      "loss": 5.6016,
      "mean_token_accuracy": 0.16967864483594894,
      "num_tokens": 67782962.0,
      "step": 33450
    },
    {
      "entropy": 5.913297462463379,
      "epoch": 2.9236214279472166,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041504980567911453,
      "loss": 5.5617,
      "mean_token_accuracy": 0.17768621146678926,
      "num_tokens": 67793168.0,
      "step": 33455
    },
    {
      "entropy": 5.906870985031128,
      "epoch": 2.9240583762999215,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041502541564553336,
      "loss": 5.561,
      "mean_token_accuracy": 0.17994209975004197,
      "num_tokens": 67803317.0,
      "step": 33460
    },
    {
      "entropy": 5.87271032333374,
      "epoch": 2.924495324652626,
      "grad_norm": 1.125,
      "learning_rate": 0.00041500102292619135,
      "loss": 5.4989,
      "mean_token_accuracy": 0.17907123118638993,
      "num_tokens": 67812594.0,
      "step": 33465
    },
    {
      "entropy": 5.871486711502075,
      "epoch": 2.924932273005331,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041497662752155637,
      "loss": 5.4727,
      "mean_token_accuracy": 0.18468204140663147,
      "num_tokens": 67822277.0,
      "step": 33470
    },
    {
      "entropy": 5.96222152709961,
      "epoch": 2.9253692213580353,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004149522294320963,
      "loss": 5.5462,
      "mean_token_accuracy": 0.17720680236816405,
      "num_tokens": 67832948.0,
      "step": 33475
    },
    {
      "entropy": 5.901156663894653,
      "epoch": 2.9258061697107403,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041492782865827914,
      "loss": 5.4607,
      "mean_token_accuracy": 0.18861417323350907,
      "num_tokens": 67843965.0,
      "step": 33480
    },
    {
      "entropy": 5.8053327083587645,
      "epoch": 2.9262431180634447,
      "grad_norm": 1.125,
      "learning_rate": 0.00041490342520057293,
      "loss": 5.5526,
      "mean_token_accuracy": 0.18480799794197084,
      "num_tokens": 67854079.0,
      "step": 33485
    },
    {
      "entropy": 5.920548772811889,
      "epoch": 2.9266800664161496,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041487901905944575,
      "loss": 5.5133,
      "mean_token_accuracy": 0.18423262238502502,
      "num_tokens": 67865218.0,
      "step": 33490
    },
    {
      "entropy": 5.9534119129180905,
      "epoch": 2.9271170147688546,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041485461023536564,
      "loss": 5.4798,
      "mean_token_accuracy": 0.19077782183885575,
      "num_tokens": 67875229.0,
      "step": 33495
    },
    {
      "entropy": 5.865934991836548,
      "epoch": 2.927553963121559,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004148301987288008,
      "loss": 5.5918,
      "mean_token_accuracy": 0.18223651647567748,
      "num_tokens": 67885795.0,
      "step": 33500
    },
    {
      "entropy": 5.861395645141601,
      "epoch": 2.9279909114742635,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004148057845402195,
      "loss": 5.4935,
      "mean_token_accuracy": 0.18686310052871705,
      "num_tokens": 67896668.0,
      "step": 33505
    },
    {
      "entropy": 5.853166723251343,
      "epoch": 2.9284278598269684,
      "grad_norm": 1.0,
      "learning_rate": 0.00041478136767009004,
      "loss": 5.4645,
      "mean_token_accuracy": 0.191537743806839,
      "num_tokens": 67907082.0,
      "step": 33510
    },
    {
      "entropy": 5.916761589050293,
      "epoch": 2.9288648081796733,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004147569481188806,
      "loss": 5.5527,
      "mean_token_accuracy": 0.1771400138735771,
      "num_tokens": 67917517.0,
      "step": 33515
    },
    {
      "entropy": 5.982824468612671,
      "epoch": 2.929301756532378,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041473252588705977,
      "loss": 5.5881,
      "mean_token_accuracy": 0.17686403244733812,
      "num_tokens": 67927332.0,
      "step": 33520
    },
    {
      "entropy": 5.940986013412475,
      "epoch": 2.9297387048850827,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041470810097509587,
      "loss": 5.5355,
      "mean_token_accuracy": 0.17937404811382293,
      "num_tokens": 67936939.0,
      "step": 33525
    },
    {
      "entropy": 5.85835108757019,
      "epoch": 2.930175653237787,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004146836733834573,
      "loss": 5.5298,
      "mean_token_accuracy": 0.18462015688419342,
      "num_tokens": 67947553.0,
      "step": 33530
    },
    {
      "entropy": 5.894081258773804,
      "epoch": 2.930612601590492,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004146592431126128,
      "loss": 5.4815,
      "mean_token_accuracy": 0.18352937400341035,
      "num_tokens": 67958489.0,
      "step": 33535
    },
    {
      "entropy": 5.8478474617004395,
      "epoch": 2.9310495499431966,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004146348101630307,
      "loss": 5.4478,
      "mean_token_accuracy": 0.18435898125171662,
      "num_tokens": 67967305.0,
      "step": 33540
    },
    {
      "entropy": 5.829212236404419,
      "epoch": 2.9314864982959015,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004146103745351799,
      "loss": 5.4839,
      "mean_token_accuracy": 0.18261103630065917,
      "num_tokens": 67977400.0,
      "step": 33545
    },
    {
      "entropy": 5.865965747833252,
      "epoch": 2.9319234466486064,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004145859362295289,
      "loss": 5.4912,
      "mean_token_accuracy": 0.18123923540115355,
      "num_tokens": 67987275.0,
      "step": 33550
    },
    {
      "entropy": 6.022046041488648,
      "epoch": 2.932360395001311,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004145614952465465,
      "loss": 5.702,
      "mean_token_accuracy": 0.1691983923316002,
      "num_tokens": 67999002.0,
      "step": 33555
    },
    {
      "entropy": 5.947755765914917,
      "epoch": 2.9327973433540153,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041453705158670147,
      "loss": 5.474,
      "mean_token_accuracy": 0.1935196340084076,
      "num_tokens": 68008504.0,
      "step": 33560
    },
    {
      "entropy": 5.884036874771118,
      "epoch": 2.9332342917067202,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041451260525046267,
      "loss": 5.5052,
      "mean_token_accuracy": 0.18326316475868226,
      "num_tokens": 68019522.0,
      "step": 33565
    },
    {
      "entropy": 5.869736242294311,
      "epoch": 2.933671240059425,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041448815623829907,
      "loss": 5.5242,
      "mean_token_accuracy": 0.1785072848200798,
      "num_tokens": 68029991.0,
      "step": 33570
    },
    {
      "entropy": 5.882711029052734,
      "epoch": 2.9341081884121296,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004144637045506795,
      "loss": 5.5537,
      "mean_token_accuracy": 0.17689082026481628,
      "num_tokens": 68040347.0,
      "step": 33575
    },
    {
      "entropy": 5.92022967338562,
      "epoch": 2.9345451367648345,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000414439250188073,
      "loss": 5.499,
      "mean_token_accuracy": 0.18971046656370164,
      "num_tokens": 68050932.0,
      "step": 33580
    },
    {
      "entropy": 5.952276802062988,
      "epoch": 2.934982085117539,
      "grad_norm": 1.0,
      "learning_rate": 0.0004144147931509486,
      "loss": 5.5283,
      "mean_token_accuracy": 0.1808869794011116,
      "num_tokens": 68061104.0,
      "step": 33585
    },
    {
      "entropy": 5.852302694320679,
      "epoch": 2.935419033470244,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041439033343977546,
      "loss": 5.4863,
      "mean_token_accuracy": 0.18833506405353545,
      "num_tokens": 68070374.0,
      "step": 33590
    },
    {
      "entropy": 5.85526294708252,
      "epoch": 2.9358559818229484,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041436587105502263,
      "loss": 5.5308,
      "mean_token_accuracy": 0.17975387573242188,
      "num_tokens": 68079996.0,
      "step": 33595
    },
    {
      "entropy": 5.895689010620117,
      "epoch": 2.9362929301756533,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004143414059971594,
      "loss": 5.5617,
      "mean_token_accuracy": 0.18129540830850602,
      "num_tokens": 68090322.0,
      "step": 33600
    },
    {
      "entropy": 5.8310462474823,
      "epoch": 2.936729878528358,
      "grad_norm": 1.171875,
      "learning_rate": 0.000414316938266655,
      "loss": 5.4046,
      "mean_token_accuracy": 0.18639412373304368,
      "num_tokens": 68099482.0,
      "step": 33605
    },
    {
      "entropy": 5.826352167129516,
      "epoch": 2.9371668268810627,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004142924678639787,
      "loss": 5.5082,
      "mean_token_accuracy": 0.18876493722200394,
      "num_tokens": 68109852.0,
      "step": 33610
    },
    {
      "entropy": 5.887367820739746,
      "epoch": 2.937603775233767,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041426799478959993,
      "loss": 5.5234,
      "mean_token_accuracy": 0.18662674874067306,
      "num_tokens": 68118721.0,
      "step": 33615
    },
    {
      "entropy": 5.939733934402466,
      "epoch": 2.938040723586472,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004142435190439879,
      "loss": 5.4762,
      "mean_token_accuracy": 0.18340552300214769,
      "num_tokens": 68129080.0,
      "step": 33620
    },
    {
      "entropy": 5.898727703094482,
      "epoch": 2.938477671939177,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041421904062761234,
      "loss": 5.52,
      "mean_token_accuracy": 0.18444158881902695,
      "num_tokens": 68138154.0,
      "step": 33625
    },
    {
      "entropy": 5.879382038116455,
      "epoch": 2.9389146202918814,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004141945595409426,
      "loss": 5.5972,
      "mean_token_accuracy": 0.17857316583395005,
      "num_tokens": 68148736.0,
      "step": 33630
    },
    {
      "entropy": 5.880358362197876,
      "epoch": 2.9393515686445864,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004141700757844483,
      "loss": 5.477,
      "mean_token_accuracy": 0.18171253204345703,
      "num_tokens": 68159041.0,
      "step": 33635
    },
    {
      "entropy": 5.904904747009278,
      "epoch": 2.939788516997291,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000414145589358599,
      "loss": 5.5031,
      "mean_token_accuracy": 0.1847042217850685,
      "num_tokens": 68168715.0,
      "step": 33640
    },
    {
      "entropy": 5.859090662002563,
      "epoch": 2.9402254653499957,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041412110026386436,
      "loss": 5.5291,
      "mean_token_accuracy": 0.18704067617654802,
      "num_tokens": 68178863.0,
      "step": 33645
    },
    {
      "entropy": 5.951573228836059,
      "epoch": 2.9406624137027,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004140966085007142,
      "loss": 5.5649,
      "mean_token_accuracy": 0.1784139856696129,
      "num_tokens": 68190131.0,
      "step": 33650
    },
    {
      "entropy": 5.882849836349488,
      "epoch": 2.941099362055405,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041407211406961815,
      "loss": 5.4872,
      "mean_token_accuracy": 0.18662316501140594,
      "num_tokens": 68200193.0,
      "step": 33655
    },
    {
      "entropy": 5.882511568069458,
      "epoch": 2.94153631040811,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041404761697104605,
      "loss": 5.4855,
      "mean_token_accuracy": 0.18276171237230301,
      "num_tokens": 68210714.0,
      "step": 33660
    },
    {
      "entropy": 5.878807735443115,
      "epoch": 2.9419732587608145,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004140231172054678,
      "loss": 5.5347,
      "mean_token_accuracy": 0.18009101152420043,
      "num_tokens": 68220809.0,
      "step": 33665
    },
    {
      "entropy": 5.911371564865112,
      "epoch": 2.942410207113519,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004139986147733534,
      "loss": 5.5278,
      "mean_token_accuracy": 0.18768562525510787,
      "num_tokens": 68229887.0,
      "step": 33670
    },
    {
      "entropy": 5.9026552677154545,
      "epoch": 2.942847155466224,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041397410967517265,
      "loss": 5.504,
      "mean_token_accuracy": 0.18385957628488542,
      "num_tokens": 68240491.0,
      "step": 33675
    },
    {
      "entropy": 5.870474433898925,
      "epoch": 2.943284103818929,
      "grad_norm": 1.125,
      "learning_rate": 0.00041394960191139566,
      "loss": 5.5956,
      "mean_token_accuracy": 0.18402131497859955,
      "num_tokens": 68250302.0,
      "step": 33680
    },
    {
      "entropy": 5.81494550704956,
      "epoch": 2.9437210521716333,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004139250914824925,
      "loss": 5.4143,
      "mean_token_accuracy": 0.1894748792052269,
      "num_tokens": 68259245.0,
      "step": 33685
    },
    {
      "entropy": 5.8681317329406735,
      "epoch": 2.944158000524338,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041390057838893335,
      "loss": 5.4594,
      "mean_token_accuracy": 0.19070049822330476,
      "num_tokens": 68269096.0,
      "step": 33690
    },
    {
      "entropy": 5.910536241531372,
      "epoch": 2.9445949488770427,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041387606263118825,
      "loss": 5.5341,
      "mean_token_accuracy": 0.18408199846744538,
      "num_tokens": 68278949.0,
      "step": 33695
    },
    {
      "entropy": 5.839140558242798,
      "epoch": 2.9450318972297476,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004138515442097275,
      "loss": 5.4526,
      "mean_token_accuracy": 0.18589754700660704,
      "num_tokens": 68288526.0,
      "step": 33700
    },
    {
      "entropy": 5.89796404838562,
      "epoch": 2.945468845582452,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004138270231250214,
      "loss": 5.4357,
      "mean_token_accuracy": 0.18863013237714768,
      "num_tokens": 68297997.0,
      "step": 33705
    },
    {
      "entropy": 5.834344911575317,
      "epoch": 2.945905793935157,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004138024993775403,
      "loss": 5.55,
      "mean_token_accuracy": 0.1832453191280365,
      "num_tokens": 68307316.0,
      "step": 33710
    },
    {
      "entropy": 5.898731136322022,
      "epoch": 2.9463427422878614,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004137779729677544,
      "loss": 5.3745,
      "mean_token_accuracy": 0.19445831328630447,
      "num_tokens": 68317124.0,
      "step": 33715
    },
    {
      "entropy": 5.832303524017334,
      "epoch": 2.9467796906405663,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004137534438961344,
      "loss": 5.5186,
      "mean_token_accuracy": 0.18706717640161513,
      "num_tokens": 68326576.0,
      "step": 33720
    },
    {
      "entropy": 5.8810125350952145,
      "epoch": 2.947216638993271,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041372891216315055,
      "loss": 5.474,
      "mean_token_accuracy": 0.19312630891799926,
      "num_tokens": 68336999.0,
      "step": 33725
    },
    {
      "entropy": 5.960949277877807,
      "epoch": 2.9476535873459757,
      "grad_norm": 1.0,
      "learning_rate": 0.00041370437776927347,
      "loss": 5.5026,
      "mean_token_accuracy": 0.18535166680812837,
      "num_tokens": 68347084.0,
      "step": 33730
    },
    {
      "entropy": 5.814162254333496,
      "epoch": 2.9480905356986806,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004136798407149738,
      "loss": 5.4844,
      "mean_token_accuracy": 0.1837248370051384,
      "num_tokens": 68357136.0,
      "step": 33735
    },
    {
      "entropy": 5.9049060344696045,
      "epoch": 2.948527484051385,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041365530100072215,
      "loss": 5.5785,
      "mean_token_accuracy": 0.173912650346756,
      "num_tokens": 68367020.0,
      "step": 33740
    },
    {
      "entropy": 5.996183824539185,
      "epoch": 2.9489644324040896,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004136307586269891,
      "loss": 5.4944,
      "mean_token_accuracy": 0.18694276213645936,
      "num_tokens": 68378436.0,
      "step": 33745
    },
    {
      "entropy": 5.880055379867554,
      "epoch": 2.9494013807567945,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041360621359424544,
      "loss": 5.4556,
      "mean_token_accuracy": 0.1833631858229637,
      "num_tokens": 68387777.0,
      "step": 33750
    },
    {
      "entropy": 5.791916704177856,
      "epoch": 2.9498383291094994,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041358166590296205,
      "loss": 5.4877,
      "mean_token_accuracy": 0.1835209459066391,
      "num_tokens": 68397648.0,
      "step": 33755
    },
    {
      "entropy": 5.952750825881958,
      "epoch": 2.950275277462204,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041355711555360965,
      "loss": 5.6345,
      "mean_token_accuracy": 0.17583155035972595,
      "num_tokens": 68409539.0,
      "step": 33760
    },
    {
      "entropy": 6.0000263214111325,
      "epoch": 2.9507122258149088,
      "grad_norm": 1.25,
      "learning_rate": 0.00041353256254665914,
      "loss": 5.5632,
      "mean_token_accuracy": 0.17954799830913543,
      "num_tokens": 68419626.0,
      "step": 33765
    },
    {
      "entropy": 5.878669261932373,
      "epoch": 2.9511491741676132,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041350800688258157,
      "loss": 5.4912,
      "mean_token_accuracy": 0.1911418154835701,
      "num_tokens": 68429289.0,
      "step": 33770
    },
    {
      "entropy": 5.892546272277832,
      "epoch": 2.951586122520318,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041348344856184785,
      "loss": 5.5642,
      "mean_token_accuracy": 0.18561024516820906,
      "num_tokens": 68440186.0,
      "step": 33775
    },
    {
      "entropy": 5.924119853973389,
      "epoch": 2.9520230708730226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041345888758492896,
      "loss": 5.4526,
      "mean_token_accuracy": 0.19092100262641906,
      "num_tokens": 68450292.0,
      "step": 33780
    },
    {
      "entropy": 5.909563255310059,
      "epoch": 2.9524600192257275,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041343432395229605,
      "loss": 5.5236,
      "mean_token_accuracy": 0.18000029325485228,
      "num_tokens": 68460419.0,
      "step": 33785
    },
    {
      "entropy": 5.862645101547241,
      "epoch": 2.9528969675784325,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004134097576644203,
      "loss": 5.4801,
      "mean_token_accuracy": 0.18572704046964644,
      "num_tokens": 68470488.0,
      "step": 33790
    },
    {
      "entropy": 5.88679780960083,
      "epoch": 2.953333915931137,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041338518872177286,
      "loss": 5.5029,
      "mean_token_accuracy": 0.18843043148517608,
      "num_tokens": 68479872.0,
      "step": 33795
    },
    {
      "entropy": 5.872550630569458,
      "epoch": 2.9537708642838414,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041336061712482497,
      "loss": 5.5441,
      "mean_token_accuracy": 0.17869024574756623,
      "num_tokens": 68491086.0,
      "step": 33800
    },
    {
      "entropy": 5.968395090103149,
      "epoch": 2.9542078126365463,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041333604287404805,
      "loss": 5.5995,
      "mean_token_accuracy": 0.17044871747493745,
      "num_tokens": 68501136.0,
      "step": 33805
    },
    {
      "entropy": 5.856628847122193,
      "epoch": 2.954644760989251,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004133114659699132,
      "loss": 5.4596,
      "mean_token_accuracy": 0.1882012277841568,
      "num_tokens": 68511579.0,
      "step": 33810
    },
    {
      "entropy": 5.878273582458496,
      "epoch": 2.9550817093419557,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041328688641289205,
      "loss": 5.4756,
      "mean_token_accuracy": 0.19038270711898803,
      "num_tokens": 68521716.0,
      "step": 33815
    },
    {
      "entropy": 5.871596002578736,
      "epoch": 2.9555186576946606,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041326230420345584,
      "loss": 5.5327,
      "mean_token_accuracy": 0.18604498654603957,
      "num_tokens": 68531814.0,
      "step": 33820
    },
    {
      "entropy": 5.83560585975647,
      "epoch": 2.955955606047365,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004132377193420763,
      "loss": 5.4171,
      "mean_token_accuracy": 0.1937512308359146,
      "num_tokens": 68541388.0,
      "step": 33825
    },
    {
      "entropy": 5.921309757232666,
      "epoch": 2.95639255440007,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041321313182922484,
      "loss": 5.6266,
      "mean_token_accuracy": 0.17518227621912957,
      "num_tokens": 68552259.0,
      "step": 33830
    },
    {
      "entropy": 5.9307914733886715,
      "epoch": 2.9568295027527745,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004131885416653731,
      "loss": 5.5056,
      "mean_token_accuracy": 0.187727190554142,
      "num_tokens": 68562001.0,
      "step": 33835
    },
    {
      "entropy": 5.953164386749267,
      "epoch": 2.9572664511054794,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004131639488509927,
      "loss": 5.4992,
      "mean_token_accuracy": 0.18477512747049332,
      "num_tokens": 68571636.0,
      "step": 33840
    },
    {
      "entropy": 5.914348220825195,
      "epoch": 2.9577033994581843,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004131393533865554,
      "loss": 5.6157,
      "mean_token_accuracy": 0.1759214922785759,
      "num_tokens": 68581448.0,
      "step": 33845
    },
    {
      "entropy": 5.986692285537719,
      "epoch": 2.9581403478108887,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041311475527253284,
      "loss": 5.6292,
      "mean_token_accuracy": 0.17320444583892822,
      "num_tokens": 68591880.0,
      "step": 33850
    },
    {
      "entropy": 5.9729211807250975,
      "epoch": 2.958577296163593,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041309015450939694,
      "loss": 5.5988,
      "mean_token_accuracy": 0.1796057015657425,
      "num_tokens": 68603025.0,
      "step": 33855
    },
    {
      "entropy": 5.921748065948487,
      "epoch": 2.959014244516298,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004130655510976196,
      "loss": 5.5884,
      "mean_token_accuracy": 0.17389858216047288,
      "num_tokens": 68613580.0,
      "step": 33860
    },
    {
      "entropy": 5.9317831039428714,
      "epoch": 2.959451192869003,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041304094503767256,
      "loss": 5.6156,
      "mean_token_accuracy": 0.17730202078819274,
      "num_tokens": 68624451.0,
      "step": 33865
    },
    {
      "entropy": 5.860482120513916,
      "epoch": 2.9598881412217075,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00041301633633002784,
      "loss": 5.5073,
      "mean_token_accuracy": 0.1859039768576622,
      "num_tokens": 68634408.0,
      "step": 33870
    },
    {
      "entropy": 5.832436609268188,
      "epoch": 2.9603250895744124,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004129917249751575,
      "loss": 5.4579,
      "mean_token_accuracy": 0.18902477025985717,
      "num_tokens": 68643795.0,
      "step": 33875
    },
    {
      "entropy": 5.891926527023315,
      "epoch": 2.960762037927117,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004129671109735336,
      "loss": 5.4661,
      "mean_token_accuracy": 0.18145961910486222,
      "num_tokens": 68653669.0,
      "step": 33880
    },
    {
      "entropy": 5.960223293304443,
      "epoch": 2.961198986279822,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041294249432562825,
      "loss": 5.5659,
      "mean_token_accuracy": 0.177358914911747,
      "num_tokens": 68663094.0,
      "step": 33885
    },
    {
      "entropy": 5.825188732147216,
      "epoch": 2.9616359346325263,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041291787503191347,
      "loss": 5.4795,
      "mean_token_accuracy": 0.19574272483587266,
      "num_tokens": 68673561.0,
      "step": 33890
    },
    {
      "entropy": 5.900087070465088,
      "epoch": 2.962072882985231,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041289325309286167,
      "loss": 5.6027,
      "mean_token_accuracy": 0.16976108700037001,
      "num_tokens": 68683624.0,
      "step": 33895
    },
    {
      "entropy": 5.96026120185852,
      "epoch": 2.962509831337936,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041286862850894497,
      "loss": 5.5578,
      "mean_token_accuracy": 0.18146113604307174,
      "num_tokens": 68693805.0,
      "step": 33900
    },
    {
      "entropy": 5.93872447013855,
      "epoch": 2.9629467796906406,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004128440012806357,
      "loss": 5.558,
      "mean_token_accuracy": 0.18098945170640945,
      "num_tokens": 68704020.0,
      "step": 33905
    },
    {
      "entropy": 5.91333236694336,
      "epoch": 2.963383728043345,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041281937140840634,
      "loss": 5.4886,
      "mean_token_accuracy": 0.1857759952545166,
      "num_tokens": 68714486.0,
      "step": 33910
    },
    {
      "entropy": 5.881474256515503,
      "epoch": 2.96382067639605,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004127947388927292,
      "loss": 5.4777,
      "mean_token_accuracy": 0.18442461937665938,
      "num_tokens": 68724759.0,
      "step": 33915
    },
    {
      "entropy": 5.8345112800598145,
      "epoch": 2.964257624748755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004127701037340767,
      "loss": 5.4316,
      "mean_token_accuracy": 0.18960844576358796,
      "num_tokens": 68734826.0,
      "step": 33920
    },
    {
      "entropy": 5.845219230651855,
      "epoch": 2.9646945731014593,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041274546593292143,
      "loss": 5.5421,
      "mean_token_accuracy": 0.18475598245859146,
      "num_tokens": 68744786.0,
      "step": 33925
    },
    {
      "entropy": 5.940587854385376,
      "epoch": 2.9651315214541643,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004127208254897361,
      "loss": 5.5515,
      "mean_token_accuracy": 0.1807931423187256,
      "num_tokens": 68754751.0,
      "step": 33930
    },
    {
      "entropy": 5.957643461227417,
      "epoch": 2.9655684698068687,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041269618240499296,
      "loss": 5.5443,
      "mean_token_accuracy": 0.17615481466054916,
      "num_tokens": 68765652.0,
      "step": 33935
    },
    {
      "entropy": 5.873692750930786,
      "epoch": 2.9660054181595736,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00041267153667916506,
      "loss": 5.488,
      "mean_token_accuracy": 0.1872515544295311,
      "num_tokens": 68775587.0,
      "step": 33940
    },
    {
      "entropy": 5.900841426849365,
      "epoch": 2.966442366512278,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041264688831272483,
      "loss": 5.5171,
      "mean_token_accuracy": 0.18013301342725754,
      "num_tokens": 68786678.0,
      "step": 33945
    },
    {
      "entropy": 5.859254693984985,
      "epoch": 2.966879314864983,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004126222373061452,
      "loss": 5.3944,
      "mean_token_accuracy": 0.18903290331363679,
      "num_tokens": 68796108.0,
      "step": 33950
    },
    {
      "entropy": 5.823514747619629,
      "epoch": 2.967316263217688,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041259758365989896,
      "loss": 5.4741,
      "mean_token_accuracy": 0.18487164974212647,
      "num_tokens": 68806364.0,
      "step": 33955
    },
    {
      "entropy": 5.959085893630982,
      "epoch": 2.9677532115703924,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004125729273744589,
      "loss": 5.5563,
      "mean_token_accuracy": 0.17691604793071747,
      "num_tokens": 68816099.0,
      "step": 33960
    },
    {
      "entropy": 5.882583284378052,
      "epoch": 2.968190159923097,
      "grad_norm": 1.125,
      "learning_rate": 0.00041254826845029806,
      "loss": 5.5482,
      "mean_token_accuracy": 0.1801913484930992,
      "num_tokens": 68826480.0,
      "step": 33965
    },
    {
      "entropy": 5.879341888427734,
      "epoch": 2.968627108275802,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041252360688788937,
      "loss": 5.4779,
      "mean_token_accuracy": 0.18534045815467834,
      "num_tokens": 68835657.0,
      "step": 33970
    },
    {
      "entropy": 5.958382940292358,
      "epoch": 2.9690640566285067,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004124989426877058,
      "loss": 5.5501,
      "mean_token_accuracy": 0.18412982374429704,
      "num_tokens": 68845115.0,
      "step": 33975
    },
    {
      "entropy": 5.8907421112060545,
      "epoch": 2.969501004981211,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004124742758502205,
      "loss": 5.5196,
      "mean_token_accuracy": 0.18077187836170197,
      "num_tokens": 68854562.0,
      "step": 33980
    },
    {
      "entropy": 5.909314966201782,
      "epoch": 2.969937953333916,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004124496063759066,
      "loss": 5.582,
      "mean_token_accuracy": 0.17953250855207442,
      "num_tokens": 68864787.0,
      "step": 33985
    },
    {
      "entropy": 5.869363594055176,
      "epoch": 2.9703749016866206,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004124249342652371,
      "loss": 5.4832,
      "mean_token_accuracy": 0.1823698416352272,
      "num_tokens": 68874518.0,
      "step": 33990
    },
    {
      "entropy": 5.962499284744263,
      "epoch": 2.9708118500393255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041240025951868545,
      "loss": 5.4576,
      "mean_token_accuracy": 0.17736574113368989,
      "num_tokens": 68883765.0,
      "step": 33995
    },
    {
      "entropy": 5.919967079162598,
      "epoch": 2.97124879839203,
      "grad_norm": 1.703125,
      "learning_rate": 0.00041237558213672477,
      "loss": 5.525,
      "mean_token_accuracy": 0.1824126273393631,
      "num_tokens": 68894191.0,
      "step": 34000
    },
    {
      "entropy": 5.852690601348877,
      "epoch": 2.971685746744735,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041235090211982846,
      "loss": 5.4849,
      "mean_token_accuracy": 0.18581901490688324,
      "num_tokens": 68905136.0,
      "step": 34005
    },
    {
      "entropy": 5.8839170932769775,
      "epoch": 2.9721226950974398,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004123262194684699,
      "loss": 5.4849,
      "mean_token_accuracy": 0.187214632332325,
      "num_tokens": 68914620.0,
      "step": 34010
    },
    {
      "entropy": 5.911972427368164,
      "epoch": 2.9725596434501442,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041230153418312244,
      "loss": 5.5147,
      "mean_token_accuracy": 0.1831081837415695,
      "num_tokens": 68923781.0,
      "step": 34015
    },
    {
      "entropy": 5.8262285709381105,
      "epoch": 2.9729965918028487,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004122768462642596,
      "loss": 5.5704,
      "mean_token_accuracy": 0.17301106601953506,
      "num_tokens": 68934283.0,
      "step": 34020
    },
    {
      "entropy": 5.862225866317749,
      "epoch": 2.9734335401555536,
      "grad_norm": 1.109375,
      "learning_rate": 0.000412252155712355,
      "loss": 5.4562,
      "mean_token_accuracy": 0.18499594926834106,
      "num_tokens": 68943799.0,
      "step": 34025
    },
    {
      "entropy": 5.958679962158203,
      "epoch": 2.9738704885082585,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041222746252788213,
      "loss": 5.5414,
      "mean_token_accuracy": 0.18549837470054625,
      "num_tokens": 68953733.0,
      "step": 34030
    },
    {
      "entropy": 5.883492279052734,
      "epoch": 2.974307436860963,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004122027667113146,
      "loss": 5.5107,
      "mean_token_accuracy": 0.18073685914278032,
      "num_tokens": 68963888.0,
      "step": 34035
    },
    {
      "entropy": 5.9608831882476805,
      "epoch": 2.974744385213668,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000412178068263126,
      "loss": 5.5758,
      "mean_token_accuracy": 0.17674157470464708,
      "num_tokens": 68974652.0,
      "step": 34040
    },
    {
      "entropy": 5.961799144744873,
      "epoch": 2.9751813335663724,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004121533671837903,
      "loss": 5.4719,
      "mean_token_accuracy": 0.1857839494943619,
      "num_tokens": 68984583.0,
      "step": 34045
    },
    {
      "entropy": 5.849625015258789,
      "epoch": 2.9756182819190773,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004121286634737811,
      "loss": 5.5776,
      "mean_token_accuracy": 0.17622963339090347,
      "num_tokens": 68995790.0,
      "step": 34050
    },
    {
      "entropy": 5.930218601226807,
      "epoch": 2.9760552302717818,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004121039571335723,
      "loss": 5.5398,
      "mean_token_accuracy": 0.17856047302484512,
      "num_tokens": 69006695.0,
      "step": 34055
    },
    {
      "entropy": 5.85222020149231,
      "epoch": 2.9764921786244867,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004120792481636378,
      "loss": 5.4586,
      "mean_token_accuracy": 0.1942053034901619,
      "num_tokens": 69017504.0,
      "step": 34060
    },
    {
      "entropy": 5.891269540786743,
      "epoch": 2.976929126977191,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004120545365644515,
      "loss": 5.6048,
      "mean_token_accuracy": 0.17422419488430024,
      "num_tokens": 69026938.0,
      "step": 34065
    },
    {
      "entropy": 5.937666273117065,
      "epoch": 2.977366075329896,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004120298223364872,
      "loss": 5.4734,
      "mean_token_accuracy": 0.18477753549814224,
      "num_tokens": 69037404.0,
      "step": 34070
    },
    {
      "entropy": 5.899027729034424,
      "epoch": 2.9778030236826005,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041200510548021935,
      "loss": 5.4897,
      "mean_token_accuracy": 0.18987805694341658,
      "num_tokens": 69047507.0,
      "step": 34075
    },
    {
      "entropy": 5.8595963478088375,
      "epoch": 2.9782399720353054,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041198038599612156,
      "loss": 5.5016,
      "mean_token_accuracy": 0.18231384754180907,
      "num_tokens": 69057626.0,
      "step": 34080
    },
    {
      "entropy": 5.913874626159668,
      "epoch": 2.9786769203880104,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041195566388466836,
      "loss": 5.5142,
      "mean_token_accuracy": 0.18241243660449982,
      "num_tokens": 69068142.0,
      "step": 34085
    },
    {
      "entropy": 5.919228553771973,
      "epoch": 2.979113868740715,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041193093914633364,
      "loss": 5.5299,
      "mean_token_accuracy": 0.18438478261232377,
      "num_tokens": 69077935.0,
      "step": 34090
    },
    {
      "entropy": 5.926045656204224,
      "epoch": 2.9795508170934193,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004119062117815918,
      "loss": 5.5883,
      "mean_token_accuracy": 0.16996269524097443,
      "num_tokens": 69087397.0,
      "step": 34095
    },
    {
      "entropy": 5.944768857955933,
      "epoch": 2.979987765446124,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004118814817909171,
      "loss": 5.5933,
      "mean_token_accuracy": 0.18321585953235625,
      "num_tokens": 69097666.0,
      "step": 34100
    },
    {
      "entropy": 5.884004402160644,
      "epoch": 2.980424713798829,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004118567491747839,
      "loss": 5.5213,
      "mean_token_accuracy": 0.17830887138843537,
      "num_tokens": 69107516.0,
      "step": 34105
    },
    {
      "entropy": 5.889015197753906,
      "epoch": 2.9808616621515336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004118320139336665,
      "loss": 5.4832,
      "mean_token_accuracy": 0.17870856076478958,
      "num_tokens": 69117462.0,
      "step": 34110
    },
    {
      "entropy": 5.940749263763427,
      "epoch": 2.9812986105042385,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004118072760680393,
      "loss": 5.4788,
      "mean_token_accuracy": 0.18703508675098418,
      "num_tokens": 69128171.0,
      "step": 34115
    },
    {
      "entropy": 5.815265655517578,
      "epoch": 2.981735558856943,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041178253557837697,
      "loss": 5.5064,
      "mean_token_accuracy": 0.18436640799045562,
      "num_tokens": 69138169.0,
      "step": 34120
    },
    {
      "entropy": 5.833314561843872,
      "epoch": 2.982172507209648,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041175779246515384,
      "loss": 5.4486,
      "mean_token_accuracy": 0.18771958351135254,
      "num_tokens": 69147518.0,
      "step": 34125
    },
    {
      "entropy": 5.8775101661682125,
      "epoch": 2.9826094555623524,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004117330467288447,
      "loss": 5.4954,
      "mean_token_accuracy": 0.18187984675168992,
      "num_tokens": 69157049.0,
      "step": 34130
    },
    {
      "entropy": 5.8465173721313475,
      "epoch": 2.9830464039150573,
      "grad_norm": 1.125,
      "learning_rate": 0.00041170829836992403,
      "loss": 5.5042,
      "mean_token_accuracy": 0.18619152158498764,
      "num_tokens": 69166502.0,
      "step": 34135
    },
    {
      "entropy": 5.930161190032959,
      "epoch": 2.983483352267762,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004116835473888665,
      "loss": 5.4621,
      "mean_token_accuracy": 0.18208789676427842,
      "num_tokens": 69176566.0,
      "step": 34140
    },
    {
      "entropy": 5.859261465072632,
      "epoch": 2.9839203006204666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004116587937861469,
      "loss": 5.4855,
      "mean_token_accuracy": 0.18066295236349106,
      "num_tokens": 69186677.0,
      "step": 34145
    },
    {
      "entropy": 5.8663520336151125,
      "epoch": 2.984357248973171,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004116340375622402,
      "loss": 5.5717,
      "mean_token_accuracy": 0.17807987332344055,
      "num_tokens": 69197595.0,
      "step": 34150
    },
    {
      "entropy": 5.892182970046997,
      "epoch": 2.984794197325876,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004116092787176209,
      "loss": 5.5023,
      "mean_token_accuracy": 0.1883601114153862,
      "num_tokens": 69207384.0,
      "step": 34155
    },
    {
      "entropy": 5.8995672225952145,
      "epoch": 2.985231145678581,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041158451725276405,
      "loss": 5.3939,
      "mean_token_accuracy": 0.1963351309299469,
      "num_tokens": 69217019.0,
      "step": 34160
    },
    {
      "entropy": 5.9743914127349855,
      "epoch": 2.9856680940312854,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004115597531681447,
      "loss": 5.6788,
      "mean_token_accuracy": 0.1667443871498108,
      "num_tokens": 69229076.0,
      "step": 34165
    },
    {
      "entropy": 5.944848299026489,
      "epoch": 2.9861050423839903,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041153498646423763,
      "loss": 5.6082,
      "mean_token_accuracy": 0.17577505856752396,
      "num_tokens": 69238771.0,
      "step": 34170
    },
    {
      "entropy": 5.927469158172608,
      "epoch": 2.986541990736695,
      "grad_norm": 1.15625,
      "learning_rate": 0.000411510217141518,
      "loss": 5.5063,
      "mean_token_accuracy": 0.18152387738227843,
      "num_tokens": 69248864.0,
      "step": 34175
    },
    {
      "entropy": 5.86846923828125,
      "epoch": 2.9869789390893997,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041148544520046087,
      "loss": 5.5796,
      "mean_token_accuracy": 0.17617340981960297,
      "num_tokens": 69259372.0,
      "step": 34180
    },
    {
      "entropy": 5.836015844345093,
      "epoch": 2.987415887442104,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004114606706415415,
      "loss": 5.4576,
      "mean_token_accuracy": 0.18259613811969758,
      "num_tokens": 69270079.0,
      "step": 34185
    },
    {
      "entropy": 5.928583431243896,
      "epoch": 2.987852835794809,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004114358934652349,
      "loss": 5.5065,
      "mean_token_accuracy": 0.1841472268104553,
      "num_tokens": 69279909.0,
      "step": 34190
    },
    {
      "entropy": 5.859051322937011,
      "epoch": 2.988289784147514,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004114111136720163,
      "loss": 5.4408,
      "mean_token_accuracy": 0.18973531872034072,
      "num_tokens": 69289598.0,
      "step": 34195
    },
    {
      "entropy": 5.945561122894287,
      "epoch": 2.9887267325002185,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004113863312623611,
      "loss": 5.5382,
      "mean_token_accuracy": 0.18094297200441362,
      "num_tokens": 69301305.0,
      "step": 34200
    },
    {
      "entropy": 5.8708984375,
      "epoch": 2.989163680852923,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004113615462367446,
      "loss": 5.4815,
      "mean_token_accuracy": 0.187975013256073,
      "num_tokens": 69310560.0,
      "step": 34205
    },
    {
      "entropy": 5.819914245605469,
      "epoch": 2.989600629205628,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041133675859564213,
      "loss": 5.5477,
      "mean_token_accuracy": 0.1835186317563057,
      "num_tokens": 69320286.0,
      "step": 34210
    },
    {
      "entropy": 5.86975793838501,
      "epoch": 2.9900375775583328,
      "grad_norm": 1.125,
      "learning_rate": 0.00041131196833952925,
      "loss": 5.4577,
      "mean_token_accuracy": 0.18435653001070024,
      "num_tokens": 69329895.0,
      "step": 34215
    },
    {
      "entropy": 5.935191631317139,
      "epoch": 2.9904745259110372,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004112871754688814,
      "loss": 5.4687,
      "mean_token_accuracy": 0.18355579525232316,
      "num_tokens": 69340122.0,
      "step": 34220
    },
    {
      "entropy": 5.866327381134033,
      "epoch": 2.990911474263742,
      "grad_norm": 1.0,
      "learning_rate": 0.00041126237998417414,
      "loss": 5.4865,
      "mean_token_accuracy": 0.1857226848602295,
      "num_tokens": 69349354.0,
      "step": 34225
    },
    {
      "entropy": 5.897889757156372,
      "epoch": 2.9913484226164466,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041123758188588296,
      "loss": 5.5744,
      "mean_token_accuracy": 0.1714955076575279,
      "num_tokens": 69360236.0,
      "step": 34230
    },
    {
      "entropy": 5.973222446441651,
      "epoch": 2.9917853709691515,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041121278117448365,
      "loss": 5.5345,
      "mean_token_accuracy": 0.18606639951467513,
      "num_tokens": 69369362.0,
      "step": 34235
    },
    {
      "entropy": 5.935163402557373,
      "epoch": 2.992222319321856,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041118797785045174,
      "loss": 5.5408,
      "mean_token_accuracy": 0.17926606833934783,
      "num_tokens": 69379403.0,
      "step": 34240
    },
    {
      "entropy": 5.858764696121216,
      "epoch": 2.992659267674561,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004111631719142631,
      "loss": 5.5153,
      "mean_token_accuracy": 0.18675494343042373,
      "num_tokens": 69388667.0,
      "step": 34245
    },
    {
      "entropy": 5.924489259719849,
      "epoch": 2.993096216027266,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041113836336639347,
      "loss": 5.4999,
      "mean_token_accuracy": 0.18277183175086975,
      "num_tokens": 69399552.0,
      "step": 34250
    },
    {
      "entropy": 5.900462675094604,
      "epoch": 2.9935331643799703,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004111135522073187,
      "loss": 5.3992,
      "mean_token_accuracy": 0.19065149277448654,
      "num_tokens": 69409323.0,
      "step": 34255
    },
    {
      "entropy": 5.872309064865112,
      "epoch": 2.9939701127326748,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004110887384375147,
      "loss": 5.5282,
      "mean_token_accuracy": 0.18019166737794876,
      "num_tokens": 69418696.0,
      "step": 34260
    },
    {
      "entropy": 5.930271959304809,
      "epoch": 2.9944070610853797,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004110639220574573,
      "loss": 5.61,
      "mean_token_accuracy": 0.17685055285692214,
      "num_tokens": 69429470.0,
      "step": 34265
    },
    {
      "entropy": 5.868441200256347,
      "epoch": 2.9948440094380846,
      "grad_norm": 1.125,
      "learning_rate": 0.00041103910306762267,
      "loss": 5.4604,
      "mean_token_accuracy": 0.1881233796477318,
      "num_tokens": 69440636.0,
      "step": 34270
    },
    {
      "entropy": 5.902501440048217,
      "epoch": 2.995280957790789,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041101428146848667,
      "loss": 5.5399,
      "mean_token_accuracy": 0.18130839616060257,
      "num_tokens": 69451819.0,
      "step": 34275
    },
    {
      "entropy": 5.916478824615479,
      "epoch": 2.995717906143494,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00041098945726052556,
      "loss": 5.6107,
      "mean_token_accuracy": 0.17824551314115525,
      "num_tokens": 69462690.0,
      "step": 34280
    },
    {
      "entropy": 5.8714793682098385,
      "epoch": 2.9961548544961984,
      "grad_norm": 1.0,
      "learning_rate": 0.0004109646304442154,
      "loss": 5.46,
      "mean_token_accuracy": 0.18791285902261734,
      "num_tokens": 69473338.0,
      "step": 34285
    },
    {
      "entropy": 5.909438037872315,
      "epoch": 2.9965918028489034,
      "grad_norm": 1.125,
      "learning_rate": 0.0004109398010200323,
      "loss": 5.495,
      "mean_token_accuracy": 0.18271346837282182,
      "num_tokens": 69482189.0,
      "step": 34290
    },
    {
      "entropy": 5.779447746276856,
      "epoch": 2.997028751201608,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004109149689884527,
      "loss": 5.443,
      "mean_token_accuracy": 0.19352191239595412,
      "num_tokens": 69492084.0,
      "step": 34295
    },
    {
      "entropy": 5.9151695728302,
      "epoch": 2.9974656995543127,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041089013434995273,
      "loss": 5.503,
      "mean_token_accuracy": 0.18986082226037979,
      "num_tokens": 69502886.0,
      "step": 34300
    },
    {
      "entropy": 5.921262502670288,
      "epoch": 2.9979026479070177,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004108652971050088,
      "loss": 5.4515,
      "mean_token_accuracy": 0.18528779447078705,
      "num_tokens": 69512670.0,
      "step": 34305
    },
    {
      "entropy": 5.811900949478149,
      "epoch": 2.998339596259722,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041084045725409726,
      "loss": 5.5424,
      "mean_token_accuracy": 0.1797345519065857,
      "num_tokens": 69522744.0,
      "step": 34310
    },
    {
      "entropy": 5.851929473876953,
      "epoch": 2.9987765446124266,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004108156147976945,
      "loss": 5.4015,
      "mean_token_accuracy": 0.1900213360786438,
      "num_tokens": 69533195.0,
      "step": 34315
    },
    {
      "entropy": 5.901485729217529,
      "epoch": 2.9992134929651315,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004107907697362772,
      "loss": 5.5694,
      "mean_token_accuracy": 0.18256947100162507,
      "num_tokens": 69543058.0,
      "step": 34320
    },
    {
      "entropy": 5.99129900932312,
      "epoch": 2.9996504413178364,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004107659220703217,
      "loss": 5.522,
      "mean_token_accuracy": 0.1841009169816971,
      "num_tokens": 69552428.0,
      "step": 34325
    },
    {
      "entropy": 5.9332949161529545,
      "epoch": 3.000087389670541,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004107410718003047,
      "loss": 5.4937,
      "mean_token_accuracy": 0.18564373254776,
      "num_tokens": 69561735.0,
      "step": 34330
    },
    {
      "entropy": 5.820268201828003,
      "epoch": 3.000524338023246,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004107162189267028,
      "loss": 5.478,
      "mean_token_accuracy": 0.18108728229999543,
      "num_tokens": 69573165.0,
      "step": 34335
    },
    {
      "entropy": 5.867929840087891,
      "epoch": 3.0009612863759503,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00041069136344999273,
      "loss": 5.4416,
      "mean_token_accuracy": 0.1842103958129883,
      "num_tokens": 69583484.0,
      "step": 34340
    },
    {
      "entropy": 5.863696002960205,
      "epoch": 3.001398234728655,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004106665053706512,
      "loss": 5.4629,
      "mean_token_accuracy": 0.18436622321605683,
      "num_tokens": 69593486.0,
      "step": 34345
    },
    {
      "entropy": 5.937929677963257,
      "epoch": 3.0018351830813597,
      "grad_norm": 1.078125,
      "learning_rate": 0.000410641644689155,
      "loss": 5.5538,
      "mean_token_accuracy": 0.1789838746190071,
      "num_tokens": 69604245.0,
      "step": 34350
    },
    {
      "entropy": 5.937907743453979,
      "epoch": 3.0022721314340646,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000410616781405981,
      "loss": 5.4662,
      "mean_token_accuracy": 0.18803012520074844,
      "num_tokens": 69614237.0,
      "step": 34355
    },
    {
      "entropy": 5.9121095657348635,
      "epoch": 3.002709079786769,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004105919155216061,
      "loss": 5.4792,
      "mean_token_accuracy": 0.18479037880897523,
      "num_tokens": 69623695.0,
      "step": 34360
    },
    {
      "entropy": 5.891081476211548,
      "epoch": 3.003146028139474,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041056704703650715,
      "loss": 5.3803,
      "mean_token_accuracy": 0.19502062499523162,
      "num_tokens": 69633299.0,
      "step": 34365
    },
    {
      "entropy": 5.8530652046203615,
      "epoch": 3.0035829764921784,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004105421759511612,
      "loss": 5.4356,
      "mean_token_accuracy": 0.1808922529220581,
      "num_tokens": 69642853.0,
      "step": 34370
    },
    {
      "entropy": 5.882161951065063,
      "epoch": 3.0040199248448833,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004105173022660453,
      "loss": 5.4749,
      "mean_token_accuracy": 0.18382043689489364,
      "num_tokens": 69652697.0,
      "step": 34375
    },
    {
      "entropy": 5.915562629699707,
      "epoch": 3.0044568731975883,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004104924259816366,
      "loss": 5.5064,
      "mean_token_accuracy": 0.19109993875026704,
      "num_tokens": 69664261.0,
      "step": 34380
    },
    {
      "entropy": 5.8996742248535154,
      "epoch": 3.0048938215502927,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004104675470984121,
      "loss": 5.4881,
      "mean_token_accuracy": 0.19279077500104905,
      "num_tokens": 69674819.0,
      "step": 34385
    },
    {
      "entropy": 5.850151538848877,
      "epoch": 3.0053307699029976,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041044266561684914,
      "loss": 5.4408,
      "mean_token_accuracy": 0.18495372086763381,
      "num_tokens": 69685546.0,
      "step": 34390
    },
    {
      "entropy": 5.79946813583374,
      "epoch": 3.005767718255702,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004104177815374248,
      "loss": 5.365,
      "mean_token_accuracy": 0.19764966666698455,
      "num_tokens": 69695192.0,
      "step": 34395
    },
    {
      "entropy": 5.8833013534545895,
      "epoch": 3.006204666608407,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004103928948606164,
      "loss": 5.4109,
      "mean_token_accuracy": 0.18652660697698592,
      "num_tokens": 69705669.0,
      "step": 34400
    },
    {
      "entropy": 5.794457340240479,
      "epoch": 3.0066416149611115,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004103680055869015,
      "loss": 5.4019,
      "mean_token_accuracy": 0.19682155400514603,
      "num_tokens": 69716274.0,
      "step": 34405
    },
    {
      "entropy": 5.89814715385437,
      "epoch": 3.0070785633138164,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004103431137167572,
      "loss": 5.4682,
      "mean_token_accuracy": 0.18899813145399094,
      "num_tokens": 69726438.0,
      "step": 34410
    },
    {
      "entropy": 5.881954193115234,
      "epoch": 3.007515511666521,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041031821925066115,
      "loss": 5.4668,
      "mean_token_accuracy": 0.19105769246816634,
      "num_tokens": 69735467.0,
      "step": 34415
    },
    {
      "entropy": 5.857857370376587,
      "epoch": 3.007952460019226,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041029332218909067,
      "loss": 5.4571,
      "mean_token_accuracy": 0.18565064072608947,
      "num_tokens": 69744886.0,
      "step": 34420
    },
    {
      "entropy": 5.862932586669922,
      "epoch": 3.0083894083719303,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041026842253252333,
      "loss": 5.4672,
      "mean_token_accuracy": 0.17915499061346055,
      "num_tokens": 69754337.0,
      "step": 34425
    },
    {
      "entropy": 5.823921823501587,
      "epoch": 3.008826356724635,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004102435202814369,
      "loss": 5.4451,
      "mean_token_accuracy": 0.1887630507349968,
      "num_tokens": 69764660.0,
      "step": 34430
    },
    {
      "entropy": 5.8515619277954105,
      "epoch": 3.00926330507734,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004102186154363088,
      "loss": 5.4173,
      "mean_token_accuracy": 0.1879889577627182,
      "num_tokens": 69774339.0,
      "step": 34435
    },
    {
      "entropy": 5.900795364379883,
      "epoch": 3.0097002534300445,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00041019370799761685,
      "loss": 5.5091,
      "mean_token_accuracy": 0.18410640954971313,
      "num_tokens": 69785272.0,
      "step": 34440
    },
    {
      "entropy": 5.881802368164062,
      "epoch": 3.0101372017827495,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004101687979658386,
      "loss": 5.5427,
      "mean_token_accuracy": 0.18309650719165801,
      "num_tokens": 69796646.0,
      "step": 34445
    },
    {
      "entropy": 5.859975671768188,
      "epoch": 3.010574150135454,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041014388534145213,
      "loss": 5.4093,
      "mean_token_accuracy": 0.19501478970050812,
      "num_tokens": 69807233.0,
      "step": 34450
    },
    {
      "entropy": 5.910800552368164,
      "epoch": 3.011011098488159,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004101189701249351,
      "loss": 5.4923,
      "mean_token_accuracy": 0.1828117236495018,
      "num_tokens": 69818131.0,
      "step": 34455
    },
    {
      "entropy": 5.952968120574951,
      "epoch": 3.0114480468408633,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004100940523167654,
      "loss": 5.5659,
      "mean_token_accuracy": 0.17652879953384398,
      "num_tokens": 69829537.0,
      "step": 34460
    },
    {
      "entropy": 5.926409101486206,
      "epoch": 3.0118849951935682,
      "grad_norm": 1.265625,
      "learning_rate": 0.000410069131917421,
      "loss": 5.5033,
      "mean_token_accuracy": 0.1793579250574112,
      "num_tokens": 69840964.0,
      "step": 34465
    },
    {
      "entropy": 5.879490184783935,
      "epoch": 3.0123219435462727,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041004420892737993,
      "loss": 5.3832,
      "mean_token_accuracy": 0.1896154671907425,
      "num_tokens": 69849919.0,
      "step": 34470
    },
    {
      "entropy": 5.824447774887085,
      "epoch": 3.0127588918989776,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004100192833471201,
      "loss": 5.4951,
      "mean_token_accuracy": 0.1861352115869522,
      "num_tokens": 69859832.0,
      "step": 34475
    },
    {
      "entropy": 5.860016870498657,
      "epoch": 3.013195840251682,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004099943551771196,
      "loss": 5.3991,
      "mean_token_accuracy": 0.19151399284601212,
      "num_tokens": 69869382.0,
      "step": 34480
    },
    {
      "entropy": 5.857220697402954,
      "epoch": 3.013632788604387,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004099694244178568,
      "loss": 5.426,
      "mean_token_accuracy": 0.18823544234037398,
      "num_tokens": 69879282.0,
      "step": 34485
    },
    {
      "entropy": 5.82499132156372,
      "epoch": 3.0140697369570915,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004099444910698096,
      "loss": 5.3822,
      "mean_token_accuracy": 0.19525910317897796,
      "num_tokens": 69889876.0,
      "step": 34490
    },
    {
      "entropy": 5.895588159561157,
      "epoch": 3.0145066853097964,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004099195551334564,
      "loss": 5.4801,
      "mean_token_accuracy": 0.1771703839302063,
      "num_tokens": 69899642.0,
      "step": 34495
    },
    {
      "entropy": 5.849583578109741,
      "epoch": 3.0149436336625013,
      "grad_norm": 2.234375,
      "learning_rate": 0.00040989461660927536,
      "loss": 5.4068,
      "mean_token_accuracy": 0.18441249281167985,
      "num_tokens": 69909877.0,
      "step": 34500
    },
    {
      "entropy": 5.875034952163697,
      "epoch": 3.0153805820152058,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040986967549774497,
      "loss": 5.4041,
      "mean_token_accuracy": 0.19133496284484863,
      "num_tokens": 69920061.0,
      "step": 34505
    },
    {
      "entropy": 5.816222763061523,
      "epoch": 3.0158175303679107,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004098447317993435,
      "loss": 5.4225,
      "mean_token_accuracy": 0.19185275137424468,
      "num_tokens": 69930739.0,
      "step": 34510
    },
    {
      "entropy": 5.766131353378296,
      "epoch": 3.016254478720615,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040981978551454947,
      "loss": 5.3449,
      "mean_token_accuracy": 0.19646138399839402,
      "num_tokens": 69939825.0,
      "step": 34515
    },
    {
      "entropy": 5.918108034133911,
      "epoch": 3.01669142707332,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040979483664384123,
      "loss": 5.6013,
      "mean_token_accuracy": 0.1722034126520157,
      "num_tokens": 69949662.0,
      "step": 34520
    },
    {
      "entropy": 5.864528369903565,
      "epoch": 3.0171283754260245,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040976988518769743,
      "loss": 5.4656,
      "mean_token_accuracy": 0.1928654581308365,
      "num_tokens": 69959950.0,
      "step": 34525
    },
    {
      "entropy": 5.904948329925537,
      "epoch": 3.0175653237787294,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040974493114659654,
      "loss": 5.504,
      "mean_token_accuracy": 0.1797540307044983,
      "num_tokens": 69970451.0,
      "step": 34530
    },
    {
      "entropy": 5.862445831298828,
      "epoch": 3.018002272131434,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004097199745210174,
      "loss": 5.4264,
      "mean_token_accuracy": 0.18797146826982497,
      "num_tokens": 69979940.0,
      "step": 34535
    },
    {
      "entropy": 5.865592813491821,
      "epoch": 3.018439220484139,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040969501531143844,
      "loss": 5.4024,
      "mean_token_accuracy": 0.19615158587694168,
      "num_tokens": 69989507.0,
      "step": 34540
    },
    {
      "entropy": 5.878801012039185,
      "epoch": 3.0188761688368433,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040967005351833863,
      "loss": 5.4343,
      "mean_token_accuracy": 0.1906401440501213,
      "num_tokens": 69999184.0,
      "step": 34545
    },
    {
      "entropy": 5.931103181838989,
      "epoch": 3.019313117189548,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004096450891421965,
      "loss": 5.5379,
      "mean_token_accuracy": 0.18082533031702042,
      "num_tokens": 70009569.0,
      "step": 34550
    },
    {
      "entropy": 5.9215014457702635,
      "epoch": 3.019750065542253,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040962012218349105,
      "loss": 5.4779,
      "mean_token_accuracy": 0.18528713136911393,
      "num_tokens": 70019824.0,
      "step": 34555
    },
    {
      "entropy": 5.812055873870849,
      "epoch": 3.0201870138949576,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004095951526427011,
      "loss": 5.3363,
      "mean_token_accuracy": 0.19426196813583374,
      "num_tokens": 70029043.0,
      "step": 34560
    },
    {
      "entropy": 5.904932069778442,
      "epoch": 3.0206239622476625,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004095701805203056,
      "loss": 5.4811,
      "mean_token_accuracy": 0.1827412948012352,
      "num_tokens": 70039794.0,
      "step": 34565
    },
    {
      "entropy": 5.898588752746582,
      "epoch": 3.021060910600367,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004095452058167837,
      "loss": 5.5018,
      "mean_token_accuracy": 0.18349916040897368,
      "num_tokens": 70049866.0,
      "step": 34570
    },
    {
      "entropy": 5.817253923416137,
      "epoch": 3.021497858953072,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000409520228532614,
      "loss": 5.4036,
      "mean_token_accuracy": 0.19412864446640016,
      "num_tokens": 70059471.0,
      "step": 34575
    },
    {
      "entropy": 5.875708293914795,
      "epoch": 3.0219348073057763,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000409495248668276,
      "loss": 5.4165,
      "mean_token_accuracy": 0.19627053588628768,
      "num_tokens": 70069003.0,
      "step": 34580
    },
    {
      "entropy": 5.895253372192383,
      "epoch": 3.0223717556584813,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040947026622424853,
      "loss": 5.4885,
      "mean_token_accuracy": 0.18647090792655946,
      "num_tokens": 70078650.0,
      "step": 34585
    },
    {
      "entropy": 5.8312928676605225,
      "epoch": 3.0228087040111857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040944528120101097,
      "loss": 5.4329,
      "mean_token_accuracy": 0.19201755672693252,
      "num_tokens": 70087361.0,
      "step": 34590
    },
    {
      "entropy": 5.85228967666626,
      "epoch": 3.0232456523638906,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040942029359904245,
      "loss": 5.4877,
      "mean_token_accuracy": 0.18682384639978408,
      "num_tokens": 70097289.0,
      "step": 34595
    },
    {
      "entropy": 5.879140853881836,
      "epoch": 3.023682600716595,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040939530341882225,
      "loss": 5.4335,
      "mean_token_accuracy": 0.19043500274419783,
      "num_tokens": 70108099.0,
      "step": 34600
    },
    {
      "entropy": 5.892002439498901,
      "epoch": 3.0241195490693,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040937031066082974,
      "loss": 5.5279,
      "mean_token_accuracy": 0.17674847692251205,
      "num_tokens": 70119243.0,
      "step": 34605
    },
    {
      "entropy": 5.857103633880615,
      "epoch": 3.0245564974220045,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004093453153255442,
      "loss": 5.4488,
      "mean_token_accuracy": 0.18824723958969117,
      "num_tokens": 70129073.0,
      "step": 34610
    },
    {
      "entropy": 5.8321010112762455,
      "epoch": 3.0249934457747094,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004093203174134452,
      "loss": 5.3863,
      "mean_token_accuracy": 0.19632358253002166,
      "num_tokens": 70139778.0,
      "step": 34615
    },
    {
      "entropy": 5.9190967082977295,
      "epoch": 3.0254303941274143,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040929531692501206,
      "loss": 5.5622,
      "mean_token_accuracy": 0.1752645716071129,
      "num_tokens": 70151088.0,
      "step": 34620
    },
    {
      "entropy": 5.806300449371338,
      "epoch": 3.025867342480119,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00040927031386072435,
      "loss": 5.4114,
      "mean_token_accuracy": 0.19497726261615753,
      "num_tokens": 70160133.0,
      "step": 34625
    },
    {
      "entropy": 5.859305715560913,
      "epoch": 3.0263042908328237,
      "grad_norm": 1.25,
      "learning_rate": 0.00040924530822106177,
      "loss": 5.4347,
      "mean_token_accuracy": 0.18380756676197052,
      "num_tokens": 70171176.0,
      "step": 34630
    },
    {
      "entropy": 5.88772873878479,
      "epoch": 3.026741239185528,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040922030000650376,
      "loss": 5.465,
      "mean_token_accuracy": 0.18267188519239425,
      "num_tokens": 70181413.0,
      "step": 34635
    },
    {
      "entropy": 5.860606384277344,
      "epoch": 3.027178187538233,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004091952892175301,
      "loss": 5.4749,
      "mean_token_accuracy": 0.18402609676122667,
      "num_tokens": 70190535.0,
      "step": 34640
    },
    {
      "entropy": 5.853446197509766,
      "epoch": 3.0276151358909376,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040917027585462044,
      "loss": 5.4457,
      "mean_token_accuracy": 0.19115019887685775,
      "num_tokens": 70199440.0,
      "step": 34645
    },
    {
      "entropy": 5.864360523223877,
      "epoch": 3.0280520842436425,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004091452599182546,
      "loss": 5.3978,
      "mean_token_accuracy": 0.18968817442655564,
      "num_tokens": 70209846.0,
      "step": 34650
    },
    {
      "entropy": 5.838032674789429,
      "epoch": 3.028489032596347,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040912024140891235,
      "loss": 5.4811,
      "mean_token_accuracy": 0.1841941311955452,
      "num_tokens": 70219910.0,
      "step": 34655
    },
    {
      "entropy": 5.876888418197632,
      "epoch": 3.028925980949052,
      "grad_norm": 1.25,
      "learning_rate": 0.0004090952203270737,
      "loss": 5.4584,
      "mean_token_accuracy": 0.18546702414751054,
      "num_tokens": 70230937.0,
      "step": 34660
    },
    {
      "entropy": 5.829709148406982,
      "epoch": 3.0293629293017563,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040907019667321833,
      "loss": 5.4744,
      "mean_token_accuracy": 0.18673080950975418,
      "num_tokens": 70240791.0,
      "step": 34665
    },
    {
      "entropy": 5.896839046478272,
      "epoch": 3.0297998776544612,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004090451704478264,
      "loss": 5.4583,
      "mean_token_accuracy": 0.19155978709459304,
      "num_tokens": 70251231.0,
      "step": 34670
    },
    {
      "entropy": 5.876859426498413,
      "epoch": 3.030236826007166,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040902014165137787,
      "loss": 5.3741,
      "mean_token_accuracy": 0.19046128839254378,
      "num_tokens": 70260993.0,
      "step": 34675
    },
    {
      "entropy": 5.9286130428314205,
      "epoch": 3.0306737743598706,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004089951102843528,
      "loss": 5.5598,
      "mean_token_accuracy": 0.1820199266076088,
      "num_tokens": 70271380.0,
      "step": 34680
    },
    {
      "entropy": 5.92327127456665,
      "epoch": 3.0311107227125755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040897007634723125,
      "loss": 5.4928,
      "mean_token_accuracy": 0.18471347093582152,
      "num_tokens": 70282614.0,
      "step": 34685
    },
    {
      "entropy": 5.884122037887574,
      "epoch": 3.03154767106528,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004089450398404935,
      "loss": 5.4296,
      "mean_token_accuracy": 0.18436165302991867,
      "num_tokens": 70292084.0,
      "step": 34690
    },
    {
      "entropy": 5.796443176269531,
      "epoch": 3.031984619417985,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040892000076461966,
      "loss": 5.3527,
      "mean_token_accuracy": 0.19039742797613143,
      "num_tokens": 70302205.0,
      "step": 34695
    },
    {
      "entropy": 5.857495307922363,
      "epoch": 3.0324215677706894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040889495912009,
      "loss": 5.5099,
      "mean_token_accuracy": 0.18505624532699586,
      "num_tokens": 70312264.0,
      "step": 34700
    },
    {
      "entropy": 5.881351041793823,
      "epoch": 3.0328585161233943,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040886991490738495,
      "loss": 5.4163,
      "mean_token_accuracy": 0.18624035716056825,
      "num_tokens": 70321733.0,
      "step": 34705
    },
    {
      "entropy": 5.85012149810791,
      "epoch": 3.0332954644760988,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040884486812698475,
      "loss": 5.5191,
      "mean_token_accuracy": 0.17983121126890184,
      "num_tokens": 70331621.0,
      "step": 34710
    },
    {
      "entropy": 5.84227499961853,
      "epoch": 3.0337324128288037,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040881981877936987,
      "loss": 5.4549,
      "mean_token_accuracy": 0.18438527584075928,
      "num_tokens": 70342247.0,
      "step": 34715
    },
    {
      "entropy": 5.833035707473755,
      "epoch": 3.034169361181508,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004087947668650207,
      "loss": 5.4279,
      "mean_token_accuracy": 0.1917800098657608,
      "num_tokens": 70352423.0,
      "step": 34720
    },
    {
      "entropy": 5.923224067687988,
      "epoch": 3.034606309534213,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004087697123844178,
      "loss": 5.5061,
      "mean_token_accuracy": 0.18222893327474593,
      "num_tokens": 70362740.0,
      "step": 34725
    },
    {
      "entropy": 5.864924669265747,
      "epoch": 3.035043257886918,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004087446553380417,
      "loss": 5.3899,
      "mean_token_accuracy": 0.1947702020406723,
      "num_tokens": 70372456.0,
      "step": 34730
    },
    {
      "entropy": 5.826031875610352,
      "epoch": 3.0354802062396224,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004087195957263731,
      "loss": 5.4169,
      "mean_token_accuracy": 0.19817046076059341,
      "num_tokens": 70382741.0,
      "step": 34735
    },
    {
      "entropy": 5.787367105484009,
      "epoch": 3.0359171545923274,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040869453354989255,
      "loss": 5.4132,
      "mean_token_accuracy": 0.18648206293582917,
      "num_tokens": 70392712.0,
      "step": 34740
    },
    {
      "entropy": 5.87633376121521,
      "epoch": 3.036354102945032,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040866946880908076,
      "loss": 5.5136,
      "mean_token_accuracy": 0.18529810309410094,
      "num_tokens": 70403655.0,
      "step": 34745
    },
    {
      "entropy": 5.9873528480529785,
      "epoch": 3.0367910512977367,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004086444015044186,
      "loss": 5.564,
      "mean_token_accuracy": 0.17308548241853713,
      "num_tokens": 70413720.0,
      "step": 34750
    },
    {
      "entropy": 5.908074760437012,
      "epoch": 3.037227999650441,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004086193316363867,
      "loss": 5.4693,
      "mean_token_accuracy": 0.18522464781999587,
      "num_tokens": 70423910.0,
      "step": 34755
    },
    {
      "entropy": 5.859823989868164,
      "epoch": 3.037664948003146,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040859425920546604,
      "loss": 5.4771,
      "mean_token_accuracy": 0.18379694521427153,
      "num_tokens": 70434593.0,
      "step": 34760
    },
    {
      "entropy": 5.864034032821655,
      "epoch": 3.0381018963558506,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004085691842121375,
      "loss": 5.4312,
      "mean_token_accuracy": 0.18675966262817384,
      "num_tokens": 70443967.0,
      "step": 34765
    },
    {
      "entropy": 5.919545984268188,
      "epoch": 3.0385388447085555,
      "grad_norm": 1.109375,
      "learning_rate": 0.000408544106656882,
      "loss": 5.4457,
      "mean_token_accuracy": 0.18316383212804793,
      "num_tokens": 70454098.0,
      "step": 34770
    },
    {
      "entropy": 5.846420049667358,
      "epoch": 3.03897579306126,
      "grad_norm": 0.984375,
      "learning_rate": 0.00040851902654018057,
      "loss": 5.4313,
      "mean_token_accuracy": 0.18939936012029648,
      "num_tokens": 70463946.0,
      "step": 34775
    },
    {
      "entropy": 5.94201021194458,
      "epoch": 3.039412741413965,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040849394386251425,
      "loss": 5.4711,
      "mean_token_accuracy": 0.1869213968515396,
      "num_tokens": 70474805.0,
      "step": 34780
    },
    {
      "entropy": 5.803517246246338,
      "epoch": 3.03984968976667,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004084688586243641,
      "loss": 5.3876,
      "mean_token_accuracy": 0.19157457798719407,
      "num_tokens": 70484585.0,
      "step": 34785
    },
    {
      "entropy": 5.874563884735108,
      "epoch": 3.0402866381193743,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004084437708262113,
      "loss": 5.4642,
      "mean_token_accuracy": 0.18424742370843888,
      "num_tokens": 70494811.0,
      "step": 34790
    },
    {
      "entropy": 5.934290266036987,
      "epoch": 3.040723586472079,
      "grad_norm": 1.0,
      "learning_rate": 0.0004084186804685371,
      "loss": 5.5058,
      "mean_token_accuracy": 0.17909509241580962,
      "num_tokens": 70505392.0,
      "step": 34795
    },
    {
      "entropy": 5.83153510093689,
      "epoch": 3.0411605348247837,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004083935875518227,
      "loss": 5.425,
      "mean_token_accuracy": 0.19531687945127488,
      "num_tokens": 70515066.0,
      "step": 34800
    },
    {
      "entropy": 5.877661323547363,
      "epoch": 3.0415974831774886,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040836849207654937,
      "loss": 5.403,
      "mean_token_accuracy": 0.1846579223871231,
      "num_tokens": 70525100.0,
      "step": 34805
    },
    {
      "entropy": 5.855971336364746,
      "epoch": 3.042034431530193,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004083433940431985,
      "loss": 5.4389,
      "mean_token_accuracy": 0.19087734669446946,
      "num_tokens": 70535286.0,
      "step": 34810
    },
    {
      "entropy": 5.872058010101318,
      "epoch": 3.042471379882898,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004083182934522514,
      "loss": 5.4806,
      "mean_token_accuracy": 0.18621524572372436,
      "num_tokens": 70545724.0,
      "step": 34815
    },
    {
      "entropy": 5.872526836395264,
      "epoch": 3.0429083282356024,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040829319030418967,
      "loss": 5.4544,
      "mean_token_accuracy": 0.18630764186382293,
      "num_tokens": 70555179.0,
      "step": 34820
    },
    {
      "entropy": 5.920904159545898,
      "epoch": 3.0433452765883073,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004082680845994946,
      "loss": 5.4734,
      "mean_token_accuracy": 0.185177181661129,
      "num_tokens": 70564355.0,
      "step": 34825
    },
    {
      "entropy": 5.796715307235718,
      "epoch": 3.043782224941012,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004082429763386479,
      "loss": 5.376,
      "mean_token_accuracy": 0.1984864056110382,
      "num_tokens": 70575367.0,
      "step": 34830
    },
    {
      "entropy": 5.763911962509155,
      "epoch": 3.0442191732937167,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040821786552213107,
      "loss": 5.406,
      "mean_token_accuracy": 0.19440398514270782,
      "num_tokens": 70585047.0,
      "step": 34835
    },
    {
      "entropy": 5.910327434539795,
      "epoch": 3.044656121646421,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040819275215042586,
      "loss": 5.4584,
      "mean_token_accuracy": 0.182684725522995,
      "num_tokens": 70594736.0,
      "step": 34840
    },
    {
      "entropy": 5.865161180496216,
      "epoch": 3.045093069999126,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040816763622401376,
      "loss": 5.4685,
      "mean_token_accuracy": 0.18439559191465377,
      "num_tokens": 70604762.0,
      "step": 34845
    },
    {
      "entropy": 5.890598583221435,
      "epoch": 3.045530018351831,
      "grad_norm": 1.59375,
      "learning_rate": 0.00040814251774337666,
      "loss": 5.4607,
      "mean_token_accuracy": 0.1873869851231575,
      "num_tokens": 70614438.0,
      "step": 34850
    },
    {
      "entropy": 5.968915128707886,
      "epoch": 3.0459669667045355,
      "grad_norm": 1.125,
      "learning_rate": 0.0004081173967089963,
      "loss": 5.5723,
      "mean_token_accuracy": 0.17724541276693345,
      "num_tokens": 70624669.0,
      "step": 34855
    },
    {
      "entropy": 5.851333570480347,
      "epoch": 3.0464039150572404,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040809227312135456,
      "loss": 5.3894,
      "mean_token_accuracy": 0.18591253757476806,
      "num_tokens": 70635082.0,
      "step": 34860
    },
    {
      "entropy": 5.875976991653443,
      "epoch": 3.046840863409945,
      "grad_norm": 1.0,
      "learning_rate": 0.0004080671469809332,
      "loss": 5.4319,
      "mean_token_accuracy": 0.1962618574500084,
      "num_tokens": 70644117.0,
      "step": 34865
    },
    {
      "entropy": 5.877263355255127,
      "epoch": 3.04727781176265,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004080420182882144,
      "loss": 5.4842,
      "mean_token_accuracy": 0.1808053657412529,
      "num_tokens": 70653299.0,
      "step": 34870
    },
    {
      "entropy": 5.867557334899902,
      "epoch": 3.0477147601153542,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040801688704367975,
      "loss": 5.4573,
      "mean_token_accuracy": 0.18565655946731568,
      "num_tokens": 70664477.0,
      "step": 34875
    },
    {
      "entropy": 5.897366094589233,
      "epoch": 3.048151708468059,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040799175324781164,
      "loss": 5.4608,
      "mean_token_accuracy": 0.18589465618133544,
      "num_tokens": 70674635.0,
      "step": 34880
    },
    {
      "entropy": 5.8088397026062015,
      "epoch": 3.0485886568207636,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000407966616901092,
      "loss": 5.4134,
      "mean_token_accuracy": 0.1904789462685585,
      "num_tokens": 70684633.0,
      "step": 34885
    },
    {
      "entropy": 5.876935291290283,
      "epoch": 3.0490256051734685,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004079414780040029,
      "loss": 5.5043,
      "mean_token_accuracy": 0.1914009615778923,
      "num_tokens": 70694416.0,
      "step": 34890
    },
    {
      "entropy": 5.896354150772095,
      "epoch": 3.049462553526173,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004079163365570267,
      "loss": 5.4205,
      "mean_token_accuracy": 0.18900664150714874,
      "num_tokens": 70703373.0,
      "step": 34895
    },
    {
      "entropy": 5.8201251983642575,
      "epoch": 3.049899501878878,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004078911925606455,
      "loss": 5.4298,
      "mean_token_accuracy": 0.18768406063318252,
      "num_tokens": 70713868.0,
      "step": 34900
    },
    {
      "entropy": 5.930911254882813,
      "epoch": 3.050336450231583,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004078660460153415,
      "loss": 5.5677,
      "mean_token_accuracy": 0.1796730026602745,
      "num_tokens": 70724306.0,
      "step": 34905
    },
    {
      "entropy": 5.923606538772583,
      "epoch": 3.0507733985842873,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040784089692159724,
      "loss": 5.5181,
      "mean_token_accuracy": 0.1862379476428032,
      "num_tokens": 70733923.0,
      "step": 34910
    },
    {
      "entropy": 5.908187198638916,
      "epoch": 3.0512103469369922,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004078157452798949,
      "loss": 5.5259,
      "mean_token_accuracy": 0.17528572231531142,
      "num_tokens": 70744964.0,
      "step": 34915
    },
    {
      "entropy": 5.834770011901855,
      "epoch": 3.0516472952896967,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00040779059109071706,
      "loss": 5.3898,
      "mean_token_accuracy": 0.19343163222074508,
      "num_tokens": 70754639.0,
      "step": 34920
    },
    {
      "entropy": 5.927265596389771,
      "epoch": 3.0520842436424016,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040776543435454597,
      "loss": 5.5257,
      "mean_token_accuracy": 0.18194843977689742,
      "num_tokens": 70764810.0,
      "step": 34925
    },
    {
      "entropy": 5.977007293701172,
      "epoch": 3.052521191995106,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004077402750718643,
      "loss": 5.4906,
      "mean_token_accuracy": 0.1836841583251953,
      "num_tokens": 70774698.0,
      "step": 34930
    },
    {
      "entropy": 5.849209547042847,
      "epoch": 3.052958140347811,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004077151132431546,
      "loss": 5.4059,
      "mean_token_accuracy": 0.19025180339813233,
      "num_tokens": 70785522.0,
      "step": 34935
    },
    {
      "entropy": 5.835637712478638,
      "epoch": 3.0533950887005155,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004076899488688996,
      "loss": 5.5147,
      "mean_token_accuracy": 0.1828167662024498,
      "num_tokens": 70797001.0,
      "step": 34940
    },
    {
      "entropy": 5.859197568893433,
      "epoch": 3.0538320370532204,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040766478194958176,
      "loss": 5.5039,
      "mean_token_accuracy": 0.18497495353221893,
      "num_tokens": 70807247.0,
      "step": 34945
    },
    {
      "entropy": 5.928295755386353,
      "epoch": 3.054268985405925,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004076396124856839,
      "loss": 5.5596,
      "mean_token_accuracy": 0.17170002907514573,
      "num_tokens": 70818269.0,
      "step": 34950
    },
    {
      "entropy": 5.918961095809936,
      "epoch": 3.0547059337586298,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004076144404776888,
      "loss": 5.5243,
      "mean_token_accuracy": 0.18123069256544114,
      "num_tokens": 70827759.0,
      "step": 34955
    },
    {
      "entropy": 5.855829906463623,
      "epoch": 3.0551428821113342,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004075892659260792,
      "loss": 5.4639,
      "mean_token_accuracy": 0.1816246762871742,
      "num_tokens": 70838296.0,
      "step": 34960
    },
    {
      "entropy": 5.869568634033203,
      "epoch": 3.055579830464039,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00040756408883133797,
      "loss": 5.4322,
      "mean_token_accuracy": 0.18416446596384048,
      "num_tokens": 70848364.0,
      "step": 34965
    },
    {
      "entropy": 5.8894655227661135,
      "epoch": 3.056016778816744,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004075389091939481,
      "loss": 5.4608,
      "mean_token_accuracy": 0.1904545545578003,
      "num_tokens": 70858845.0,
      "step": 34970
    },
    {
      "entropy": 5.7828514099121096,
      "epoch": 3.0564537271694485,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004075137270143925,
      "loss": 5.5191,
      "mean_token_accuracy": 0.1940263479948044,
      "num_tokens": 70868976.0,
      "step": 34975
    },
    {
      "entropy": 5.814946794509888,
      "epoch": 3.0568906755221534,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004074885422931542,
      "loss": 5.4256,
      "mean_token_accuracy": 0.19346295595169066,
      "num_tokens": 70878287.0,
      "step": 34980
    },
    {
      "entropy": 5.851599168777466,
      "epoch": 3.057327623874858,
      "grad_norm": 1.421875,
      "learning_rate": 0.00040746335503071625,
      "loss": 5.4514,
      "mean_token_accuracy": 0.19102429002523422,
      "num_tokens": 70889374.0,
      "step": 34985
    },
    {
      "entropy": 5.809772777557373,
      "epoch": 3.057764572227563,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040743816522756165,
      "loss": 5.4636,
      "mean_token_accuracy": 0.19167284667491913,
      "num_tokens": 70899741.0,
      "step": 34990
    },
    {
      "entropy": 5.861270761489868,
      "epoch": 3.0582015205802673,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040741297288417373,
      "loss": 5.5011,
      "mean_token_accuracy": 0.18776473701000213,
      "num_tokens": 70909962.0,
      "step": 34995
    },
    {
      "entropy": 5.930247068405151,
      "epoch": 3.058638468932972,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004073877780010356,
      "loss": 5.4634,
      "mean_token_accuracy": 0.18305508345365523,
      "num_tokens": 70921129.0,
      "step": 35000
    },
    {
      "entropy": 5.846339082717895,
      "epoch": 3.0590754172856767,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004073625805786305,
      "loss": 5.4196,
      "mean_token_accuracy": 0.19005023688077927,
      "num_tokens": 70930492.0,
      "step": 35005
    },
    {
      "entropy": 5.9128814220428465,
      "epoch": 3.0595123656383816,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004073373806174417,
      "loss": 5.465,
      "mean_token_accuracy": 0.18901586085557937,
      "num_tokens": 70940667.0,
      "step": 35010
    },
    {
      "entropy": 5.919613790512085,
      "epoch": 3.059949313991086,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004073121781179527,
      "loss": 5.4261,
      "mean_token_accuracy": 0.1888587534427643,
      "num_tokens": 70949815.0,
      "step": 35015
    },
    {
      "entropy": 5.906631898880005,
      "epoch": 3.060386262343791,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040728697308064674,
      "loss": 5.5642,
      "mean_token_accuracy": 0.17448007613420485,
      "num_tokens": 70959931.0,
      "step": 35020
    },
    {
      "entropy": 5.877073478698731,
      "epoch": 3.060823210696496,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004072617655060073,
      "loss": 5.4167,
      "mean_token_accuracy": 0.18893867135047912,
      "num_tokens": 70970204.0,
      "step": 35025
    },
    {
      "entropy": 5.92132568359375,
      "epoch": 3.0612601590492003,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040723655539451787,
      "loss": 5.5381,
      "mean_token_accuracy": 0.1827617809176445,
      "num_tokens": 70981841.0,
      "step": 35030
    },
    {
      "entropy": 5.866485500335694,
      "epoch": 3.0616971074019053,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004072113427466621,
      "loss": 5.4763,
      "mean_token_accuracy": 0.18325523734092714,
      "num_tokens": 70992281.0,
      "step": 35035
    },
    {
      "entropy": 5.825808334350586,
      "epoch": 3.0621340557546097,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040718612756292347,
      "loss": 5.399,
      "mean_token_accuracy": 0.19346917420625687,
      "num_tokens": 71002737.0,
      "step": 35040
    },
    {
      "entropy": 5.893586206436157,
      "epoch": 3.0625710041073146,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004071609098437856,
      "loss": 5.4752,
      "mean_token_accuracy": 0.18585843443870545,
      "num_tokens": 71013604.0,
      "step": 35045
    },
    {
      "entropy": 5.835874795913696,
      "epoch": 3.063007952460019,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040713568958973225,
      "loss": 5.4499,
      "mean_token_accuracy": 0.18740849196910858,
      "num_tokens": 71023278.0,
      "step": 35050
    },
    {
      "entropy": 5.830988931655884,
      "epoch": 3.063444900812724,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040711046680124715,
      "loss": 5.4878,
      "mean_token_accuracy": 0.18432799428701402,
      "num_tokens": 71033823.0,
      "step": 35055
    },
    {
      "entropy": 5.863263845443726,
      "epoch": 3.0638818491654285,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004070852414788141,
      "loss": 5.4864,
      "mean_token_accuracy": 0.18738189488649368,
      "num_tokens": 71043802.0,
      "step": 35060
    },
    {
      "entropy": 5.933799886703492,
      "epoch": 3.0643187975181334,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004070600136229168,
      "loss": 5.5314,
      "mean_token_accuracy": 0.17922853380441667,
      "num_tokens": 71054152.0,
      "step": 35065
    },
    {
      "entropy": 5.858313131332397,
      "epoch": 3.064755745870838,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040703478323403944,
      "loss": 5.4498,
      "mean_token_accuracy": 0.18769308924674988,
      "num_tokens": 71065346.0,
      "step": 35070
    },
    {
      "entropy": 5.856303071975708,
      "epoch": 3.065192694223543,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040700955031266556,
      "loss": 5.4551,
      "mean_token_accuracy": 0.18548957854509354,
      "num_tokens": 71074855.0,
      "step": 35075
    },
    {
      "entropy": 5.880418920516968,
      "epoch": 3.0656296425762477,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004069843148592794,
      "loss": 5.4367,
      "mean_token_accuracy": 0.1882682591676712,
      "num_tokens": 71084372.0,
      "step": 35080
    },
    {
      "entropy": 5.872654104232788,
      "epoch": 3.066066590928952,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000406959076874365,
      "loss": 5.4796,
      "mean_token_accuracy": 0.18999163955450057,
      "num_tokens": 71093365.0,
      "step": 35085
    },
    {
      "entropy": 5.810344982147217,
      "epoch": 3.066503539281657,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040693383635840625,
      "loss": 5.3907,
      "mean_token_accuracy": 0.19006227254867553,
      "num_tokens": 71102748.0,
      "step": 35090
    },
    {
      "entropy": 5.891453504562378,
      "epoch": 3.0669404876343616,
      "grad_norm": 1.0,
      "learning_rate": 0.0004069085933118874,
      "loss": 5.5016,
      "mean_token_accuracy": 0.18308556079864502,
      "num_tokens": 71113878.0,
      "step": 35095
    },
    {
      "entropy": 5.949921655654907,
      "epoch": 3.0673774359870665,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004068833477352927,
      "loss": 5.559,
      "mean_token_accuracy": 0.18142566084861755,
      "num_tokens": 71124849.0,
      "step": 35100
    },
    {
      "entropy": 5.891891527175903,
      "epoch": 3.067814384339771,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040685809962910614,
      "loss": 5.4439,
      "mean_token_accuracy": 0.18950872421264647,
      "num_tokens": 71134064.0,
      "step": 35105
    },
    {
      "entropy": 5.774101829528808,
      "epoch": 3.068251332692476,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040683284899381225,
      "loss": 5.3337,
      "mean_token_accuracy": 0.1950444057583809,
      "num_tokens": 71144270.0,
      "step": 35110
    },
    {
      "entropy": 5.793408489227295,
      "epoch": 3.0686882810451803,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040680759582989516,
      "loss": 5.3575,
      "mean_token_accuracy": 0.19225072115659714,
      "num_tokens": 71153949.0,
      "step": 35115
    },
    {
      "entropy": 5.859307718276978,
      "epoch": 3.0691252293978852,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00040678234013783945,
      "loss": 5.4774,
      "mean_token_accuracy": 0.1842327207326889,
      "num_tokens": 71164858.0,
      "step": 35120
    },
    {
      "entropy": 5.881254529953003,
      "epoch": 3.0695621777505897,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004067570819181293,
      "loss": 5.3794,
      "mean_token_accuracy": 0.19075289666652678,
      "num_tokens": 71174832.0,
      "step": 35125
    },
    {
      "entropy": 5.826876401901245,
      "epoch": 3.0699991261032946,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004067318211712492,
      "loss": 5.4929,
      "mean_token_accuracy": 0.18648180961608887,
      "num_tokens": 71185089.0,
      "step": 35130
    },
    {
      "entropy": 5.95692458152771,
      "epoch": 3.070436074455999,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004067065578976839,
      "loss": 5.475,
      "mean_token_accuracy": 0.17997392117977143,
      "num_tokens": 71195889.0,
      "step": 35135
    },
    {
      "entropy": 5.913473987579346,
      "epoch": 3.070873022808704,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040668129209791773,
      "loss": 5.5065,
      "mean_token_accuracy": 0.18207223266363143,
      "num_tokens": 71205825.0,
      "step": 35140
    },
    {
      "entropy": 5.795118808746338,
      "epoch": 3.071309971161409,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004066560237724354,
      "loss": 5.3758,
      "mean_token_accuracy": 0.190970578789711,
      "num_tokens": 71215201.0,
      "step": 35145
    },
    {
      "entropy": 5.747096967697144,
      "epoch": 3.0717469195141134,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004066307529217215,
      "loss": 5.3461,
      "mean_token_accuracy": 0.198968006670475,
      "num_tokens": 71225329.0,
      "step": 35150
    },
    {
      "entropy": 5.86203670501709,
      "epoch": 3.0721838678668183,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040660547954626076,
      "loss": 5.3592,
      "mean_token_accuracy": 0.19612170457839967,
      "num_tokens": 71235301.0,
      "step": 35155
    },
    {
      "entropy": 5.834159421920776,
      "epoch": 3.0726208162195228,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040658020364653803,
      "loss": 5.4458,
      "mean_token_accuracy": 0.18469666838645935,
      "num_tokens": 71244991.0,
      "step": 35160
    },
    {
      "entropy": 5.807538414001465,
      "epoch": 3.0730577645722277,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000406554925223038,
      "loss": 5.4415,
      "mean_token_accuracy": 0.18883248865604402,
      "num_tokens": 71255480.0,
      "step": 35165
    },
    {
      "entropy": 5.960793209075928,
      "epoch": 3.073494712924932,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004065296442762456,
      "loss": 5.5743,
      "mean_token_accuracy": 0.17740367650985717,
      "num_tokens": 71265996.0,
      "step": 35170
    },
    {
      "entropy": 5.951339244842529,
      "epoch": 3.073931661277637,
      "grad_norm": 1.0,
      "learning_rate": 0.00040650436080664576,
      "loss": 5.4705,
      "mean_token_accuracy": 0.18343302309513093,
      "num_tokens": 71277293.0,
      "step": 35175
    },
    {
      "entropy": 5.8408478736877445,
      "epoch": 3.0743686096303415,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040647907481472333,
      "loss": 5.3845,
      "mean_token_accuracy": 0.18985203951597213,
      "num_tokens": 71285903.0,
      "step": 35180
    },
    {
      "entropy": 5.7687753200531,
      "epoch": 3.0748055579830464,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004064537863009633,
      "loss": 5.3852,
      "mean_token_accuracy": 0.19397914707660674,
      "num_tokens": 71294979.0,
      "step": 35185
    },
    {
      "entropy": 5.825279903411865,
      "epoch": 3.075242506335751,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040642849526585087,
      "loss": 5.4151,
      "mean_token_accuracy": 0.1972356766462326,
      "num_tokens": 71306223.0,
      "step": 35190
    },
    {
      "entropy": 5.992163515090942,
      "epoch": 3.075679454688456,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004064032017098709,
      "loss": 5.55,
      "mean_token_accuracy": 0.17758143544197083,
      "num_tokens": 71316984.0,
      "step": 35195
    },
    {
      "entropy": 5.931327295303345,
      "epoch": 3.0761164030411607,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040637790563350886,
      "loss": 5.5392,
      "mean_token_accuracy": 0.18824919313192368,
      "num_tokens": 71327855.0,
      "step": 35200
    },
    {
      "entropy": 5.8311848640441895,
      "epoch": 3.076553351393865,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004063526070372496,
      "loss": 5.4281,
      "mean_token_accuracy": 0.1904354363679886,
      "num_tokens": 71338115.0,
      "step": 35205
    },
    {
      "entropy": 5.8081964492797855,
      "epoch": 3.07699029974657,
      "grad_norm": 1.125,
      "learning_rate": 0.0004063273059215785,
      "loss": 5.4118,
      "mean_token_accuracy": 0.18790787160396577,
      "num_tokens": 71348533.0,
      "step": 35210
    },
    {
      "entropy": 5.808724308013916,
      "epoch": 3.0774272480992746,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004063020022869809,
      "loss": 5.2709,
      "mean_token_accuracy": 0.20252985507249832,
      "num_tokens": 71358368.0,
      "step": 35215
    },
    {
      "entropy": 5.878724908828735,
      "epoch": 3.0778641964519795,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040627669613394206,
      "loss": 5.4816,
      "mean_token_accuracy": 0.18611304759979247,
      "num_tokens": 71368443.0,
      "step": 35220
    },
    {
      "entropy": 5.8636805534362795,
      "epoch": 3.078301144804684,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040625138746294747,
      "loss": 5.4868,
      "mean_token_accuracy": 0.17684431970119477,
      "num_tokens": 71378508.0,
      "step": 35225
    },
    {
      "entropy": 5.868913221359253,
      "epoch": 3.078738093157389,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004062260762744825,
      "loss": 5.4044,
      "mean_token_accuracy": 0.19094925075769426,
      "num_tokens": 71388512.0,
      "step": 35230
    },
    {
      "entropy": 5.889388179779052,
      "epoch": 3.0791750415100934,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00040620076256903256,
      "loss": 5.4548,
      "mean_token_accuracy": 0.18293111771345139,
      "num_tokens": 71397908.0,
      "step": 35235
    },
    {
      "entropy": 5.895798683166504,
      "epoch": 3.0796119898627983,
      "grad_norm": 1.0,
      "learning_rate": 0.0004061754463470833,
      "loss": 5.5169,
      "mean_token_accuracy": 0.17885506004095078,
      "num_tokens": 71408287.0,
      "step": 35240
    },
    {
      "entropy": 5.913955783843994,
      "epoch": 3.0800489382155027,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004061501276091202,
      "loss": 5.5309,
      "mean_token_accuracy": 0.18348852843046187,
      "num_tokens": 71419227.0,
      "step": 35245
    },
    {
      "entropy": 5.923487854003906,
      "epoch": 3.0804858865682077,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000406124806355629,
      "loss": 5.4634,
      "mean_token_accuracy": 0.19020348638296128,
      "num_tokens": 71429398.0,
      "step": 35250
    },
    {
      "entropy": 5.817701292037964,
      "epoch": 3.080922834920912,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040609948258709517,
      "loss": 5.4499,
      "mean_token_accuracy": 0.18519534319639205,
      "num_tokens": 71440185.0,
      "step": 35255
    },
    {
      "entropy": 5.783143806457519,
      "epoch": 3.081359783273617,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004060741563040047,
      "loss": 5.3958,
      "mean_token_accuracy": 0.20128493756055832,
      "num_tokens": 71451160.0,
      "step": 35260
    },
    {
      "entropy": 5.94634690284729,
      "epoch": 3.081796731626322,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004060488275068432,
      "loss": 5.5079,
      "mean_token_accuracy": 0.18072105050086976,
      "num_tokens": 71461459.0,
      "step": 35265
    },
    {
      "entropy": 5.846947717666626,
      "epoch": 3.0822336799790264,
      "grad_norm": 1.25,
      "learning_rate": 0.00040602349619609646,
      "loss": 5.4276,
      "mean_token_accuracy": 0.18646549433469772,
      "num_tokens": 71470806.0,
      "step": 35270
    },
    {
      "entropy": 5.768787145614624,
      "epoch": 3.0826706283317313,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040599816237225044,
      "loss": 5.4161,
      "mean_token_accuracy": 0.19090855717658997,
      "num_tokens": 71480929.0,
      "step": 35275
    },
    {
      "entropy": 5.861230897903442,
      "epoch": 3.083107576684436,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040597282603579103,
      "loss": 5.4458,
      "mean_token_accuracy": 0.18850542306900026,
      "num_tokens": 71490360.0,
      "step": 35280
    },
    {
      "entropy": 5.820180177688599,
      "epoch": 3.0835445250371407,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040594748718720424,
      "loss": 5.4998,
      "mean_token_accuracy": 0.1834468886256218,
      "num_tokens": 71500355.0,
      "step": 35285
    },
    {
      "entropy": 5.818678951263427,
      "epoch": 3.083981473389845,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040592214582697595,
      "loss": 5.4288,
      "mean_token_accuracy": 0.19288876950740813,
      "num_tokens": 71509581.0,
      "step": 35290
    },
    {
      "entropy": 5.838703584671021,
      "epoch": 3.08441842174255,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004058968019555923,
      "loss": 5.4852,
      "mean_token_accuracy": 0.1803950324654579,
      "num_tokens": 71519967.0,
      "step": 35295
    },
    {
      "entropy": 5.823473739624023,
      "epoch": 3.0848553700952546,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00040587145557353943,
      "loss": 5.3897,
      "mean_token_accuracy": 0.19313610345125198,
      "num_tokens": 71530619.0,
      "step": 35300
    },
    {
      "entropy": 5.930850172042847,
      "epoch": 3.0852923184479595,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004058461066813035,
      "loss": 5.5157,
      "mean_token_accuracy": 0.18246853947639466,
      "num_tokens": 71541682.0,
      "step": 35305
    },
    {
      "entropy": 5.852738857269287,
      "epoch": 3.085729266800664,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040582075527937066,
      "loss": 5.4246,
      "mean_token_accuracy": 0.18853125125169753,
      "num_tokens": 71552971.0,
      "step": 35310
    },
    {
      "entropy": 5.776291036605835,
      "epoch": 3.086166215153369,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004057954013682272,
      "loss": 5.4383,
      "mean_token_accuracy": 0.18492306470870973,
      "num_tokens": 71562932.0,
      "step": 35315
    },
    {
      "entropy": 5.859170436859131,
      "epoch": 3.0866031635060738,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040577004494835934,
      "loss": 5.3935,
      "mean_token_accuracy": 0.18994652181863786,
      "num_tokens": 71572662.0,
      "step": 35320
    },
    {
      "entropy": 5.853063774108887,
      "epoch": 3.0870401118587782,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00040574468602025355,
      "loss": 5.4018,
      "mean_token_accuracy": 0.1911722481250763,
      "num_tokens": 71582503.0,
      "step": 35325
    },
    {
      "entropy": 5.866520977020263,
      "epoch": 3.087477060211483,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040571932458439617,
      "loss": 5.4733,
      "mean_token_accuracy": 0.18429113924503326,
      "num_tokens": 71592261.0,
      "step": 35330
    },
    {
      "entropy": 5.874917697906494,
      "epoch": 3.0879140085641876,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040569396064127355,
      "loss": 5.4351,
      "mean_token_accuracy": 0.19435689151287078,
      "num_tokens": 71602514.0,
      "step": 35335
    },
    {
      "entropy": 5.808575010299682,
      "epoch": 3.0883509569168925,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004056685941913724,
      "loss": 5.4668,
      "mean_token_accuracy": 0.19165320992469786,
      "num_tokens": 71612912.0,
      "step": 35340
    },
    {
      "entropy": 5.773308897018433,
      "epoch": 3.088787905269597,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040564322523517905,
      "loss": 5.3657,
      "mean_token_accuracy": 0.1928522989153862,
      "num_tokens": 71622673.0,
      "step": 35345
    },
    {
      "entropy": 5.860588884353637,
      "epoch": 3.089224853622302,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004056178537731803,
      "loss": 5.4412,
      "mean_token_accuracy": 0.1828494369983673,
      "num_tokens": 71633315.0,
      "step": 35350
    },
    {
      "entropy": 5.889090538024902,
      "epoch": 3.0896618019750064,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004055924798058626,
      "loss": 5.5085,
      "mean_token_accuracy": 0.1810545116662979,
      "num_tokens": 71644159.0,
      "step": 35355
    },
    {
      "entropy": 5.863624382019043,
      "epoch": 3.0900987503277113,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040556710333371275,
      "loss": 5.4288,
      "mean_token_accuracy": 0.18586842864751815,
      "num_tokens": 71654227.0,
      "step": 35360
    },
    {
      "entropy": 5.8453648567199705,
      "epoch": 3.0905356986804158,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004055417243572174,
      "loss": 5.3965,
      "mean_token_accuracy": 0.19125329852104186,
      "num_tokens": 71664711.0,
      "step": 35365
    },
    {
      "entropy": 5.862816333770752,
      "epoch": 3.0909726470331207,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040551634287686343,
      "loss": 5.4413,
      "mean_token_accuracy": 0.19379470199346543,
      "num_tokens": 71674880.0,
      "step": 35370
    },
    {
      "entropy": 5.886863327026367,
      "epoch": 3.0914095953858256,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004054909588931375,
      "loss": 5.4825,
      "mean_token_accuracy": 0.17882474958896638,
      "num_tokens": 71685386.0,
      "step": 35375
    },
    {
      "entropy": 5.881377077102661,
      "epoch": 3.09184654373853,
      "grad_norm": 1.625,
      "learning_rate": 0.0004054655724065268,
      "loss": 5.5807,
      "mean_token_accuracy": 0.17672258615493774,
      "num_tokens": 71696008.0,
      "step": 35380
    },
    {
      "entropy": 5.8615912914276125,
      "epoch": 3.092283492091235,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004054401834175179,
      "loss": 5.4271,
      "mean_token_accuracy": 0.1822238326072693,
      "num_tokens": 71705548.0,
      "step": 35385
    },
    {
      "entropy": 5.876466274261475,
      "epoch": 3.0927204404439395,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040541479192659797,
      "loss": 5.4637,
      "mean_token_accuracy": 0.18268050402402877,
      "num_tokens": 71716891.0,
      "step": 35390
    },
    {
      "entropy": 5.908016872406006,
      "epoch": 3.0931573887966444,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040538939793425394,
      "loss": 5.5346,
      "mean_token_accuracy": 0.17592580616474152,
      "num_tokens": 71727954.0,
      "step": 35395
    },
    {
      "entropy": 5.868587589263916,
      "epoch": 3.093594337149349,
      "grad_norm": 0.984375,
      "learning_rate": 0.000405364001440973,
      "loss": 5.4769,
      "mean_token_accuracy": 0.18494230806827544,
      "num_tokens": 71738282.0,
      "step": 35400
    },
    {
      "entropy": 5.901454877853394,
      "epoch": 3.0940312855020538,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004053386024472422,
      "loss": 5.5095,
      "mean_token_accuracy": 0.17985223233699799,
      "num_tokens": 71747895.0,
      "step": 35405
    },
    {
      "entropy": 5.851915454864502,
      "epoch": 3.094468233854758,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004053132009535487,
      "loss": 5.4278,
      "mean_token_accuracy": 0.18873768895864487,
      "num_tokens": 71758107.0,
      "step": 35410
    },
    {
      "entropy": 5.912807369232178,
      "epoch": 3.094905182207463,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004052877969603797,
      "loss": 5.5233,
      "mean_token_accuracy": 0.18391186743974686,
      "num_tokens": 71767661.0,
      "step": 35415
    },
    {
      "entropy": 5.874825429916382,
      "epoch": 3.0953421305601676,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004052623904682225,
      "loss": 5.4728,
      "mean_token_accuracy": 0.18574604839086534,
      "num_tokens": 71777908.0,
      "step": 35420
    },
    {
      "entropy": 5.828262615203857,
      "epoch": 3.0957790789128725,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040523698147756443,
      "loss": 5.4324,
      "mean_token_accuracy": 0.18684030920267106,
      "num_tokens": 71786809.0,
      "step": 35425
    },
    {
      "entropy": 5.82130012512207,
      "epoch": 3.0962160272655774,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040521156998889283,
      "loss": 5.4272,
      "mean_token_accuracy": 0.18421821743249894,
      "num_tokens": 71797299.0,
      "step": 35430
    },
    {
      "entropy": 5.89860167503357,
      "epoch": 3.096652975618282,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040518615600269506,
      "loss": 5.4477,
      "mean_token_accuracy": 0.1899898663163185,
      "num_tokens": 71807565.0,
      "step": 35435
    },
    {
      "entropy": 5.788535070419312,
      "epoch": 3.097089923970987,
      "grad_norm": 1.75,
      "learning_rate": 0.0004051607395194586,
      "loss": 5.439,
      "mean_token_accuracy": 0.1851774200797081,
      "num_tokens": 71817288.0,
      "step": 35440
    },
    {
      "entropy": 5.84379825592041,
      "epoch": 3.0975268723236913,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040513532053967086,
      "loss": 5.4294,
      "mean_token_accuracy": 0.18729934245347976,
      "num_tokens": 71827371.0,
      "step": 35445
    },
    {
      "entropy": 5.957208156585693,
      "epoch": 3.097963820676396,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040510989906381955,
      "loss": 5.5656,
      "mean_token_accuracy": 0.18198092579841613,
      "num_tokens": 71837872.0,
      "step": 35450
    },
    {
      "entropy": 5.84923529624939,
      "epoch": 3.0984007690291007,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040508447509239224,
      "loss": 5.391,
      "mean_token_accuracy": 0.19416451305150986,
      "num_tokens": 71848507.0,
      "step": 35455
    },
    {
      "entropy": 5.828889560699463,
      "epoch": 3.0988377173818056,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004050590486258764,
      "loss": 5.4823,
      "mean_token_accuracy": 0.18050789684057236,
      "num_tokens": 71860495.0,
      "step": 35460
    },
    {
      "entropy": 5.8970386505126955,
      "epoch": 3.09927466573451,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00040503361966476005,
      "loss": 5.4382,
      "mean_token_accuracy": 0.19410718977451324,
      "num_tokens": 71870081.0,
      "step": 35465
    },
    {
      "entropy": 5.878225564956665,
      "epoch": 3.099711614087215,
      "grad_norm": 1.0,
      "learning_rate": 0.0004050081882095307,
      "loss": 5.4239,
      "mean_token_accuracy": 0.18616491705179214,
      "num_tokens": 71880072.0,
      "step": 35470
    },
    {
      "entropy": 5.794767570495606,
      "epoch": 3.1001485624399194,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004049827542606761,
      "loss": 5.4144,
      "mean_token_accuracy": 0.18828898668289185,
      "num_tokens": 71889225.0,
      "step": 35475
    },
    {
      "entropy": 5.845514726638794,
      "epoch": 3.1005855107926243,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004049573178186842,
      "loss": 5.4378,
      "mean_token_accuracy": 0.19298908859491348,
      "num_tokens": 71898601.0,
      "step": 35480
    },
    {
      "entropy": 5.906799221038819,
      "epoch": 3.101022459145329,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00040493187888404283,
      "loss": 5.4597,
      "mean_token_accuracy": 0.1852088376879692,
      "num_tokens": 71909267.0,
      "step": 35485
    },
    {
      "entropy": 5.8738672733306885,
      "epoch": 3.1014594074980337,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040490643745724,
      "loss": 5.4755,
      "mean_token_accuracy": 0.1829742670059204,
      "num_tokens": 71919066.0,
      "step": 35490
    },
    {
      "entropy": 5.8627238273620605,
      "epoch": 3.1018963558507386,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004048809935387635,
      "loss": 5.4406,
      "mean_token_accuracy": 0.19029994010925294,
      "num_tokens": 71929067.0,
      "step": 35495
    },
    {
      "entropy": 5.881176900863648,
      "epoch": 3.102333304203443,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040485554712910155,
      "loss": 5.3747,
      "mean_token_accuracy": 0.1878790259361267,
      "num_tokens": 71938044.0,
      "step": 35500
    },
    {
      "entropy": 5.791514205932617,
      "epoch": 3.102770252556148,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040483009822874216,
      "loss": 5.4758,
      "mean_token_accuracy": 0.18812976628541947,
      "num_tokens": 71947714.0,
      "step": 35505
    },
    {
      "entropy": 5.840854358673096,
      "epoch": 3.1032072009088525,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004048046468381735,
      "loss": 5.4039,
      "mean_token_accuracy": 0.19614460021257402,
      "num_tokens": 71957660.0,
      "step": 35510
    },
    {
      "entropy": 5.827871417999267,
      "epoch": 3.1036441492615574,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004047791929578837,
      "loss": 5.3628,
      "mean_token_accuracy": 0.20278186798095704,
      "num_tokens": 71967063.0,
      "step": 35515
    },
    {
      "entropy": 5.771059894561768,
      "epoch": 3.104081097614262,
      "grad_norm": 1.125,
      "learning_rate": 0.0004047537365883609,
      "loss": 5.4743,
      "mean_token_accuracy": 0.19025885462760925,
      "num_tokens": 71977331.0,
      "step": 35520
    },
    {
      "entropy": 5.804490470886231,
      "epoch": 3.104518045966967,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040472827773009346,
      "loss": 5.37,
      "mean_token_accuracy": 0.20070275962352752,
      "num_tokens": 71987047.0,
      "step": 35525
    },
    {
      "entropy": 5.877826356887818,
      "epoch": 3.1049549943196713,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040470281638356974,
      "loss": 5.3762,
      "mean_token_accuracy": 0.18822022527456284,
      "num_tokens": 71997057.0,
      "step": 35530
    },
    {
      "entropy": 5.884373664855957,
      "epoch": 3.105391942672376,
      "grad_norm": 1.09375,
      "learning_rate": 0.000404677352549278,
      "loss": 5.4911,
      "mean_token_accuracy": 0.1840697392821312,
      "num_tokens": 72007092.0,
      "step": 35535
    },
    {
      "entropy": 5.891128063201904,
      "epoch": 3.1058288910250806,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040465188622770663,
      "loss": 5.5305,
      "mean_token_accuracy": 0.18721354901790618,
      "num_tokens": 72016847.0,
      "step": 35540
    },
    {
      "entropy": 5.820114707946777,
      "epoch": 3.1062658393777856,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004046264174193443,
      "loss": 5.4081,
      "mean_token_accuracy": 0.18912160843610765,
      "num_tokens": 72026706.0,
      "step": 35545
    },
    {
      "entropy": 5.9288201332092285,
      "epoch": 3.1067027877304905,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004046009461246792,
      "loss": 5.5588,
      "mean_token_accuracy": 0.1741600975394249,
      "num_tokens": 72036358.0,
      "step": 35550
    },
    {
      "entropy": 5.886380243301391,
      "epoch": 3.107139736083195,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004045754723442001,
      "loss": 5.5276,
      "mean_token_accuracy": 0.17912737876176835,
      "num_tokens": 72049156.0,
      "step": 35555
    },
    {
      "entropy": 5.862275552749634,
      "epoch": 3.1075766844359,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004045499960783955,
      "loss": 5.411,
      "mean_token_accuracy": 0.18819085210561753,
      "num_tokens": 72058986.0,
      "step": 35560
    },
    {
      "entropy": 5.887662172317505,
      "epoch": 3.1080136327886043,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004045245173277543,
      "loss": 5.4279,
      "mean_token_accuracy": 0.19309579730033874,
      "num_tokens": 72069141.0,
      "step": 35565
    },
    {
      "entropy": 5.835042572021484,
      "epoch": 3.1084505811413092,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040449903609276474,
      "loss": 5.3794,
      "mean_token_accuracy": 0.20587885230779648,
      "num_tokens": 72079009.0,
      "step": 35570
    },
    {
      "entropy": 5.853797340393067,
      "epoch": 3.1088875294940137,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004044735523739159,
      "loss": 5.4474,
      "mean_token_accuracy": 0.18504127264022827,
      "num_tokens": 72089551.0,
      "step": 35575
    },
    {
      "entropy": 5.890784454345703,
      "epoch": 3.1093244778467186,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040444806617169655,
      "loss": 5.4199,
      "mean_token_accuracy": 0.1920565962791443,
      "num_tokens": 72098965.0,
      "step": 35580
    },
    {
      "entropy": 5.849766540527344,
      "epoch": 3.109761426199423,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040442257748659537,
      "loss": 5.4706,
      "mean_token_accuracy": 0.1859883949160576,
      "num_tokens": 72109568.0,
      "step": 35585
    },
    {
      "entropy": 5.801813316345215,
      "epoch": 3.110198374552128,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040439708631910136,
      "loss": 5.4463,
      "mean_token_accuracy": 0.19174514710903168,
      "num_tokens": 72119988.0,
      "step": 35590
    },
    {
      "entropy": 5.849910497665405,
      "epoch": 3.1106353229048325,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004043715926697034,
      "loss": 5.528,
      "mean_token_accuracy": 0.18924249857664108,
      "num_tokens": 72130515.0,
      "step": 35595
    },
    {
      "entropy": 5.936064386367798,
      "epoch": 3.1110722712575374,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004043460965388906,
      "loss": 5.4626,
      "mean_token_accuracy": 0.18305448591709136,
      "num_tokens": 72140362.0,
      "step": 35600
    },
    {
      "entropy": 5.875641393661499,
      "epoch": 3.111509219610242,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004043205979271518,
      "loss": 5.4541,
      "mean_token_accuracy": 0.18899966031312943,
      "num_tokens": 72150726.0,
      "step": 35605
    },
    {
      "entropy": 5.866143369674683,
      "epoch": 3.1119461679629468,
      "grad_norm": 1.125,
      "learning_rate": 0.0004042950968349762,
      "loss": 5.4305,
      "mean_token_accuracy": 0.1904611185193062,
      "num_tokens": 72160264.0,
      "step": 35610
    },
    {
      "entropy": 5.841701650619507,
      "epoch": 3.1123831163156517,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004042695932628529,
      "loss": 5.4685,
      "mean_token_accuracy": 0.18492915034294127,
      "num_tokens": 72171407.0,
      "step": 35615
    },
    {
      "entropy": 5.896761751174926,
      "epoch": 3.112820064668356,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004042440872112711,
      "loss": 5.496,
      "mean_token_accuracy": 0.18653163313865662,
      "num_tokens": 72182132.0,
      "step": 35620
    },
    {
      "entropy": 5.855418920516968,
      "epoch": 3.113257013021061,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00040421857868071996,
      "loss": 5.4544,
      "mean_token_accuracy": 0.1820968583226204,
      "num_tokens": 72192129.0,
      "step": 35625
    },
    {
      "entropy": 5.798604154586792,
      "epoch": 3.1136939613737655,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00040419306767168874,
      "loss": 5.4263,
      "mean_token_accuracy": 0.19110185652971268,
      "num_tokens": 72202053.0,
      "step": 35630
    },
    {
      "entropy": 5.893521690368653,
      "epoch": 3.1141309097264704,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004041675541846668,
      "loss": 5.4059,
      "mean_token_accuracy": 0.18940211981534957,
      "num_tokens": 72211352.0,
      "step": 35635
    },
    {
      "entropy": 5.844776916503906,
      "epoch": 3.114567858079175,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040414203822014343,
      "loss": 5.4104,
      "mean_token_accuracy": 0.19314993917942047,
      "num_tokens": 72220816.0,
      "step": 35640
    },
    {
      "entropy": 5.902501249313355,
      "epoch": 3.11500480643188,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040411651977860817,
      "loss": 5.5023,
      "mean_token_accuracy": 0.18199446350336074,
      "num_tokens": 72231318.0,
      "step": 35645
    },
    {
      "entropy": 5.841136312484741,
      "epoch": 3.1154417547845843,
      "grad_norm": 1.125,
      "learning_rate": 0.00040409099886055044,
      "loss": 5.4308,
      "mean_token_accuracy": 0.18875985741615295,
      "num_tokens": 72241035.0,
      "step": 35650
    },
    {
      "entropy": 5.892470455169677,
      "epoch": 3.115878703137289,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004040654754664596,
      "loss": 5.3813,
      "mean_token_accuracy": 0.18797800391912461,
      "num_tokens": 72249967.0,
      "step": 35655
    },
    {
      "entropy": 5.8706605434417725,
      "epoch": 3.1163156514899937,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040403994959682546,
      "loss": 5.4733,
      "mean_token_accuracy": 0.19094394743442536,
      "num_tokens": 72260114.0,
      "step": 35660
    },
    {
      "entropy": 5.877473211288452,
      "epoch": 3.1167525998426986,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040401442125213735,
      "loss": 5.3995,
      "mean_token_accuracy": 0.18755018413066865,
      "num_tokens": 72270257.0,
      "step": 35665
    },
    {
      "entropy": 5.817257833480835,
      "epoch": 3.1171895481954035,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040398889043288507,
      "loss": 5.3783,
      "mean_token_accuracy": 0.19247139990329742,
      "num_tokens": 72279628.0,
      "step": 35670
    },
    {
      "entropy": 5.823062467575073,
      "epoch": 3.117626496548108,
      "grad_norm": 1.125,
      "learning_rate": 0.0004039633571395583,
      "loss": 5.4304,
      "mean_token_accuracy": 0.18707530498504638,
      "num_tokens": 72289493.0,
      "step": 35675
    },
    {
      "entropy": 5.828316640853882,
      "epoch": 3.118063444900813,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040393782137264676,
      "loss": 5.3372,
      "mean_token_accuracy": 0.20060081481933595,
      "num_tokens": 72300007.0,
      "step": 35680
    },
    {
      "entropy": 5.823051643371582,
      "epoch": 3.1185003932535174,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004039122831326402,
      "loss": 5.4094,
      "mean_token_accuracy": 0.19643742889165877,
      "num_tokens": 72310596.0,
      "step": 35685
    },
    {
      "entropy": 5.863248825073242,
      "epoch": 3.1189373416062223,
      "grad_norm": 1.125,
      "learning_rate": 0.0004038867424200285,
      "loss": 5.4806,
      "mean_token_accuracy": 0.18402898460626602,
      "num_tokens": 72320156.0,
      "step": 35690
    },
    {
      "entropy": 5.871545362472534,
      "epoch": 3.1193742899589267,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004038611992353016,
      "loss": 5.4299,
      "mean_token_accuracy": 0.18922303318977357,
      "num_tokens": 72329608.0,
      "step": 35695
    },
    {
      "entropy": 5.810177278518677,
      "epoch": 3.1198112383116317,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004038356535789494,
      "loss": 5.3765,
      "mean_token_accuracy": 0.19252126514911652,
      "num_tokens": 72339565.0,
      "step": 35700
    },
    {
      "entropy": 5.843631982803345,
      "epoch": 3.120248186664336,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00040381010545146174,
      "loss": 5.4964,
      "mean_token_accuracy": 0.1872357115149498,
      "num_tokens": 72349566.0,
      "step": 35705
    },
    {
      "entropy": 5.906213665008545,
      "epoch": 3.120685135017041,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004037845548533289,
      "loss": 5.402,
      "mean_token_accuracy": 0.1897497221827507,
      "num_tokens": 72360366.0,
      "step": 35710
    },
    {
      "entropy": 5.930982637405395,
      "epoch": 3.1211220833697455,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040375900178504066,
      "loss": 5.551,
      "mean_token_accuracy": 0.1859034299850464,
      "num_tokens": 72369893.0,
      "step": 35715
    },
    {
      "entropy": 5.862224388122558,
      "epoch": 3.1215590317224504,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004037334462470874,
      "loss": 5.4659,
      "mean_token_accuracy": 0.18382886201143264,
      "num_tokens": 72380484.0,
      "step": 35720
    },
    {
      "entropy": 5.883748388290405,
      "epoch": 3.1219959800751553,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004037078882399591,
      "loss": 5.3783,
      "mean_token_accuracy": 0.19710464477539064,
      "num_tokens": 72391088.0,
      "step": 35725
    },
    {
      "entropy": 5.86829137802124,
      "epoch": 3.12243292842786,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004036823277641462,
      "loss": 5.4911,
      "mean_token_accuracy": 0.17640205323696137,
      "num_tokens": 72401253.0,
      "step": 35730
    },
    {
      "entropy": 5.868426036834717,
      "epoch": 3.1228698767805647,
      "grad_norm": 1.125,
      "learning_rate": 0.00040365676482013877,
      "loss": 5.3777,
      "mean_token_accuracy": 0.1960926592350006,
      "num_tokens": 72411235.0,
      "step": 35735
    },
    {
      "entropy": 5.849258613586426,
      "epoch": 3.123306825133269,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004036311994084271,
      "loss": 5.4956,
      "mean_token_accuracy": 0.18652531057596206,
      "num_tokens": 72421387.0,
      "step": 35740
    },
    {
      "entropy": 5.850615739822388,
      "epoch": 3.123743773485974,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00040360563152950173,
      "loss": 5.4618,
      "mean_token_accuracy": 0.1851579427719116,
      "num_tokens": 72431431.0,
      "step": 35745
    },
    {
      "entropy": 5.875936651229859,
      "epoch": 3.1241807218386786,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004035800611838529,
      "loss": 5.4505,
      "mean_token_accuracy": 0.19443041384220122,
      "num_tokens": 72441286.0,
      "step": 35750
    },
    {
      "entropy": 5.884691333770752,
      "epoch": 3.1246176701913835,
      "grad_norm": 1.125,
      "learning_rate": 0.00040355448837197115,
      "loss": 5.4665,
      "mean_token_accuracy": 0.1852572739124298,
      "num_tokens": 72451661.0,
      "step": 35755
    },
    {
      "entropy": 5.830942392349243,
      "epoch": 3.125054618544088,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040352891309434693,
      "loss": 5.4491,
      "mean_token_accuracy": 0.18423787504434586,
      "num_tokens": 72462013.0,
      "step": 35760
    },
    {
      "entropy": 5.831071043014527,
      "epoch": 3.125491566896793,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004035033353514708,
      "loss": 5.3998,
      "mean_token_accuracy": 0.18952567130327225,
      "num_tokens": 72472804.0,
      "step": 35765
    },
    {
      "entropy": 5.886786365509034,
      "epoch": 3.1259285152494973,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040347775514383344,
      "loss": 5.4478,
      "mean_token_accuracy": 0.17945829927921295,
      "num_tokens": 72482893.0,
      "step": 35770
    },
    {
      "entropy": 5.83762936592102,
      "epoch": 3.1263654636022022,
      "grad_norm": 1.5625,
      "learning_rate": 0.00040345217247192536,
      "loss": 5.4574,
      "mean_token_accuracy": 0.18942423164844513,
      "num_tokens": 72492882.0,
      "step": 35775
    },
    {
      "entropy": 5.780650091171265,
      "epoch": 3.126802411954907,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004034265873362373,
      "loss": 5.3588,
      "mean_token_accuracy": 0.1967799723148346,
      "num_tokens": 72502322.0,
      "step": 35780
    },
    {
      "entropy": 5.862298917770386,
      "epoch": 3.1272393603076116,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040340099973726,
      "loss": 5.4487,
      "mean_token_accuracy": 0.19084275066852568,
      "num_tokens": 72511401.0,
      "step": 35785
    },
    {
      "entropy": 5.854264163970948,
      "epoch": 3.1276763086603165,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040337540967548427,
      "loss": 5.5011,
      "mean_token_accuracy": 0.18514941036701202,
      "num_tokens": 72522071.0,
      "step": 35790
    },
    {
      "entropy": 5.932526445388794,
      "epoch": 3.128113257013021,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040334981715140086,
      "loss": 5.5507,
      "mean_token_accuracy": 0.18204782605171205,
      "num_tokens": 72532899.0,
      "step": 35795
    },
    {
      "entropy": 5.921161222457886,
      "epoch": 3.128550205365726,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040332422216550067,
      "loss": 5.4971,
      "mean_token_accuracy": 0.1872641324996948,
      "num_tokens": 72543372.0,
      "step": 35800
    },
    {
      "entropy": 5.790947437286377,
      "epoch": 3.1289871537184304,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004032986247182748,
      "loss": 5.4206,
      "mean_token_accuracy": 0.19119456708431243,
      "num_tokens": 72553014.0,
      "step": 35805
    },
    {
      "entropy": 5.770578765869141,
      "epoch": 3.1294241020711353,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040327302481021394,
      "loss": 5.3663,
      "mean_token_accuracy": 0.1992931142449379,
      "num_tokens": 72562281.0,
      "step": 35810
    },
    {
      "entropy": 5.821314716339112,
      "epoch": 3.1298610504238398,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040324742244180937,
      "loss": 5.3851,
      "mean_token_accuracy": 0.1953117460012436,
      "num_tokens": 72571817.0,
      "step": 35815
    },
    {
      "entropy": 5.835093402862549,
      "epoch": 3.1302979987765447,
      "grad_norm": 1.0,
      "learning_rate": 0.00040322181761355194,
      "loss": 5.4148,
      "mean_token_accuracy": 0.19512415528297425,
      "num_tokens": 72582935.0,
      "step": 35820
    },
    {
      "entropy": 5.8475573539733885,
      "epoch": 3.130734947129249,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004031962103259329,
      "loss": 5.4449,
      "mean_token_accuracy": 0.1922880932688713,
      "num_tokens": 72593280.0,
      "step": 35825
    },
    {
      "entropy": 5.792855930328369,
      "epoch": 3.131171895481954,
      "grad_norm": 1.25,
      "learning_rate": 0.0004031706005794434,
      "loss": 5.3986,
      "mean_token_accuracy": 0.19114818572998046,
      "num_tokens": 72603193.0,
      "step": 35830
    },
    {
      "entropy": 5.776505184173584,
      "epoch": 3.1316088438346585,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004031449883745746,
      "loss": 5.4385,
      "mean_token_accuracy": 0.19290281981229782,
      "num_tokens": 72613363.0,
      "step": 35835
    },
    {
      "entropy": 5.863107204437256,
      "epoch": 3.1320457921873635,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00040311937371181774,
      "loss": 5.5843,
      "mean_token_accuracy": 0.17718100994825364,
      "num_tokens": 72624151.0,
      "step": 35840
    },
    {
      "entropy": 5.902103567123413,
      "epoch": 3.1324827405400684,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040309375659166414,
      "loss": 5.3711,
      "mean_token_accuracy": 0.19757779091596603,
      "num_tokens": 72633692.0,
      "step": 35845
    },
    {
      "entropy": 5.822600889205932,
      "epoch": 3.132919688892773,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00040306813701460525,
      "loss": 5.4501,
      "mean_token_accuracy": 0.18791510164737701,
      "num_tokens": 72644016.0,
      "step": 35850
    },
    {
      "entropy": 5.777342176437378,
      "epoch": 3.1333566372454777,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004030425149811324,
      "loss": 5.4548,
      "mean_token_accuracy": 0.19180996865034103,
      "num_tokens": 72654033.0,
      "step": 35855
    },
    {
      "entropy": 5.920959043502807,
      "epoch": 3.133793585598182,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004030168904917369,
      "loss": 5.3957,
      "mean_token_accuracy": 0.1838674306869507,
      "num_tokens": 72663452.0,
      "step": 35860
    },
    {
      "entropy": 5.916935825347901,
      "epoch": 3.134230533950887,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040299126354691043,
      "loss": 5.5163,
      "mean_token_accuracy": 0.18636248409748077,
      "num_tokens": 72674766.0,
      "step": 35865
    },
    {
      "entropy": 5.881098794937134,
      "epoch": 3.1346674823035916,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040296563414714444,
      "loss": 5.4251,
      "mean_token_accuracy": 0.19027300477027892,
      "num_tokens": 72683816.0,
      "step": 35870
    },
    {
      "entropy": 5.789319038391113,
      "epoch": 3.1351044306562965,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004029400022929306,
      "loss": 5.3861,
      "mean_token_accuracy": 0.19673891365528107,
      "num_tokens": 72693604.0,
      "step": 35875
    },
    {
      "entropy": 5.81574010848999,
      "epoch": 3.135541379009001,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004029143679847604,
      "loss": 5.4494,
      "mean_token_accuracy": 0.1897513434290886,
      "num_tokens": 72703936.0,
      "step": 35880
    },
    {
      "entropy": 5.913729429244995,
      "epoch": 3.135978327361706,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004028887312231256,
      "loss": 5.3869,
      "mean_token_accuracy": 0.19369083940982817,
      "num_tokens": 72713646.0,
      "step": 35885
    },
    {
      "entropy": 5.890940523147583,
      "epoch": 3.1364152757144104,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040286309200851793,
      "loss": 5.3841,
      "mean_token_accuracy": 0.19595580101013182,
      "num_tokens": 72723573.0,
      "step": 35890
    },
    {
      "entropy": 5.823127555847168,
      "epoch": 3.1368522240671153,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004028374503414292,
      "loss": 5.4746,
      "mean_token_accuracy": 0.18271387070417405,
      "num_tokens": 72733805.0,
      "step": 35895
    },
    {
      "entropy": 5.809091424942016,
      "epoch": 3.1372891724198197,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004028118062223512,
      "loss": 5.4088,
      "mean_token_accuracy": 0.18584075570106506,
      "num_tokens": 72744254.0,
      "step": 35900
    },
    {
      "entropy": 5.8975952625274655,
      "epoch": 3.1377261207725247,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004027861596517757,
      "loss": 5.4799,
      "mean_token_accuracy": 0.17911557108163834,
      "num_tokens": 72754267.0,
      "step": 35905
    },
    {
      "entropy": 5.717033910751343,
      "epoch": 3.1381630691252296,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004027605106301948,
      "loss": 5.3362,
      "mean_token_accuracy": 0.20857569128274916,
      "num_tokens": 72764598.0,
      "step": 35910
    },
    {
      "entropy": 5.854325437545777,
      "epoch": 3.138600017477934,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040273485915810026,
      "loss": 5.4837,
      "mean_token_accuracy": 0.18674635738134385,
      "num_tokens": 72774145.0,
      "step": 35915
    },
    {
      "entropy": 5.98345422744751,
      "epoch": 3.139036965830639,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040270920523598425,
      "loss": 5.5603,
      "mean_token_accuracy": 0.17322797179222107,
      "num_tokens": 72783766.0,
      "step": 35920
    },
    {
      "entropy": 5.856675863265991,
      "epoch": 3.1394739141833434,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004026835488643387,
      "loss": 5.4347,
      "mean_token_accuracy": 0.19477100819349288,
      "num_tokens": 72793907.0,
      "step": 35925
    },
    {
      "entropy": 5.8837562084198,
      "epoch": 3.1399108625360483,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040265789004365586,
      "loss": 5.5441,
      "mean_token_accuracy": 0.1777719646692276,
      "num_tokens": 72802927.0,
      "step": 35930
    },
    {
      "entropy": 5.779984188079834,
      "epoch": 3.140347810888753,
      "grad_norm": 1.125,
      "learning_rate": 0.0004026322287744277,
      "loss": 5.3055,
      "mean_token_accuracy": 0.20318931043148042,
      "num_tokens": 72813387.0,
      "step": 35935
    },
    {
      "entropy": 5.860019159317017,
      "epoch": 3.1407847592414577,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004026065650571466,
      "loss": 5.3789,
      "mean_token_accuracy": 0.1905522018671036,
      "num_tokens": 72824109.0,
      "step": 35940
    },
    {
      "entropy": 5.841941928863525,
      "epoch": 3.141221707594162,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004025808988923047,
      "loss": 5.3986,
      "mean_token_accuracy": 0.18848884105682373,
      "num_tokens": 72834107.0,
      "step": 35945
    },
    {
      "entropy": 5.856906414031982,
      "epoch": 3.141658655946867,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040255523028039433,
      "loss": 5.4163,
      "mean_token_accuracy": 0.19228577017784118,
      "num_tokens": 72843726.0,
      "step": 35950
    },
    {
      "entropy": 5.897414445877075,
      "epoch": 3.1420956042995716,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040252955922190776,
      "loss": 5.4902,
      "mean_token_accuracy": 0.18848942816257477,
      "num_tokens": 72854203.0,
      "step": 35955
    },
    {
      "entropy": 5.854227733612061,
      "epoch": 3.1425325526522765,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040250388571733744,
      "loss": 5.4303,
      "mean_token_accuracy": 0.18690014481544495,
      "num_tokens": 72864689.0,
      "step": 35960
    },
    {
      "entropy": 5.799973058700561,
      "epoch": 3.1429695010049814,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004024782097671758,
      "loss": 5.4539,
      "mean_token_accuracy": 0.19120018184185028,
      "num_tokens": 72875040.0,
      "step": 35965
    },
    {
      "entropy": 5.807091140747071,
      "epoch": 3.143406449357686,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004024525313719153,
      "loss": 5.4154,
      "mean_token_accuracy": 0.19494519382715225,
      "num_tokens": 72885174.0,
      "step": 35970
    },
    {
      "entropy": 5.8590374946594235,
      "epoch": 3.143843397710391,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040242685053204844,
      "loss": 5.4126,
      "mean_token_accuracy": 0.19475630670785904,
      "num_tokens": 72895344.0,
      "step": 35975
    },
    {
      "entropy": 5.897257137298584,
      "epoch": 3.1442803460630953,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004024011672480678,
      "loss": 5.4906,
      "mean_token_accuracy": 0.18921309858560562,
      "num_tokens": 72905950.0,
      "step": 35980
    },
    {
      "entropy": 5.8964169979095455,
      "epoch": 3.1447172944158,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040237548152046603,
      "loss": 5.4309,
      "mean_token_accuracy": 0.1892562821507454,
      "num_tokens": 72916828.0,
      "step": 35985
    },
    {
      "entropy": 5.868967342376709,
      "epoch": 3.1451542427685046,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00040234979334973577,
      "loss": 5.488,
      "mean_token_accuracy": 0.19181353449821473,
      "num_tokens": 72926951.0,
      "step": 35990
    },
    {
      "entropy": 5.8276783466339115,
      "epoch": 3.1455911911212096,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004023241027363698,
      "loss": 5.3736,
      "mean_token_accuracy": 0.19330652207136154,
      "num_tokens": 72937909.0,
      "step": 35995
    },
    {
      "entropy": 5.884532690048218,
      "epoch": 3.146028139473914,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004022984096808608,
      "loss": 5.4624,
      "mean_token_accuracy": 0.18094021677970887,
      "num_tokens": 72949898.0,
      "step": 36000
    },
    {
      "epoch": 3.146028139473914,
      "eval_entropy": 5.617149824917845,
      "eval_loss": 5.519400119781494,
      "eval_mean_token_accuracy": 0.19237325754502124,
      "eval_num_tokens": 72949898.0,
      "eval_runtime": 28.3609,
      "eval_samples_per_second": 1305.108,
      "eval_steps_per_second": 163.147,
      "step": 36000
    },
    {
      "entropy": 5.871783685684204,
      "epoch": 3.146465087826619,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040227271418370156,
      "loss": 5.4613,
      "mean_token_accuracy": 0.1893881857395172,
      "num_tokens": 72959643.0,
      "step": 36005
    },
    {
      "entropy": 5.829525518417358,
      "epoch": 3.1469020361793234,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000402247016245385,
      "loss": 5.438,
      "mean_token_accuracy": 0.18668801486492156,
      "num_tokens": 72969745.0,
      "step": 36010
    },
    {
      "entropy": 5.899714851379395,
      "epoch": 3.1473389845320283,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040222131586640397,
      "loss": 5.4111,
      "mean_token_accuracy": 0.19071101248264313,
      "num_tokens": 72979892.0,
      "step": 36015
    },
    {
      "entropy": 5.799844789505005,
      "epoch": 3.1477759328847332,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040219561304725147,
      "loss": 5.4237,
      "mean_token_accuracy": 0.19662487506866455,
      "num_tokens": 72989526.0,
      "step": 36020
    },
    {
      "entropy": 5.785911750793457,
      "epoch": 3.1482128812374377,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040216990778842046,
      "loss": 5.4581,
      "mean_token_accuracy": 0.18913525342941284,
      "num_tokens": 72999771.0,
      "step": 36025
    },
    {
      "entropy": 5.914534711837769,
      "epoch": 3.1486498295901426,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040214420009040394,
      "loss": 5.4422,
      "mean_token_accuracy": 0.18669335395097733,
      "num_tokens": 73011602.0,
      "step": 36030
    },
    {
      "entropy": 5.9396764755249025,
      "epoch": 3.149086777942847,
      "grad_norm": 0.984375,
      "learning_rate": 0.000402118489953695,
      "loss": 5.4537,
      "mean_token_accuracy": 0.19270148426294326,
      "num_tokens": 73021838.0,
      "step": 36035
    },
    {
      "entropy": 5.876261758804321,
      "epoch": 3.149523726295552,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040209277737878683,
      "loss": 5.441,
      "mean_token_accuracy": 0.18790773153305054,
      "num_tokens": 73032242.0,
      "step": 36040
    },
    {
      "entropy": 5.885119152069092,
      "epoch": 3.1499606746482565,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004020670623661727,
      "loss": 5.5234,
      "mean_token_accuracy": 0.17745808511972427,
      "num_tokens": 73042903.0,
      "step": 36045
    },
    {
      "entropy": 5.845907354354859,
      "epoch": 3.1503976230009614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004020413449163457,
      "loss": 5.3854,
      "mean_token_accuracy": 0.1966857358813286,
      "num_tokens": 73054083.0,
      "step": 36050
    },
    {
      "entropy": 5.975423860549927,
      "epoch": 3.150834571353666,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004020156250297991,
      "loss": 5.4891,
      "mean_token_accuracy": 0.18366118967533113,
      "num_tokens": 73064215.0,
      "step": 36055
    },
    {
      "entropy": 5.765280389785767,
      "epoch": 3.1512715197063708,
      "grad_norm": 1.25,
      "learning_rate": 0.0004019899027070262,
      "loss": 5.3853,
      "mean_token_accuracy": 0.201210056245327,
      "num_tokens": 73074319.0,
      "step": 36060
    },
    {
      "entropy": 5.826645755767823,
      "epoch": 3.1517084680590752,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004019641779485206,
      "loss": 5.5021,
      "mean_token_accuracy": 0.1811123088002205,
      "num_tokens": 73083972.0,
      "step": 36065
    },
    {
      "entropy": 5.846661996841431,
      "epoch": 3.15214541641178,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040193845075477537,
      "loss": 5.467,
      "mean_token_accuracy": 0.1835094451904297,
      "num_tokens": 73094495.0,
      "step": 36070
    },
    {
      "entropy": 5.818940877914429,
      "epoch": 3.152582364764485,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004019127211262842,
      "loss": 5.3115,
      "mean_token_accuracy": 0.20215406715869905,
      "num_tokens": 73104355.0,
      "step": 36075
    },
    {
      "entropy": 5.795654916763306,
      "epoch": 3.1530193131171895,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040188698906354055,
      "loss": 5.4316,
      "mean_token_accuracy": 0.18712959438562393,
      "num_tokens": 73115795.0,
      "step": 36080
    },
    {
      "entropy": 5.859730386734009,
      "epoch": 3.1534562614698944,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004018612545670379,
      "loss": 5.3396,
      "mean_token_accuracy": 0.19802376627922058,
      "num_tokens": 73125351.0,
      "step": 36085
    },
    {
      "entropy": 5.806928205490112,
      "epoch": 3.153893209822599,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040183551763727,
      "loss": 5.4044,
      "mean_token_accuracy": 0.19477193355560302,
      "num_tokens": 73135087.0,
      "step": 36090
    },
    {
      "entropy": 5.756876277923584,
      "epoch": 3.154330158175304,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040180977827473035,
      "loss": 5.4083,
      "mean_token_accuracy": 0.18655415773391723,
      "num_tokens": 73145382.0,
      "step": 36095
    },
    {
      "entropy": 5.898734188079834,
      "epoch": 3.1547671065280083,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00040178403647991276,
      "loss": 5.527,
      "mean_token_accuracy": 0.17898927628993988,
      "num_tokens": 73155689.0,
      "step": 36100
    },
    {
      "entropy": 5.966485452651978,
      "epoch": 3.155204054880713,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004017582922533109,
      "loss": 5.4655,
      "mean_token_accuracy": 0.1802283450961113,
      "num_tokens": 73165803.0,
      "step": 36105
    },
    {
      "entropy": 5.888410663604736,
      "epoch": 3.1556410032334177,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040173254559541853,
      "loss": 5.4603,
      "mean_token_accuracy": 0.18909046947956085,
      "num_tokens": 73176029.0,
      "step": 36110
    },
    {
      "entropy": 5.813517713546753,
      "epoch": 3.1560779515861226,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004017067965067296,
      "loss": 5.4584,
      "mean_token_accuracy": 0.1917307823896408,
      "num_tokens": 73187708.0,
      "step": 36115
    },
    {
      "entropy": 5.851143264770508,
      "epoch": 3.156514899938827,
      "grad_norm": 1.125,
      "learning_rate": 0.0004016810449877379,
      "loss": 5.4468,
      "mean_token_accuracy": 0.184252567589283,
      "num_tokens": 73197458.0,
      "step": 36120
    },
    {
      "entropy": 5.868219470977783,
      "epoch": 3.156951848291532,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004016552910389373,
      "loss": 5.4333,
      "mean_token_accuracy": 0.18814577758312226,
      "num_tokens": 73206931.0,
      "step": 36125
    },
    {
      "entropy": 5.825252342224121,
      "epoch": 3.157388796644237,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040162953466082194,
      "loss": 5.4344,
      "mean_token_accuracy": 0.19071153700351715,
      "num_tokens": 73217048.0,
      "step": 36130
    },
    {
      "entropy": 5.806840658187866,
      "epoch": 3.1578257449969414,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004016037758538857,
      "loss": 5.4344,
      "mean_token_accuracy": 0.1875745102763176,
      "num_tokens": 73226649.0,
      "step": 36135
    },
    {
      "entropy": 5.849632358551025,
      "epoch": 3.1582626933496463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004015780146186227,
      "loss": 5.3653,
      "mean_token_accuracy": 0.1972366601228714,
      "num_tokens": 73236822.0,
      "step": 36140
    },
    {
      "entropy": 5.924367809295655,
      "epoch": 3.1586996417023507,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004015522509555271,
      "loss": 5.5329,
      "mean_token_accuracy": 0.18512776494026184,
      "num_tokens": 73247078.0,
      "step": 36145
    },
    {
      "entropy": 5.752648782730103,
      "epoch": 3.1591365900550556,
      "grad_norm": 1.140625,
      "learning_rate": 0.000401526484865093,
      "loss": 5.2716,
      "mean_token_accuracy": 0.20524507910013198,
      "num_tokens": 73257001.0,
      "step": 36150
    },
    {
      "entropy": 5.766736459732056,
      "epoch": 3.15957353840776,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004015007163478146,
      "loss": 5.3613,
      "mean_token_accuracy": 0.19027575254440307,
      "num_tokens": 73266922.0,
      "step": 36155
    },
    {
      "entropy": 5.858849954605103,
      "epoch": 3.160010486760465,
      "grad_norm": 0.96875,
      "learning_rate": 0.00040147494540418617,
      "loss": 5.4319,
      "mean_token_accuracy": 0.18907323479652405,
      "num_tokens": 73277858.0,
      "step": 36160
    },
    {
      "entropy": 5.8421402931213375,
      "epoch": 3.1604474351131695,
      "grad_norm": 1.0,
      "learning_rate": 0.00040144917203470203,
      "loss": 5.4291,
      "mean_token_accuracy": 0.19131535589694976,
      "num_tokens": 73287361.0,
      "step": 36165
    },
    {
      "entropy": 5.832982778549194,
      "epoch": 3.1608843834658744,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004014233962398565,
      "loss": 5.3662,
      "mean_token_accuracy": 0.19357559978961944,
      "num_tokens": 73296521.0,
      "step": 36170
    },
    {
      "entropy": 5.805889749526978,
      "epoch": 3.161321331818579,
      "grad_norm": 1.046875,
      "learning_rate": 0.000401397618020144,
      "loss": 5.382,
      "mean_token_accuracy": 0.19540658444166184,
      "num_tokens": 73306257.0,
      "step": 36175
    },
    {
      "entropy": 5.790313100814819,
      "epoch": 3.161758280171284,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000401371837376059,
      "loss": 5.3468,
      "mean_token_accuracy": 0.19507648199796676,
      "num_tokens": 73315939.0,
      "step": 36180
    },
    {
      "entropy": 5.8079413890838625,
      "epoch": 3.1621952285239883,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004013460543080959,
      "loss": 5.3993,
      "mean_token_accuracy": 0.1893674299120903,
      "num_tokens": 73326407.0,
      "step": 36185
    },
    {
      "entropy": 5.8535205841064455,
      "epoch": 3.162632176876693,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004013202688167493,
      "loss": 5.437,
      "mean_token_accuracy": 0.18583192229270934,
      "num_tokens": 73336188.0,
      "step": 36190
    },
    {
      "entropy": 5.91915283203125,
      "epoch": 3.163069125229398,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004012944809025136,
      "loss": 5.466,
      "mean_token_accuracy": 0.18343381732702255,
      "num_tokens": 73346824.0,
      "step": 36195
    },
    {
      "entropy": 5.842930889129638,
      "epoch": 3.1635060735821026,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040126869056588375,
      "loss": 5.4191,
      "mean_token_accuracy": 0.1893055647611618,
      "num_tokens": 73356688.0,
      "step": 36200
    },
    {
      "entropy": 5.854706335067749,
      "epoch": 3.1639430219348075,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004012428978073543,
      "loss": 5.5194,
      "mean_token_accuracy": 0.182698355615139,
      "num_tokens": 73366905.0,
      "step": 36205
    },
    {
      "entropy": 5.921902465820312,
      "epoch": 3.164379970287512,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040121710262741974,
      "loss": 5.5258,
      "mean_token_accuracy": 0.17652927786111833,
      "num_tokens": 73376870.0,
      "step": 36210
    },
    {
      "entropy": 5.804768753051758,
      "epoch": 3.164816918640217,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004011913050265751,
      "loss": 5.3818,
      "mean_token_accuracy": 0.1895315632224083,
      "num_tokens": 73386153.0,
      "step": 36215
    },
    {
      "entropy": 5.806338977813721,
      "epoch": 3.1652538669929213,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040116550500531506,
      "loss": 5.4549,
      "mean_token_accuracy": 0.18664862364530563,
      "num_tokens": 73395598.0,
      "step": 36220
    },
    {
      "entropy": 5.875331449508667,
      "epoch": 3.1656908153456262,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040113970256413465,
      "loss": 5.4072,
      "mean_token_accuracy": 0.1881243348121643,
      "num_tokens": 73406883.0,
      "step": 36225
    },
    {
      "entropy": 5.842551851272583,
      "epoch": 3.1661277636983307,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00040111389770352846,
      "loss": 5.4298,
      "mean_token_accuracy": 0.1982238456606865,
      "num_tokens": 73418504.0,
      "step": 36230
    },
    {
      "entropy": 5.843555021286011,
      "epoch": 3.1665647120510356,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00040108809042399183,
      "loss": 5.4428,
      "mean_token_accuracy": 0.18668546676635742,
      "num_tokens": 73429228.0,
      "step": 36235
    },
    {
      "entropy": 5.804152202606201,
      "epoch": 3.16700166040374,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004010622807260194,
      "loss": 5.3933,
      "mean_token_accuracy": 0.193875952064991,
      "num_tokens": 73439269.0,
      "step": 36240
    },
    {
      "entropy": 5.828520965576172,
      "epoch": 3.167438608756445,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004010364686101064,
      "loss": 5.4481,
      "mean_token_accuracy": 0.18732474595308304,
      "num_tokens": 73449554.0,
      "step": 36245
    },
    {
      "entropy": 5.830991601943969,
      "epoch": 3.1678755571091495,
      "grad_norm": 1.046875,
      "learning_rate": 0.000401010654076748,
      "loss": 5.4016,
      "mean_token_accuracy": 0.19742897748947144,
      "num_tokens": 73460453.0,
      "step": 36250
    },
    {
      "entropy": 5.860224294662475,
      "epoch": 3.1683125054618544,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004009848371264391,
      "loss": 5.3499,
      "mean_token_accuracy": 0.19428931325674056,
      "num_tokens": 73469905.0,
      "step": 36255
    },
    {
      "entropy": 5.817387866973877,
      "epoch": 3.1687494538145593,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040095901775967507,
      "loss": 5.5193,
      "mean_token_accuracy": 0.18399060964584352,
      "num_tokens": 73480363.0,
      "step": 36260
    },
    {
      "entropy": 5.852455186843872,
      "epoch": 3.1691864021672638,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040093319597695105,
      "loss": 5.4475,
      "mean_token_accuracy": 0.19178780168294907,
      "num_tokens": 73490080.0,
      "step": 36265
    },
    {
      "entropy": 5.953781223297119,
      "epoch": 3.1696233505199687,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004009073717787624,
      "loss": 5.5002,
      "mean_token_accuracy": 0.1815382406115532,
      "num_tokens": 73500464.0,
      "step": 36270
    },
    {
      "entropy": 5.800949764251709,
      "epoch": 3.170060298872673,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004008815451656043,
      "loss": 5.4186,
      "mean_token_accuracy": 0.19171562790870667,
      "num_tokens": 73509834.0,
      "step": 36275
    },
    {
      "entropy": 5.83073239326477,
      "epoch": 3.170497247225378,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040085571613797226,
      "loss": 5.5188,
      "mean_token_accuracy": 0.17625172585248947,
      "num_tokens": 73520345.0,
      "step": 36280
    },
    {
      "entropy": 5.882139205932617,
      "epoch": 3.1709341955780825,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004008298846963617,
      "loss": 5.4759,
      "mean_token_accuracy": 0.18977874368429185,
      "num_tokens": 73530769.0,
      "step": 36285
    },
    {
      "entropy": 5.909966278076172,
      "epoch": 3.1713711439307875,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004008040508412678,
      "loss": 5.4103,
      "mean_token_accuracy": 0.18235939890146255,
      "num_tokens": 73539989.0,
      "step": 36290
    },
    {
      "entropy": 5.865775632858276,
      "epoch": 3.171808092283492,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004007782145731865,
      "loss": 5.4585,
      "mean_token_accuracy": 0.1839989870786667,
      "num_tokens": 73551344.0,
      "step": 36295
    },
    {
      "entropy": 5.802339172363281,
      "epoch": 3.172245040636197,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040075237589261303,
      "loss": 5.3729,
      "mean_token_accuracy": 0.19287455677986146,
      "num_tokens": 73560580.0,
      "step": 36300
    },
    {
      "entropy": 5.817125225067139,
      "epoch": 3.1726819889889013,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004007265348000432,
      "loss": 5.395,
      "mean_token_accuracy": 0.1842587947845459,
      "num_tokens": 73569500.0,
      "step": 36305
    },
    {
      "entropy": 5.887022542953491,
      "epoch": 3.173118937341606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040070069129597243,
      "loss": 5.4961,
      "mean_token_accuracy": 0.1828480690717697,
      "num_tokens": 73578658.0,
      "step": 36310
    },
    {
      "entropy": 5.829466819763184,
      "epoch": 3.173555885694311,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040067484538089653,
      "loss": 5.411,
      "mean_token_accuracy": 0.19207530170679094,
      "num_tokens": 73588842.0,
      "step": 36315
    },
    {
      "entropy": 5.883447504043579,
      "epoch": 3.1739928340470156,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004006489970553114,
      "loss": 5.4739,
      "mean_token_accuracy": 0.18624911159276963,
      "num_tokens": 73598962.0,
      "step": 36320
    },
    {
      "entropy": 5.8783985614776615,
      "epoch": 3.1744297823997205,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040062314631971255,
      "loss": 5.491,
      "mean_token_accuracy": 0.18129534423351287,
      "num_tokens": 73609890.0,
      "step": 36325
    },
    {
      "entropy": 5.8336437225341795,
      "epoch": 3.174866730752425,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004005972931745959,
      "loss": 5.3985,
      "mean_token_accuracy": 0.19030542075634002,
      "num_tokens": 73620453.0,
      "step": 36330
    },
    {
      "entropy": 5.843562269210816,
      "epoch": 3.17530367910513,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040057143762045746,
      "loss": 5.4284,
      "mean_token_accuracy": 0.1888924479484558,
      "num_tokens": 73630385.0,
      "step": 36335
    },
    {
      "entropy": 5.864139556884766,
      "epoch": 3.1757406274578344,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040054557965779294,
      "loss": 5.4511,
      "mean_token_accuracy": 0.18499307334423065,
      "num_tokens": 73640081.0,
      "step": 36340
    },
    {
      "entropy": 5.854496192932129,
      "epoch": 3.1761775758105393,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00040051971928709853,
      "loss": 5.4363,
      "mean_token_accuracy": 0.1853249728679657,
      "num_tokens": 73650934.0,
      "step": 36345
    },
    {
      "entropy": 5.799710416793824,
      "epoch": 3.1766145241632437,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00040049385650887004,
      "loss": 5.4161,
      "mean_token_accuracy": 0.188490791618824,
      "num_tokens": 73661835.0,
      "step": 36350
    },
    {
      "entropy": 5.874617862701416,
      "epoch": 3.1770514725159487,
      "grad_norm": 1.0,
      "learning_rate": 0.0004004679913236036,
      "loss": 5.4662,
      "mean_token_accuracy": 0.1872565492987633,
      "num_tokens": 73671738.0,
      "step": 36355
    },
    {
      "entropy": 5.904565477371216,
      "epoch": 3.177488420868653,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040044212373179547,
      "loss": 5.4771,
      "mean_token_accuracy": 0.18387358486652375,
      "num_tokens": 73681834.0,
      "step": 36360
    },
    {
      "entropy": 5.842971420288086,
      "epoch": 3.177925369221358,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040041625373394153,
      "loss": 5.4137,
      "mean_token_accuracy": 0.18949237167835237,
      "num_tokens": 73692137.0,
      "step": 36365
    },
    {
      "entropy": 5.840159368515015,
      "epoch": 3.178362317574063,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004003903813305383,
      "loss": 5.4813,
      "mean_token_accuracy": 0.18105137199163437,
      "num_tokens": 73702263.0,
      "step": 36370
    },
    {
      "entropy": 5.8602252960205075,
      "epoch": 3.1787992659267674,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004003645065220817,
      "loss": 5.4875,
      "mean_token_accuracy": 0.17990028560161592,
      "num_tokens": 73711945.0,
      "step": 36375
    },
    {
      "entropy": 5.97814974784851,
      "epoch": 3.1792362142794723,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004003386293090682,
      "loss": 5.5144,
      "mean_token_accuracy": 0.18343086689710617,
      "num_tokens": 73722205.0,
      "step": 36380
    },
    {
      "entropy": 5.877620363235474,
      "epoch": 3.179673162632177,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004003127496919942,
      "loss": 5.4941,
      "mean_token_accuracy": 0.18424209207296371,
      "num_tokens": 73732049.0,
      "step": 36385
    },
    {
      "entropy": 5.829897069931031,
      "epoch": 3.1801101109848817,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004002868676713559,
      "loss": 5.484,
      "mean_token_accuracy": 0.1852597862482071,
      "num_tokens": 73741559.0,
      "step": 36390
    },
    {
      "entropy": 5.897328519821167,
      "epoch": 3.180547059337586,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040026098324764985,
      "loss": 5.5534,
      "mean_token_accuracy": 0.18329022824764252,
      "num_tokens": 73751576.0,
      "step": 36395
    },
    {
      "entropy": 5.933443403244018,
      "epoch": 3.180984007690291,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004002350964213725,
      "loss": 5.5404,
      "mean_token_accuracy": 0.18242388665676118,
      "num_tokens": 73762155.0,
      "step": 36400
    },
    {
      "entropy": 5.8557204246521,
      "epoch": 3.1814209560429956,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004002092071930204,
      "loss": 5.4182,
      "mean_token_accuracy": 0.18378239423036574,
      "num_tokens": 73772427.0,
      "step": 36405
    },
    {
      "entropy": 5.8316703796386715,
      "epoch": 3.1818579043957005,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004001833155630901,
      "loss": 5.3901,
      "mean_token_accuracy": 0.19545504599809646,
      "num_tokens": 73782868.0,
      "step": 36410
    },
    {
      "entropy": 5.792856073379516,
      "epoch": 3.182294852748405,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00040015742153207823,
      "loss": 5.4086,
      "mean_token_accuracy": 0.19503268748521804,
      "num_tokens": 73792480.0,
      "step": 36415
    },
    {
      "entropy": 5.870983362197876,
      "epoch": 3.18273180110111,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004001315251004814,
      "loss": 5.4639,
      "mean_token_accuracy": 0.184905306994915,
      "num_tokens": 73803264.0,
      "step": 36420
    },
    {
      "entropy": 5.838406038284302,
      "epoch": 3.183168749453815,
      "grad_norm": 1.125,
      "learning_rate": 0.00040010562626879636,
      "loss": 5.3228,
      "mean_token_accuracy": 0.19821180850267411,
      "num_tokens": 73812491.0,
      "step": 36425
    },
    {
      "entropy": 5.84129958152771,
      "epoch": 3.1836056978065193,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040007972503751984,
      "loss": 5.4011,
      "mean_token_accuracy": 0.1877223700284958,
      "num_tokens": 73821672.0,
      "step": 36430
    },
    {
      "entropy": 5.85011134147644,
      "epoch": 3.184042646159224,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040005382140714867,
      "loss": 5.3955,
      "mean_token_accuracy": 0.1932191953063011,
      "num_tokens": 73831709.0,
      "step": 36435
    },
    {
      "entropy": 5.806841897964477,
      "epoch": 3.1844795945119286,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004000279153781797,
      "loss": 5.4439,
      "mean_token_accuracy": 0.18317051082849503,
      "num_tokens": 73841121.0,
      "step": 36440
    },
    {
      "entropy": 5.837855005264283,
      "epoch": 3.1849165428646335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040000200695110977,
      "loss": 5.4756,
      "mean_token_accuracy": 0.18285713493824005,
      "num_tokens": 73851299.0,
      "step": 36445
    },
    {
      "entropy": 5.933409357070923,
      "epoch": 3.185353491217338,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003999760961264359,
      "loss": 5.4826,
      "mean_token_accuracy": 0.1844177544116974,
      "num_tokens": 73862020.0,
      "step": 36450
    },
    {
      "entropy": 5.842948770523071,
      "epoch": 3.185790439570043,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00039995018290465495,
      "loss": 5.4524,
      "mean_token_accuracy": 0.18225977271795274,
      "num_tokens": 73872786.0,
      "step": 36455
    },
    {
      "entropy": 5.790694761276245,
      "epoch": 3.1862273879227474,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003999242672862641,
      "loss": 5.3733,
      "mean_token_accuracy": 0.18996161967515945,
      "num_tokens": 73882303.0,
      "step": 36460
    },
    {
      "entropy": 5.800373506546021,
      "epoch": 3.1866643362754523,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003998983492717602,
      "loss": 5.3942,
      "mean_token_accuracy": 0.19621922075748444,
      "num_tokens": 73892124.0,
      "step": 36465
    },
    {
      "entropy": 5.795248365402221,
      "epoch": 3.187101284628157,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003998724288616406,
      "loss": 5.3822,
      "mean_token_accuracy": 0.19776120632886887,
      "num_tokens": 73901803.0,
      "step": 36470
    },
    {
      "entropy": 5.813984155654907,
      "epoch": 3.1875382329808617,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039984650605640244,
      "loss": 5.3305,
      "mean_token_accuracy": 0.19902557879686356,
      "num_tokens": 73911767.0,
      "step": 36475
    },
    {
      "entropy": 5.868917798995971,
      "epoch": 3.187975181333566,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003998205808565428,
      "loss": 5.5091,
      "mean_token_accuracy": 0.17924338281154634,
      "num_tokens": 73921746.0,
      "step": 36480
    },
    {
      "entropy": 5.763257074356079,
      "epoch": 3.188412129686271,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039979465326255904,
      "loss": 5.4086,
      "mean_token_accuracy": 0.1917228013277054,
      "num_tokens": 73932282.0,
      "step": 36485
    },
    {
      "entropy": 5.899643230438232,
      "epoch": 3.188849078038976,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003997687232749484,
      "loss": 5.4988,
      "mean_token_accuracy": 0.1837580382823944,
      "num_tokens": 73942739.0,
      "step": 36490
    },
    {
      "entropy": 5.839309310913086,
      "epoch": 3.1892860263916805,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039974279089420833,
      "loss": 5.4468,
      "mean_token_accuracy": 0.19092161655426027,
      "num_tokens": 73953739.0,
      "step": 36495
    },
    {
      "entropy": 5.868335628509522,
      "epoch": 3.1897229747443854,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039971685612083616,
      "loss": 5.4602,
      "mean_token_accuracy": 0.18539584279060364,
      "num_tokens": 73963808.0,
      "step": 36500
    },
    {
      "entropy": 5.912225246429443,
      "epoch": 3.19015992309709,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039969091895532933,
      "loss": 5.523,
      "mean_token_accuracy": 0.18287118822336196,
      "num_tokens": 73973083.0,
      "step": 36505
    },
    {
      "entropy": 5.779699182510376,
      "epoch": 3.1905968714497948,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003996649793981853,
      "loss": 5.3361,
      "mean_token_accuracy": 0.20301671624183654,
      "num_tokens": 73983966.0,
      "step": 36510
    },
    {
      "entropy": 5.9046773433685305,
      "epoch": 3.1910338198024992,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039963903744990166,
      "loss": 5.4924,
      "mean_token_accuracy": 0.1869341567158699,
      "num_tokens": 73993874.0,
      "step": 36515
    },
    {
      "entropy": 5.827912855148315,
      "epoch": 3.191470768155204,
      "grad_norm": 1.09375,
      "learning_rate": 0.000399613093110976,
      "loss": 5.4582,
      "mean_token_accuracy": 0.181590473651886,
      "num_tokens": 74004179.0,
      "step": 36520
    },
    {
      "entropy": 5.842043542861939,
      "epoch": 3.1919077165079086,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003995871463819059,
      "loss": 5.4522,
      "mean_token_accuracy": 0.18821569830179213,
      "num_tokens": 74014941.0,
      "step": 36525
    },
    {
      "entropy": 5.871576023101807,
      "epoch": 3.1923446648606135,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039956119726318907,
      "loss": 5.5292,
      "mean_token_accuracy": 0.17607279121875763,
      "num_tokens": 74026620.0,
      "step": 36530
    },
    {
      "entropy": 5.868998765945435,
      "epoch": 3.192781613213318,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039953524575532317,
      "loss": 5.4567,
      "mean_token_accuracy": 0.18326786309480667,
      "num_tokens": 74036341.0,
      "step": 36535
    },
    {
      "entropy": 5.833324289321899,
      "epoch": 3.193218561566023,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039950929185880604,
      "loss": 5.4151,
      "mean_token_accuracy": 0.19172460585832596,
      "num_tokens": 74046884.0,
      "step": 36540
    },
    {
      "entropy": 5.748606157302857,
      "epoch": 3.1936555099187274,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003994833355741354,
      "loss": 5.3868,
      "mean_token_accuracy": 0.19317857027053834,
      "num_tokens": 74056706.0,
      "step": 36545
    },
    {
      "entropy": 5.673843717575073,
      "epoch": 3.1940924582714323,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039945737690180917,
      "loss": 5.2951,
      "mean_token_accuracy": 0.2110726058483124,
      "num_tokens": 74067046.0,
      "step": 36550
    },
    {
      "entropy": 5.879336214065551,
      "epoch": 3.194529406624137,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003994314158423252,
      "loss": 5.4033,
      "mean_token_accuracy": 0.185566945374012,
      "num_tokens": 74076884.0,
      "step": 36555
    },
    {
      "entropy": 5.831163835525513,
      "epoch": 3.1949663549768417,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003994054523961816,
      "loss": 5.4774,
      "mean_token_accuracy": 0.18525121361017227,
      "num_tokens": 74086482.0,
      "step": 36560
    },
    {
      "entropy": 5.761620235443115,
      "epoch": 3.1954033033295466,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039937948656387616,
      "loss": 5.379,
      "mean_token_accuracy": 0.1954501375555992,
      "num_tokens": 74095916.0,
      "step": 36565
    },
    {
      "entropy": 5.89311146736145,
      "epoch": 3.195840251682251,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039935351834590696,
      "loss": 5.4953,
      "mean_token_accuracy": 0.184486523270607,
      "num_tokens": 74106054.0,
      "step": 36570
    },
    {
      "entropy": 5.779586744308472,
      "epoch": 3.196277200034956,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039932754774277214,
      "loss": 5.4286,
      "mean_token_accuracy": 0.19447319060564042,
      "num_tokens": 74115641.0,
      "step": 36575
    },
    {
      "entropy": 5.838117170333862,
      "epoch": 3.1967141483876604,
      "grad_norm": 1.0,
      "learning_rate": 0.0003993015747549698,
      "loss": 5.4167,
      "mean_token_accuracy": 0.19196174442768096,
      "num_tokens": 74125370.0,
      "step": 36580
    },
    {
      "entropy": 5.86352767944336,
      "epoch": 3.1971510967403654,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039927559938299814,
      "loss": 5.409,
      "mean_token_accuracy": 0.1933903217315674,
      "num_tokens": 74134959.0,
      "step": 36585
    },
    {
      "entropy": 5.8933563232421875,
      "epoch": 3.19758804509307,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003992496216273554,
      "loss": 5.5352,
      "mean_token_accuracy": 0.18155248016119002,
      "num_tokens": 74144939.0,
      "step": 36590
    },
    {
      "entropy": 5.849273777008056,
      "epoch": 3.1980249934457747,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039922364148853973,
      "loss": 5.3243,
      "mean_token_accuracy": 0.1987498238682747,
      "num_tokens": 74154336.0,
      "step": 36595
    },
    {
      "entropy": 5.82608437538147,
      "epoch": 3.198461941798479,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003991976589670496,
      "loss": 5.4278,
      "mean_token_accuracy": 0.18709995448589326,
      "num_tokens": 74164022.0,
      "step": 36600
    },
    {
      "entropy": 5.9119469165802006,
      "epoch": 3.198898890151184,
      "grad_norm": 1.0,
      "learning_rate": 0.0003991716740633832,
      "loss": 5.421,
      "mean_token_accuracy": 0.18675192445516586,
      "num_tokens": 74174186.0,
      "step": 36605
    },
    {
      "entropy": 5.803337144851684,
      "epoch": 3.199335838503889,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003991456867780391,
      "loss": 5.4066,
      "mean_token_accuracy": 0.19343821257352828,
      "num_tokens": 74183913.0,
      "step": 36610
    },
    {
      "entropy": 5.823154544830322,
      "epoch": 3.1997727868565935,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003991196971115157,
      "loss": 5.4742,
      "mean_token_accuracy": 0.19044262319803237,
      "num_tokens": 74194307.0,
      "step": 36615
    },
    {
      "entropy": 5.900401782989502,
      "epoch": 3.2002097352092984,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039909370506431146,
      "loss": 5.4356,
      "mean_token_accuracy": 0.1860241949558258,
      "num_tokens": 74204672.0,
      "step": 36620
    },
    {
      "entropy": 5.876846408843994,
      "epoch": 3.200646683562003,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003990677106369249,
      "loss": 5.4391,
      "mean_token_accuracy": 0.19138703346252442,
      "num_tokens": 74214022.0,
      "step": 36625
    },
    {
      "entropy": 5.752668380737305,
      "epoch": 3.201083631914708,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039904171382985465,
      "loss": 5.3812,
      "mean_token_accuracy": 0.18940427005290986,
      "num_tokens": 74223736.0,
      "step": 36630
    },
    {
      "entropy": 5.8290290355682375,
      "epoch": 3.2015205802674123,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003990157146435993,
      "loss": 5.4234,
      "mean_token_accuracy": 0.18826052248477937,
      "num_tokens": 74234742.0,
      "step": 36635
    },
    {
      "entropy": 5.929099273681641,
      "epoch": 3.201957528620117,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003989897130786577,
      "loss": 5.4181,
      "mean_token_accuracy": 0.189544478058815,
      "num_tokens": 74244949.0,
      "step": 36640
    },
    {
      "entropy": 5.849696302413941,
      "epoch": 3.2023944769728216,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039896370913552825,
      "loss": 5.4696,
      "mean_token_accuracy": 0.18906817138195037,
      "num_tokens": 74255280.0,
      "step": 36645
    },
    {
      "entropy": 5.680950164794922,
      "epoch": 3.2028314253255266,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039893770281471005,
      "loss": 5.3113,
      "mean_token_accuracy": 0.20798212736845018,
      "num_tokens": 74265615.0,
      "step": 36650
    },
    {
      "entropy": 5.784150886535644,
      "epoch": 3.203268373678231,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003989116941167016,
      "loss": 5.4475,
      "mean_token_accuracy": 0.1942515790462494,
      "num_tokens": 74275791.0,
      "step": 36655
    },
    {
      "entropy": 5.9032903671264645,
      "epoch": 3.203705322030936,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039888568304200206,
      "loss": 5.4317,
      "mean_token_accuracy": 0.18731134831905366,
      "num_tokens": 74286295.0,
      "step": 36660
    },
    {
      "entropy": 5.778033256530762,
      "epoch": 3.204142270383641,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039885966959111015,
      "loss": 5.3588,
      "mean_token_accuracy": 0.19440564066171645,
      "num_tokens": 74295163.0,
      "step": 36665
    },
    {
      "entropy": 5.785285711288452,
      "epoch": 3.2045792187363453,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003988336537645249,
      "loss": 5.4124,
      "mean_token_accuracy": 0.18809611201286316,
      "num_tokens": 74304855.0,
      "step": 36670
    },
    {
      "entropy": 5.885551643371582,
      "epoch": 3.2050161670890502,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003988076355627453,
      "loss": 5.4627,
      "mean_token_accuracy": 0.1849295675754547,
      "num_tokens": 74314584.0,
      "step": 36675
    },
    {
      "entropy": 5.827053880691528,
      "epoch": 3.2054531154417547,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003987816149862703,
      "loss": 5.4329,
      "mean_token_accuracy": 0.19325069189071656,
      "num_tokens": 74324992.0,
      "step": 36680
    },
    {
      "entropy": 5.762261962890625,
      "epoch": 3.2058900637944596,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039875559203559913,
      "loss": 5.3855,
      "mean_token_accuracy": 0.19105874598026276,
      "num_tokens": 74336180.0,
      "step": 36685
    },
    {
      "entropy": 5.7992791652679445,
      "epoch": 3.206327012147164,
      "grad_norm": 1.71875,
      "learning_rate": 0.0003987295667112309,
      "loss": 5.312,
      "mean_token_accuracy": 0.19887296706438065,
      "num_tokens": 74346073.0,
      "step": 36690
    },
    {
      "entropy": 5.783569860458374,
      "epoch": 3.206763960499869,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003987035390136647,
      "loss": 5.3608,
      "mean_token_accuracy": 0.19162001013755797,
      "num_tokens": 74356470.0,
      "step": 36695
    },
    {
      "entropy": 5.782414960861206,
      "epoch": 3.2072009088525735,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039867750894339974,
      "loss": 5.3665,
      "mean_token_accuracy": 0.19755895137786866,
      "num_tokens": 74367035.0,
      "step": 36700
    },
    {
      "entropy": 5.79992184638977,
      "epoch": 3.2076378572052784,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003986514765009354,
      "loss": 5.4192,
      "mean_token_accuracy": 0.19174507707357408,
      "num_tokens": 74377264.0,
      "step": 36705
    },
    {
      "entropy": 5.904406404495239,
      "epoch": 3.208074805557983,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039862544168677103,
      "loss": 5.5385,
      "mean_token_accuracy": 0.1770267367362976,
      "num_tokens": 74389053.0,
      "step": 36710
    },
    {
      "entropy": 5.878494310379028,
      "epoch": 3.2085117539106878,
      "grad_norm": 1.125,
      "learning_rate": 0.0003985994045014058,
      "loss": 5.4114,
      "mean_token_accuracy": 0.19287653416395187,
      "num_tokens": 74399546.0,
      "step": 36715
    },
    {
      "entropy": 5.8094587326049805,
      "epoch": 3.2089487022633927,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039857336494533924,
      "loss": 5.421,
      "mean_token_accuracy": 0.19216849356889726,
      "num_tokens": 74409905.0,
      "step": 36720
    },
    {
      "entropy": 5.7590197086334225,
      "epoch": 3.209385650616097,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003985473230190708,
      "loss": 5.3937,
      "mean_token_accuracy": 0.19165326952934264,
      "num_tokens": 74419893.0,
      "step": 36725
    },
    {
      "entropy": 5.833619499206543,
      "epoch": 3.209822598968802,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003985212787231,
      "loss": 5.3876,
      "mean_token_accuracy": 0.19299955368041993,
      "num_tokens": 74430286.0,
      "step": 36730
    },
    {
      "entropy": 5.843651580810547,
      "epoch": 3.2102595473215065,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003984952320579263,
      "loss": 5.3832,
      "mean_token_accuracy": 0.1913555607199669,
      "num_tokens": 74439601.0,
      "step": 36735
    },
    {
      "entropy": 5.86542649269104,
      "epoch": 3.2106964956742114,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003984691830240494,
      "loss": 5.437,
      "mean_token_accuracy": 0.1882453069090843,
      "num_tokens": 74448763.0,
      "step": 36740
    },
    {
      "entropy": 5.890333080291748,
      "epoch": 3.211133444026916,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039844313162196876,
      "loss": 5.5308,
      "mean_token_accuracy": 0.18493567109107972,
      "num_tokens": 74458357.0,
      "step": 36745
    },
    {
      "entropy": 5.8515290260314945,
      "epoch": 3.211570392379621,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039841707785218426,
      "loss": 5.3974,
      "mean_token_accuracy": 0.1921614170074463,
      "num_tokens": 74468280.0,
      "step": 36750
    },
    {
      "entropy": 5.750741767883301,
      "epoch": 3.2120073407323253,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039839102171519556,
      "loss": 5.3235,
      "mean_token_accuracy": 0.20177869945764543,
      "num_tokens": 74478103.0,
      "step": 36755
    },
    {
      "entropy": 5.808017253875732,
      "epoch": 3.21244428908503,
      "grad_norm": 1.0,
      "learning_rate": 0.0003983649632115024,
      "loss": 5.3895,
      "mean_token_accuracy": 0.1880021408200264,
      "num_tokens": 74487903.0,
      "step": 36760
    },
    {
      "entropy": 5.773322868347168,
      "epoch": 3.2128812374377347,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003983389023416045,
      "loss": 5.4186,
      "mean_token_accuracy": 0.19036293774843216,
      "num_tokens": 74497998.0,
      "step": 36765
    },
    {
      "entropy": 5.815274477005005,
      "epoch": 3.2133181857904396,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039831283910600197,
      "loss": 5.4053,
      "mean_token_accuracy": 0.18694098442792892,
      "num_tokens": 74507313.0,
      "step": 36770
    },
    {
      "entropy": 5.868163633346557,
      "epoch": 3.2137551341431445,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003982867735051945,
      "loss": 5.4302,
      "mean_token_accuracy": 0.1850226640701294,
      "num_tokens": 74516983.0,
      "step": 36775
    },
    {
      "entropy": 5.825831460952759,
      "epoch": 3.214192082495849,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003982607055396821,
      "loss": 5.3957,
      "mean_token_accuracy": 0.18945649415254592,
      "num_tokens": 74527061.0,
      "step": 36780
    },
    {
      "entropy": 5.810884141921997,
      "epoch": 3.214629030848554,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00039823463520996484,
      "loss": 5.436,
      "mean_token_accuracy": 0.1953621104359627,
      "num_tokens": 74537782.0,
      "step": 36785
    },
    {
      "entropy": 5.885597467422485,
      "epoch": 3.2150659792012584,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003982085625165426,
      "loss": 5.4181,
      "mean_token_accuracy": 0.19593511521816254,
      "num_tokens": 74548082.0,
      "step": 36790
    },
    {
      "entropy": 5.808909368515015,
      "epoch": 3.2155029275539633,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003981824874599157,
      "loss": 5.4267,
      "mean_token_accuracy": 0.18909573853015899,
      "num_tokens": 74559286.0,
      "step": 36795
    },
    {
      "entropy": 5.806400632858276,
      "epoch": 3.2159398759066677,
      "grad_norm": 1.046875,
      "learning_rate": 0.000398156410040584,
      "loss": 5.4724,
      "mean_token_accuracy": 0.18630594313144683,
      "num_tokens": 74569306.0,
      "step": 36800
    },
    {
      "entropy": 5.90838212966919,
      "epoch": 3.2163768242593727,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039813033025904793,
      "loss": 5.4922,
      "mean_token_accuracy": 0.17828610241413118,
      "num_tokens": 74579798.0,
      "step": 36805
    },
    {
      "entropy": 5.844003009796142,
      "epoch": 3.216813772612077,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00039810424811580756,
      "loss": 5.4163,
      "mean_token_accuracy": 0.19036271274089814,
      "num_tokens": 74590068.0,
      "step": 36810
    },
    {
      "entropy": 5.857073354721069,
      "epoch": 3.217250720964782,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039807816361136323,
      "loss": 5.4966,
      "mean_token_accuracy": 0.18037588149309158,
      "num_tokens": 74600341.0,
      "step": 36815
    },
    {
      "entropy": 5.807110023498535,
      "epoch": 3.2176876693174865,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003980520767462152,
      "loss": 5.4577,
      "mean_token_accuracy": 0.19197207391262056,
      "num_tokens": 74609548.0,
      "step": 36820
    },
    {
      "entropy": 5.8638286113739015,
      "epoch": 3.2181246176701914,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003980259875208639,
      "loss": 5.3525,
      "mean_token_accuracy": 0.19665785431861876,
      "num_tokens": 74619590.0,
      "step": 36825
    },
    {
      "entropy": 5.864514684677124,
      "epoch": 3.218561566022896,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003979998959358097,
      "loss": 5.4063,
      "mean_token_accuracy": 0.19129252433776855,
      "num_tokens": 74629664.0,
      "step": 36830
    },
    {
      "entropy": 5.810523796081543,
      "epoch": 3.218998514375601,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003979738019915529,
      "loss": 5.429,
      "mean_token_accuracy": 0.19319439828395843,
      "num_tokens": 74639114.0,
      "step": 36835
    },
    {
      "entropy": 5.757863616943359,
      "epoch": 3.2194354627283057,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003979477056885943,
      "loss": 5.3414,
      "mean_token_accuracy": 0.19792929589748381,
      "num_tokens": 74648730.0,
      "step": 36840
    },
    {
      "entropy": 5.894167137145996,
      "epoch": 3.21987241108101,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039792160702743426,
      "loss": 5.4508,
      "mean_token_accuracy": 0.1869723990559578,
      "num_tokens": 74659102.0,
      "step": 36845
    },
    {
      "entropy": 5.872484111785889,
      "epoch": 3.220309359433715,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00039789550600857336,
      "loss": 5.5516,
      "mean_token_accuracy": 0.17435552924871445,
      "num_tokens": 74669344.0,
      "step": 36850
    },
    {
      "entropy": 5.900939130783081,
      "epoch": 3.2207463077864196,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003978694026325123,
      "loss": 5.4047,
      "mean_token_accuracy": 0.1892764687538147,
      "num_tokens": 74678524.0,
      "step": 36855
    },
    {
      "entropy": 5.8379453182220455,
      "epoch": 3.2211832561391245,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003978432968997517,
      "loss": 5.4679,
      "mean_token_accuracy": 0.19009905010461808,
      "num_tokens": 74688584.0,
      "step": 36860
    },
    {
      "entropy": 5.9131293296813965,
      "epoch": 3.221620204491829,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00039781718881079227,
      "loss": 5.4805,
      "mean_token_accuracy": 0.18255394250154494,
      "num_tokens": 74699233.0,
      "step": 36865
    },
    {
      "entropy": 5.914335536956787,
      "epoch": 3.222057152844534,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039779107836613475,
      "loss": 5.4689,
      "mean_token_accuracy": 0.18598252832889556,
      "num_tokens": 74710144.0,
      "step": 36870
    },
    {
      "entropy": 5.800966453552246,
      "epoch": 3.2224941011972383,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039776496556628014,
      "loss": 5.4921,
      "mean_token_accuracy": 0.184961798787117,
      "num_tokens": 74720289.0,
      "step": 36875
    },
    {
      "entropy": 5.837220668792725,
      "epoch": 3.2229310495499432,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039773885041172907,
      "loss": 5.4827,
      "mean_token_accuracy": 0.18109067231416703,
      "num_tokens": 74732395.0,
      "step": 36880
    },
    {
      "entropy": 5.885636568069458,
      "epoch": 3.2233679979026477,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039771273290298254,
      "loss": 5.4169,
      "mean_token_accuracy": 0.1956194519996643,
      "num_tokens": 74741986.0,
      "step": 36885
    },
    {
      "entropy": 5.756183242797851,
      "epoch": 3.2238049462553526,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003976866130405415,
      "loss": 5.3452,
      "mean_token_accuracy": 0.18690635710954667,
      "num_tokens": 74751540.0,
      "step": 36890
    },
    {
      "entropy": 5.838868618011475,
      "epoch": 3.224241894608057,
      "grad_norm": 1.125,
      "learning_rate": 0.00039766049082490696,
      "loss": 5.4026,
      "mean_token_accuracy": 0.19039017111063003,
      "num_tokens": 74761875.0,
      "step": 36895
    },
    {
      "entropy": 5.814045333862305,
      "epoch": 3.224678842960762,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039763436625657987,
      "loss": 5.3268,
      "mean_token_accuracy": 0.1943443775177002,
      "num_tokens": 74771273.0,
      "step": 36900
    },
    {
      "entropy": 5.76508526802063,
      "epoch": 3.225115791313467,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003976082393360614,
      "loss": 5.3361,
      "mean_token_accuracy": 0.1976249948143959,
      "num_tokens": 74781176.0,
      "step": 36905
    },
    {
      "entropy": 5.829538583755493,
      "epoch": 3.2255527396661714,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039758211006385273,
      "loss": 5.4284,
      "mean_token_accuracy": 0.19186076521873474,
      "num_tokens": 74790731.0,
      "step": 36910
    },
    {
      "entropy": 5.803663015365601,
      "epoch": 3.2259896880188763,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039755597844045486,
      "loss": 5.3483,
      "mean_token_accuracy": 0.19559736400842667,
      "num_tokens": 74799810.0,
      "step": 36915
    },
    {
      "entropy": 5.838222503662109,
      "epoch": 3.226426636371581,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003975298444663692,
      "loss": 5.4572,
      "mean_token_accuracy": 0.1856011062860489,
      "num_tokens": 74809701.0,
      "step": 36920
    },
    {
      "entropy": 5.848658609390259,
      "epoch": 3.2268635847242857,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039750370814209683,
      "loss": 5.4074,
      "mean_token_accuracy": 0.19630872905254365,
      "num_tokens": 74820061.0,
      "step": 36925
    },
    {
      "entropy": 5.8622682094573975,
      "epoch": 3.22730053307699,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003974775694681392,
      "loss": 5.381,
      "mean_token_accuracy": 0.1906675934791565,
      "num_tokens": 74830360.0,
      "step": 36930
    },
    {
      "entropy": 5.858479785919189,
      "epoch": 3.227737481429695,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003974514284449975,
      "loss": 5.4814,
      "mean_token_accuracy": 0.18756469935178757,
      "num_tokens": 74841855.0,
      "step": 36935
    },
    {
      "entropy": 5.753161334991455,
      "epoch": 3.2281744297823995,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00039742528507317337,
      "loss": 5.3772,
      "mean_token_accuracy": 0.1966838926076889,
      "num_tokens": 74851955.0,
      "step": 36940
    },
    {
      "entropy": 5.82631630897522,
      "epoch": 3.2286113781351045,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003973991393531681,
      "loss": 5.3944,
      "mean_token_accuracy": 0.1975926175713539,
      "num_tokens": 74861720.0,
      "step": 36945
    },
    {
      "entropy": 5.815298366546631,
      "epoch": 3.229048326487809,
      "grad_norm": 0.984375,
      "learning_rate": 0.00039737299128548317,
      "loss": 5.3942,
      "mean_token_accuracy": 0.18705493807792664,
      "num_tokens": 74872136.0,
      "step": 36950
    },
    {
      "entropy": 5.871900272369385,
      "epoch": 3.229485274840514,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003973468408706202,
      "loss": 5.424,
      "mean_token_accuracy": 0.18512572795152665,
      "num_tokens": 74881683.0,
      "step": 36955
    },
    {
      "entropy": 5.86396050453186,
      "epoch": 3.2299222231932188,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039732068810908067,
      "loss": 5.363,
      "mean_token_accuracy": 0.19684022516012192,
      "num_tokens": 74892175.0,
      "step": 36960
    },
    {
      "entropy": 5.811765813827515,
      "epoch": 3.2303591715459232,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003972945330013662,
      "loss": 5.3881,
      "mean_token_accuracy": 0.1931113436818123,
      "num_tokens": 74903547.0,
      "step": 36965
    },
    {
      "entropy": 5.795230722427368,
      "epoch": 3.230796119898628,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039726837554797856,
      "loss": 5.3791,
      "mean_token_accuracy": 0.1960186704993248,
      "num_tokens": 74912990.0,
      "step": 36970
    },
    {
      "entropy": 5.797853469848633,
      "epoch": 3.2312330682513326,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039724221574941943,
      "loss": 5.4187,
      "mean_token_accuracy": 0.18693312257528305,
      "num_tokens": 74922306.0,
      "step": 36975
    },
    {
      "entropy": 5.850541687011718,
      "epoch": 3.2316700166040375,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039721605360619054,
      "loss": 5.4078,
      "mean_token_accuracy": 0.18802638500928878,
      "num_tokens": 74932207.0,
      "step": 36980
    },
    {
      "entropy": 5.8073728561401365,
      "epoch": 3.232106964956742,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039718988911879367,
      "loss": 5.4512,
      "mean_token_accuracy": 0.18739273697137832,
      "num_tokens": 74941933.0,
      "step": 36985
    },
    {
      "entropy": 5.891920566558838,
      "epoch": 3.232543913309447,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039716372228773073,
      "loss": 5.501,
      "mean_token_accuracy": 0.18195506930351257,
      "num_tokens": 74953418.0,
      "step": 36990
    },
    {
      "entropy": 5.899511337280273,
      "epoch": 3.2329808616621514,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039713755311350356,
      "loss": 5.4384,
      "mean_token_accuracy": 0.18737725168466568,
      "num_tokens": 74963381.0,
      "step": 36995
    },
    {
      "entropy": 5.868180656433106,
      "epoch": 3.2334178100148563,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039711138159661413,
      "loss": 5.4438,
      "mean_token_accuracy": 0.19173811376094818,
      "num_tokens": 74974072.0,
      "step": 37000
    },
    {
      "entropy": 5.7446081161499025,
      "epoch": 3.2338547583675608,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003970852077375644,
      "loss": 5.3296,
      "mean_token_accuracy": 0.19905803948640824,
      "num_tokens": 74984191.0,
      "step": 37005
    },
    {
      "entropy": 5.769464588165283,
      "epoch": 3.2342917067202657,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003970590315368564,
      "loss": 5.4298,
      "mean_token_accuracy": 0.18582020848989486,
      "num_tokens": 74994278.0,
      "step": 37010
    },
    {
      "entropy": 5.808700323104858,
      "epoch": 3.2347286550729706,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003970328529949923,
      "loss": 5.3829,
      "mean_token_accuracy": 0.20149502605199815,
      "num_tokens": 75004150.0,
      "step": 37015
    },
    {
      "entropy": 5.827130889892578,
      "epoch": 3.235165603425675,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003970066721124741,
      "loss": 5.3795,
      "mean_token_accuracy": 0.1961563393473625,
      "num_tokens": 75013632.0,
      "step": 37020
    },
    {
      "entropy": 5.805138015747071,
      "epoch": 3.23560255177838,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003969804888898039,
      "loss": 5.4658,
      "mean_token_accuracy": 0.18235530257225036,
      "num_tokens": 75023217.0,
      "step": 37025
    },
    {
      "entropy": 5.836316680908203,
      "epoch": 3.2360395001310844,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003969543033274841,
      "loss": 5.3997,
      "mean_token_accuracy": 0.19394033253192902,
      "num_tokens": 75033771.0,
      "step": 37030
    },
    {
      "entropy": 5.894764375686646,
      "epoch": 3.2364764484837893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039692811542601684,
      "loss": 5.4627,
      "mean_token_accuracy": 0.18851110935211182,
      "num_tokens": 75044151.0,
      "step": 37035
    },
    {
      "entropy": 5.759392738342285,
      "epoch": 3.236913396836494,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003969019251859045,
      "loss": 5.3443,
      "mean_token_accuracy": 0.20115939676761627,
      "num_tokens": 75054129.0,
      "step": 37040
    },
    {
      "entropy": 5.799824380874634,
      "epoch": 3.2373503451891987,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039687573260764933,
      "loss": 5.3875,
      "mean_token_accuracy": 0.1927237719297409,
      "num_tokens": 75063147.0,
      "step": 37045
    },
    {
      "entropy": 5.75469388961792,
      "epoch": 3.237787293541903,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039684953769175376,
      "loss": 5.4465,
      "mean_token_accuracy": 0.1843480721116066,
      "num_tokens": 75072389.0,
      "step": 37050
    },
    {
      "entropy": 5.800268840789795,
      "epoch": 3.238224241894608,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003968233404387201,
      "loss": 5.3971,
      "mean_token_accuracy": 0.191850945353508,
      "num_tokens": 75081253.0,
      "step": 37055
    },
    {
      "entropy": 5.79422926902771,
      "epoch": 3.2386611902473126,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000396797140849051,
      "loss": 5.3839,
      "mean_token_accuracy": 0.20136246383190154,
      "num_tokens": 75091316.0,
      "step": 37060
    },
    {
      "entropy": 5.88466854095459,
      "epoch": 3.2390981386000175,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039677093892324895,
      "loss": 5.5004,
      "mean_token_accuracy": 0.184152489900589,
      "num_tokens": 75101204.0,
      "step": 37065
    },
    {
      "entropy": 5.770840740203857,
      "epoch": 3.2395350869527224,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003967447346618164,
      "loss": 5.3985,
      "mean_token_accuracy": 0.18585017174482346,
      "num_tokens": 75111524.0,
      "step": 37070
    },
    {
      "entropy": 5.785634565353393,
      "epoch": 3.239972035305427,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039671852806525613,
      "loss": 5.3936,
      "mean_token_accuracy": 0.18716720193624498,
      "num_tokens": 75120699.0,
      "step": 37075
    },
    {
      "entropy": 5.915038299560547,
      "epoch": 3.240408983658132,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039669231913407065,
      "loss": 5.4054,
      "mean_token_accuracy": 0.19166503995656967,
      "num_tokens": 75130892.0,
      "step": 37080
    },
    {
      "entropy": 5.857251405715942,
      "epoch": 3.2408459320108363,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003966661078687628,
      "loss": 5.4925,
      "mean_token_accuracy": 0.18793762922286988,
      "num_tokens": 75140994.0,
      "step": 37085
    },
    {
      "entropy": 5.82598991394043,
      "epoch": 3.241282880363541,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003966398942698352,
      "loss": 5.4949,
      "mean_token_accuracy": 0.1899473175406456,
      "num_tokens": 75151084.0,
      "step": 37090
    },
    {
      "entropy": 5.868063640594483,
      "epoch": 3.2417198287162456,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003966136783377907,
      "loss": 5.4189,
      "mean_token_accuracy": 0.18817341923713685,
      "num_tokens": 75160815.0,
      "step": 37095
    },
    {
      "entropy": 5.895303773880005,
      "epoch": 3.2421567770689506,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003965874600731321,
      "loss": 5.4002,
      "mean_token_accuracy": 0.18996490985155107,
      "num_tokens": 75170827.0,
      "step": 37100
    },
    {
      "entropy": 5.848257255554199,
      "epoch": 3.242593725421655,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003965612394763623,
      "loss": 5.4309,
      "mean_token_accuracy": 0.1922006592154503,
      "num_tokens": 75180250.0,
      "step": 37105
    },
    {
      "entropy": 5.784821653366089,
      "epoch": 3.24303067377436,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003965350165479842,
      "loss": 5.4276,
      "mean_token_accuracy": 0.18810931891202926,
      "num_tokens": 75190862.0,
      "step": 37110
    },
    {
      "entropy": 5.791286611557007,
      "epoch": 3.2434676221270644,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003965087912885009,
      "loss": 5.4061,
      "mean_token_accuracy": 0.18527624160051345,
      "num_tokens": 75199632.0,
      "step": 37115
    },
    {
      "entropy": 5.814180088043213,
      "epoch": 3.2439045704797693,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039648256369841523,
      "loss": 5.3511,
      "mean_token_accuracy": 0.18868542164564134,
      "num_tokens": 75209626.0,
      "step": 37120
    },
    {
      "entropy": 5.836448717117309,
      "epoch": 3.2443415188324742,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003964563337782305,
      "loss": 5.4606,
      "mean_token_accuracy": 0.18329747915267944,
      "num_tokens": 75220682.0,
      "step": 37125
    },
    {
      "entropy": 5.814003133773804,
      "epoch": 3.2447784671851787,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039643010152844946,
      "loss": 5.4198,
      "mean_token_accuracy": 0.1914539158344269,
      "num_tokens": 75230330.0,
      "step": 37130
    },
    {
      "entropy": 5.855144929885864,
      "epoch": 3.2452154155378836,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039640386694957553,
      "loss": 5.412,
      "mean_token_accuracy": 0.18614706248044968,
      "num_tokens": 75240485.0,
      "step": 37135
    },
    {
      "entropy": 5.8156530380249025,
      "epoch": 3.245652363890588,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039637763004211175,
      "loss": 5.4333,
      "mean_token_accuracy": 0.1904906749725342,
      "num_tokens": 75250670.0,
      "step": 37140
    },
    {
      "entropy": 5.850723361968994,
      "epoch": 3.246089312243293,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039635139080656146,
      "loss": 5.4081,
      "mean_token_accuracy": 0.1980605974793434,
      "num_tokens": 75260312.0,
      "step": 37145
    },
    {
      "entropy": 5.85810170173645,
      "epoch": 3.2465262605959975,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00039632514924342794,
      "loss": 5.4315,
      "mean_token_accuracy": 0.1837716057896614,
      "num_tokens": 75269978.0,
      "step": 37150
    },
    {
      "entropy": 5.809707546234131,
      "epoch": 3.2469632089487024,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039629890535321444,
      "loss": 5.4303,
      "mean_token_accuracy": 0.1919853374361992,
      "num_tokens": 75279781.0,
      "step": 37155
    },
    {
      "entropy": 5.894627714157105,
      "epoch": 3.247400157301407,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039627265913642435,
      "loss": 5.494,
      "mean_token_accuracy": 0.1845795378088951,
      "num_tokens": 75290352.0,
      "step": 37160
    },
    {
      "entropy": 5.894602632522583,
      "epoch": 3.2478371056541118,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003962464105935612,
      "loss": 5.4089,
      "mean_token_accuracy": 0.1833391457796097,
      "num_tokens": 75300511.0,
      "step": 37165
    },
    {
      "entropy": 5.8232816696167,
      "epoch": 3.2482740540068162,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039622015972512833,
      "loss": 5.4245,
      "mean_token_accuracy": 0.1961887702345848,
      "num_tokens": 75310717.0,
      "step": 37170
    },
    {
      "entropy": 5.847046613693237,
      "epoch": 3.248711002359521,
      "grad_norm": 1.0,
      "learning_rate": 0.0003961939065316293,
      "loss": 5.437,
      "mean_token_accuracy": 0.189336659014225,
      "num_tokens": 75322402.0,
      "step": 37175
    },
    {
      "entropy": 5.758520698547363,
      "epoch": 3.2491479507122256,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039616765101356755,
      "loss": 5.3355,
      "mean_token_accuracy": 0.19559289664030075,
      "num_tokens": 75331866.0,
      "step": 37180
    },
    {
      "entropy": 5.81611008644104,
      "epoch": 3.2495848990649305,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039614139317144676,
      "loss": 5.3369,
      "mean_token_accuracy": 0.19735849052667617,
      "num_tokens": 75341125.0,
      "step": 37185
    },
    {
      "entropy": 5.793008089065552,
      "epoch": 3.250021847417635,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003961151330057706,
      "loss": 5.4123,
      "mean_token_accuracy": 0.18975307494401933,
      "num_tokens": 75350518.0,
      "step": 37190
    },
    {
      "entropy": 5.85109314918518,
      "epoch": 3.25045879577034,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003960888705170428,
      "loss": 5.464,
      "mean_token_accuracy": 0.1838880106806755,
      "num_tokens": 75359330.0,
      "step": 37195
    },
    {
      "entropy": 5.919355630874634,
      "epoch": 3.250895744123045,
      "grad_norm": 1.0625,
      "learning_rate": 0.000396062605705767,
      "loss": 5.4922,
      "mean_token_accuracy": 0.1850689619779587,
      "num_tokens": 75368771.0,
      "step": 37200
    },
    {
      "entropy": 5.912691211700439,
      "epoch": 3.2513326924757493,
      "grad_norm": 1.03125,
      "learning_rate": 0.000396036338572447,
      "loss": 5.4483,
      "mean_token_accuracy": 0.18154437839984894,
      "num_tokens": 75379656.0,
      "step": 37205
    },
    {
      "entropy": 5.807328748703003,
      "epoch": 3.251769640828454,
      "grad_norm": 1.640625,
      "learning_rate": 0.00039601006911758646,
      "loss": 5.4478,
      "mean_token_accuracy": 0.18588249385356903,
      "num_tokens": 75390248.0,
      "step": 37210
    },
    {
      "entropy": 5.848774576187134,
      "epoch": 3.2522065891811587,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003959837973416895,
      "loss": 5.4624,
      "mean_token_accuracy": 0.18216035068035125,
      "num_tokens": 75400090.0,
      "step": 37215
    },
    {
      "entropy": 5.882959175109863,
      "epoch": 3.2526435375338636,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039595752324525984,
      "loss": 5.4519,
      "mean_token_accuracy": 0.18328630924224854,
      "num_tokens": 75410380.0,
      "step": 37220
    },
    {
      "entropy": 5.903407859802246,
      "epoch": 3.253080485886568,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003959312468288016,
      "loss": 5.3622,
      "mean_token_accuracy": 0.19270256012678147,
      "num_tokens": 75421265.0,
      "step": 37225
    },
    {
      "entropy": 5.832548379898071,
      "epoch": 3.253517434239273,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003959049680928187,
      "loss": 5.4049,
      "mean_token_accuracy": 0.19235543459653853,
      "num_tokens": 75432049.0,
      "step": 37230
    },
    {
      "entropy": 5.8176836490631105,
      "epoch": 3.2539543825919774,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003958786870378151,
      "loss": 5.4766,
      "mean_token_accuracy": 0.18628327548503876,
      "num_tokens": 75441364.0,
      "step": 37235
    },
    {
      "entropy": 5.857245349884034,
      "epoch": 3.2543913309446824,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003958524036642949,
      "loss": 5.4636,
      "mean_token_accuracy": 0.18515577167272568,
      "num_tokens": 75452019.0,
      "step": 37240
    },
    {
      "entropy": 5.88499116897583,
      "epoch": 3.254828279297387,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039582611797276237,
      "loss": 5.4553,
      "mean_token_accuracy": 0.18042528480291367,
      "num_tokens": 75463053.0,
      "step": 37245
    },
    {
      "entropy": 5.767717075347901,
      "epoch": 3.2552652276500917,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003957998299637215,
      "loss": 5.4132,
      "mean_token_accuracy": 0.1876517578959465,
      "num_tokens": 75472757.0,
      "step": 37250
    },
    {
      "entropy": 5.802687311172486,
      "epoch": 3.2557021760027967,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003957735396376766,
      "loss": 5.3699,
      "mean_token_accuracy": 0.1966856300830841,
      "num_tokens": 75482228.0,
      "step": 37255
    },
    {
      "entropy": 5.848437833786011,
      "epoch": 3.256139124355501,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003957472469951319,
      "loss": 5.4679,
      "mean_token_accuracy": 0.18747503459453582,
      "num_tokens": 75491344.0,
      "step": 37260
    },
    {
      "entropy": 5.829001426696777,
      "epoch": 3.256576072708206,
      "grad_norm": 1.0,
      "learning_rate": 0.00039572095203659177,
      "loss": 5.5058,
      "mean_token_accuracy": 0.18515388071537017,
      "num_tokens": 75502042.0,
      "step": 37265
    },
    {
      "entropy": 5.847824764251709,
      "epoch": 3.2570130210609105,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003956946547625606,
      "loss": 5.3775,
      "mean_token_accuracy": 0.19141202569007873,
      "num_tokens": 75512116.0,
      "step": 37270
    },
    {
      "entropy": 5.813665390014648,
      "epoch": 3.2574499694136154,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039566835517354266,
      "loss": 5.354,
      "mean_token_accuracy": 0.19387280195951462,
      "num_tokens": 75521220.0,
      "step": 37275
    },
    {
      "entropy": 5.875240468978882,
      "epoch": 3.25788691776632,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003956420532700423,
      "loss": 5.4558,
      "mean_token_accuracy": 0.18877315819263457,
      "num_tokens": 75530556.0,
      "step": 37280
    },
    {
      "entropy": 5.803584480285645,
      "epoch": 3.258323866119025,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003956157490525643,
      "loss": 5.4175,
      "mean_token_accuracy": 0.19395726770162583,
      "num_tokens": 75539843.0,
      "step": 37285
    },
    {
      "entropy": 5.780316638946533,
      "epoch": 3.2587608144717293,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000395589442521613,
      "loss": 5.3727,
      "mean_token_accuracy": 0.19289171397686006,
      "num_tokens": 75549945.0,
      "step": 37290
    },
    {
      "entropy": 5.852723503112793,
      "epoch": 3.259197762824434,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039556313367769297,
      "loss": 5.5079,
      "mean_token_accuracy": 0.18268150389194487,
      "num_tokens": 75560808.0,
      "step": 37295
    },
    {
      "entropy": 5.944702196121216,
      "epoch": 3.2596347111771387,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039553682252130886,
      "loss": 5.4819,
      "mean_token_accuracy": 0.18111545443534852,
      "num_tokens": 75571431.0,
      "step": 37300
    },
    {
      "entropy": 5.824877452850342,
      "epoch": 3.2600716595298436,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00039551050905296543,
      "loss": 5.4064,
      "mean_token_accuracy": 0.1875980406999588,
      "num_tokens": 75582543.0,
      "step": 37305
    },
    {
      "entropy": 5.80038366317749,
      "epoch": 3.2605086078825485,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039548419327316715,
      "loss": 5.3777,
      "mean_token_accuracy": 0.1991400644183159,
      "num_tokens": 75591951.0,
      "step": 37310
    },
    {
      "entropy": 5.748033761978149,
      "epoch": 3.260945556235253,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000395457875182419,
      "loss": 5.3893,
      "mean_token_accuracy": 0.19145141541957855,
      "num_tokens": 75602107.0,
      "step": 37315
    },
    {
      "entropy": 5.7901694774627686,
      "epoch": 3.261382504587958,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039543155478122564,
      "loss": 5.3414,
      "mean_token_accuracy": 0.19562146812677383,
      "num_tokens": 75612449.0,
      "step": 37320
    },
    {
      "entropy": 5.844538927078247,
      "epoch": 3.2618194529406623,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000395405232070092,
      "loss": 5.4897,
      "mean_token_accuracy": 0.18488170802593232,
      "num_tokens": 75622078.0,
      "step": 37325
    },
    {
      "entropy": 5.843072414398193,
      "epoch": 3.2622564012933672,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039537890704952294,
      "loss": 5.4863,
      "mean_token_accuracy": 0.18956813663244249,
      "num_tokens": 75632672.0,
      "step": 37330
    },
    {
      "entropy": 5.851376628875732,
      "epoch": 3.2626933496460717,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039535257972002325,
      "loss": 5.4144,
      "mean_token_accuracy": 0.193308088183403,
      "num_tokens": 75641633.0,
      "step": 37335
    },
    {
      "entropy": 5.810356187820434,
      "epoch": 3.2631302979987766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039532625008209814,
      "loss": 5.337,
      "mean_token_accuracy": 0.1972610667347908,
      "num_tokens": 75651109.0,
      "step": 37340
    },
    {
      "entropy": 5.8001209735870365,
      "epoch": 3.263567246351481,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003952999181362525,
      "loss": 5.3948,
      "mean_token_accuracy": 0.19432218223810196,
      "num_tokens": 75661176.0,
      "step": 37345
    },
    {
      "entropy": 5.882689428329468,
      "epoch": 3.264004194704186,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039527358388299135,
      "loss": 5.462,
      "mean_token_accuracy": 0.18037614673376084,
      "num_tokens": 75670838.0,
      "step": 37350
    },
    {
      "entropy": 5.822360372543335,
      "epoch": 3.2644411430568905,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003952472473228198,
      "loss": 5.3943,
      "mean_token_accuracy": 0.189251808822155,
      "num_tokens": 75681079.0,
      "step": 37355
    },
    {
      "entropy": 5.810986709594727,
      "epoch": 3.2648780914095954,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000395220908456243,
      "loss": 5.3062,
      "mean_token_accuracy": 0.20153491497039794,
      "num_tokens": 75691748.0,
      "step": 37360
    },
    {
      "entropy": 5.814336347579956,
      "epoch": 3.2653150397623003,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003951945672837663,
      "loss": 5.4254,
      "mean_token_accuracy": 0.1947813019156456,
      "num_tokens": 75702751.0,
      "step": 37365
    },
    {
      "entropy": 5.826012229919433,
      "epoch": 3.265751988115005,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039516822380589476,
      "loss": 5.413,
      "mean_token_accuracy": 0.19117974489927292,
      "num_tokens": 75712107.0,
      "step": 37370
    },
    {
      "entropy": 5.840648889541626,
      "epoch": 3.2661889364677097,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003951418780231338,
      "loss": 5.4074,
      "mean_token_accuracy": 0.19359344393014907,
      "num_tokens": 75721595.0,
      "step": 37375
    },
    {
      "entropy": 5.805277395248413,
      "epoch": 3.266625884820414,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003951155299359886,
      "loss": 5.3504,
      "mean_token_accuracy": 0.19202811270952225,
      "num_tokens": 75731332.0,
      "step": 37380
    },
    {
      "entropy": 5.8092375755310055,
      "epoch": 3.267062833173119,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003950891795449646,
      "loss": 5.3482,
      "mean_token_accuracy": 0.19849546253681183,
      "num_tokens": 75740425.0,
      "step": 37385
    },
    {
      "entropy": 5.750690364837647,
      "epoch": 3.2674997815258235,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003950628268505672,
      "loss": 5.3976,
      "mean_token_accuracy": 0.19477996230125427,
      "num_tokens": 75750091.0,
      "step": 37390
    },
    {
      "entropy": 5.835951089859009,
      "epoch": 3.2679367298785285,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039503647185330185,
      "loss": 5.4692,
      "mean_token_accuracy": 0.18670892268419265,
      "num_tokens": 75760780.0,
      "step": 37395
    },
    {
      "entropy": 5.86471152305603,
      "epoch": 3.268373678231233,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00039501011455367406,
      "loss": 5.429,
      "mean_token_accuracy": 0.18893443793058395,
      "num_tokens": 75771056.0,
      "step": 37400
    },
    {
      "entropy": 5.789533567428589,
      "epoch": 3.268810626583938,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003949837549521895,
      "loss": 5.3724,
      "mean_token_accuracy": 0.19256228655576707,
      "num_tokens": 75781265.0,
      "step": 37405
    },
    {
      "entropy": 5.848575496673584,
      "epoch": 3.2692475749366423,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00039495739304935357,
      "loss": 5.4557,
      "mean_token_accuracy": 0.1879527285695076,
      "num_tokens": 75791911.0,
      "step": 37410
    },
    {
      "entropy": 5.8031199932098385,
      "epoch": 3.2696845232893472,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000394931028845672,
      "loss": 5.3459,
      "mean_token_accuracy": 0.19446660578250885,
      "num_tokens": 75802085.0,
      "step": 37415
    },
    {
      "entropy": 5.815721130371093,
      "epoch": 3.270121471642052,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003949046623416504,
      "loss": 5.3484,
      "mean_token_accuracy": 0.19391589015722274,
      "num_tokens": 75810740.0,
      "step": 37420
    },
    {
      "entropy": 5.829553937911987,
      "epoch": 3.2705584199947566,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039487829353779464,
      "loss": 5.3677,
      "mean_token_accuracy": 0.19643450677394866,
      "num_tokens": 75820260.0,
      "step": 37425
    },
    {
      "entropy": 5.807199573516845,
      "epoch": 3.2709953683474615,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039485192243461035,
      "loss": 5.3715,
      "mean_token_accuracy": 0.19701554179191588,
      "num_tokens": 75830152.0,
      "step": 37430
    },
    {
      "entropy": 5.757636022567749,
      "epoch": 3.271432316700166,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003948255490326034,
      "loss": 5.3525,
      "mean_token_accuracy": 0.19661375880241394,
      "num_tokens": 75840617.0,
      "step": 37435
    },
    {
      "entropy": 5.762553596496582,
      "epoch": 3.271869265052871,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003947991733322796,
      "loss": 5.3325,
      "mean_token_accuracy": 0.19835250526666642,
      "num_tokens": 75850114.0,
      "step": 37440
    },
    {
      "entropy": 5.780206108093262,
      "epoch": 3.2723062134055754,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003947727953341449,
      "loss": 5.3802,
      "mean_token_accuracy": 0.20173653215169907,
      "num_tokens": 75861202.0,
      "step": 37445
    },
    {
      "entropy": 5.904019737243653,
      "epoch": 3.2727431617582803,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039474641503870524,
      "loss": 5.3769,
      "mean_token_accuracy": 0.1848931774497032,
      "num_tokens": 75870613.0,
      "step": 37450
    },
    {
      "entropy": 5.795709562301636,
      "epoch": 3.2731801101109848,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039472003244646656,
      "loss": 5.3525,
      "mean_token_accuracy": 0.19306788146495818,
      "num_tokens": 75880550.0,
      "step": 37455
    },
    {
      "entropy": 5.799242734909058,
      "epoch": 3.2736170584636897,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003946936475579349,
      "loss": 5.3607,
      "mean_token_accuracy": 0.1966223269701004,
      "num_tokens": 75890838.0,
      "step": 37460
    },
    {
      "entropy": 5.817186689376831,
      "epoch": 3.274054006816394,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003946672603736164,
      "loss": 5.4571,
      "mean_token_accuracy": 0.19143808484077454,
      "num_tokens": 75900834.0,
      "step": 37465
    },
    {
      "entropy": 5.7711199760437015,
      "epoch": 3.274490955169099,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039464087089401714,
      "loss": 5.4341,
      "mean_token_accuracy": 0.19387325495481492,
      "num_tokens": 75911442.0,
      "step": 37470
    },
    {
      "entropy": 5.8888054370880125,
      "epoch": 3.274927903521804,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039461447911964327,
      "loss": 5.4751,
      "mean_token_accuracy": 0.18436464071273803,
      "num_tokens": 75922824.0,
      "step": 37475
    },
    {
      "entropy": 5.865142917633056,
      "epoch": 3.2753648518745084,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003945880850510011,
      "loss": 5.4158,
      "mean_token_accuracy": 0.18718547523021697,
      "num_tokens": 75932784.0,
      "step": 37480
    },
    {
      "entropy": 5.75543532371521,
      "epoch": 3.275801800227213,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039456168868859663,
      "loss": 5.3435,
      "mean_token_accuracy": 0.1974824845790863,
      "num_tokens": 75942176.0,
      "step": 37485
    },
    {
      "entropy": 5.8076784133911135,
      "epoch": 3.276238748579918,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039453529003293647,
      "loss": 5.3813,
      "mean_token_accuracy": 0.19438507109880448,
      "num_tokens": 75952602.0,
      "step": 37490
    },
    {
      "entropy": 5.790217161178589,
      "epoch": 3.2766756969326227,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039450888908452677,
      "loss": 5.3636,
      "mean_token_accuracy": 0.19400166273117064,
      "num_tokens": 75962971.0,
      "step": 37495
    },
    {
      "entropy": 5.819219541549683,
      "epoch": 3.277112645285327,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003944824858438739,
      "loss": 5.4811,
      "mean_token_accuracy": 0.1897004947066307,
      "num_tokens": 75972751.0,
      "step": 37500
    },
    {
      "entropy": 5.858637523651123,
      "epoch": 3.277549593638032,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003944560803114845,
      "loss": 5.552,
      "mean_token_accuracy": 0.18106737583875657,
      "num_tokens": 75984653.0,
      "step": 37505
    },
    {
      "entropy": 5.914376974105835,
      "epoch": 3.2779865419907366,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003944296724878647,
      "loss": 5.427,
      "mean_token_accuracy": 0.1915833532810211,
      "num_tokens": 75994211.0,
      "step": 37510
    },
    {
      "entropy": 5.798228120803833,
      "epoch": 3.2784234903434415,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003944032623735213,
      "loss": 5.3517,
      "mean_token_accuracy": 0.19447022080421447,
      "num_tokens": 76003891.0,
      "step": 37515
    },
    {
      "entropy": 5.813213539123535,
      "epoch": 3.278860438696146,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003943768499689607,
      "loss": 5.402,
      "mean_token_accuracy": 0.191982664167881,
      "num_tokens": 76014625.0,
      "step": 37520
    },
    {
      "entropy": 5.817114448547363,
      "epoch": 3.279297387048851,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039435043527468963,
      "loss": 5.3586,
      "mean_token_accuracy": 0.19056814312934875,
      "num_tokens": 76024773.0,
      "step": 37525
    },
    {
      "entropy": 5.8201123714447025,
      "epoch": 3.2797343354015553,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039432401829121467,
      "loss": 5.2934,
      "mean_token_accuracy": 0.19588046222925187,
      "num_tokens": 76035085.0,
      "step": 37530
    },
    {
      "entropy": 5.811058807373047,
      "epoch": 3.2801712837542603,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003942975990190425,
      "loss": 5.4477,
      "mean_token_accuracy": 0.18448801785707475,
      "num_tokens": 76046055.0,
      "step": 37535
    },
    {
      "entropy": 5.7524158477783205,
      "epoch": 3.2806082321069647,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003942711774586799,
      "loss": 5.3094,
      "mean_token_accuracy": 0.2029389977455139,
      "num_tokens": 76055761.0,
      "step": 37540
    },
    {
      "entropy": 5.868040275573731,
      "epoch": 3.2810451804596696,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003942447536106335,
      "loss": 5.5168,
      "mean_token_accuracy": 0.1853923961520195,
      "num_tokens": 76065844.0,
      "step": 37545
    },
    {
      "entropy": 5.844596195220947,
      "epoch": 3.2814821288123746,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039421832747541034,
      "loss": 5.4181,
      "mean_token_accuracy": 0.18712867945432662,
      "num_tokens": 76076048.0,
      "step": 37550
    },
    {
      "entropy": 5.783387327194214,
      "epoch": 3.281919077165079,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003941918990535172,
      "loss": 5.3698,
      "mean_token_accuracy": 0.1944403573870659,
      "num_tokens": 76086775.0,
      "step": 37555
    },
    {
      "entropy": 5.833826065063477,
      "epoch": 3.282356025517784,
      "grad_norm": 1.0,
      "learning_rate": 0.0003941654683454609,
      "loss": 5.4423,
      "mean_token_accuracy": 0.18768578320741652,
      "num_tokens": 76097492.0,
      "step": 37560
    },
    {
      "entropy": 5.809890556335449,
      "epoch": 3.2827929738704884,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039413903535174855,
      "loss": 5.4079,
      "mean_token_accuracy": 0.18823193162679672,
      "num_tokens": 76106709.0,
      "step": 37565
    },
    {
      "entropy": 5.835576343536377,
      "epoch": 3.2832299222231933,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000394112600072887,
      "loss": 5.3898,
      "mean_token_accuracy": 0.18989737927913666,
      "num_tokens": 76117189.0,
      "step": 37570
    },
    {
      "entropy": 5.9123759269714355,
      "epoch": 3.283666870575898,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00039408616250938353,
      "loss": 5.457,
      "mean_token_accuracy": 0.18956712186336516,
      "num_tokens": 76127624.0,
      "step": 37575
    },
    {
      "entropy": 5.851936769485474,
      "epoch": 3.2841038189286027,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039405972266174494,
      "loss": 5.4245,
      "mean_token_accuracy": 0.18692323416471482,
      "num_tokens": 76138138.0,
      "step": 37580
    },
    {
      "entropy": 5.730501270294189,
      "epoch": 3.284540767281307,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00039403328053047847,
      "loss": 5.2883,
      "mean_token_accuracy": 0.20671509355306625,
      "num_tokens": 76148780.0,
      "step": 37585
    },
    {
      "entropy": 5.7327533721923825,
      "epoch": 3.284977715634012,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003940068361160913,
      "loss": 5.3336,
      "mean_token_accuracy": 0.2021001935005188,
      "num_tokens": 76158468.0,
      "step": 37590
    },
    {
      "entropy": 5.86924991607666,
      "epoch": 3.2854146639867166,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039398038941909066,
      "loss": 5.4581,
      "mean_token_accuracy": 0.1869281202554703,
      "num_tokens": 76168789.0,
      "step": 37595
    },
    {
      "entropy": 5.8451268672943115,
      "epoch": 3.2858516123394215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003939539404399839,
      "loss": 5.3757,
      "mean_token_accuracy": 0.18859937489032746,
      "num_tokens": 76179355.0,
      "step": 37600
    },
    {
      "entropy": 5.898609209060669,
      "epoch": 3.2862885606921264,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003939274891792781,
      "loss": 5.4437,
      "mean_token_accuracy": 0.18224112689495087,
      "num_tokens": 76189235.0,
      "step": 37605
    },
    {
      "entropy": 5.7635050296783445,
      "epoch": 3.286725509044831,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039390103563748086,
      "loss": 5.3573,
      "mean_token_accuracy": 0.19729894399642944,
      "num_tokens": 76200146.0,
      "step": 37610
    },
    {
      "entropy": 5.811476469039917,
      "epoch": 3.2871624573975358,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039387457981509937,
      "loss": 5.3949,
      "mean_token_accuracy": 0.19568008929491043,
      "num_tokens": 76209519.0,
      "step": 37615
    },
    {
      "entropy": 5.840524196624756,
      "epoch": 3.2875994057502402,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003938481217126412,
      "loss": 5.3746,
      "mean_token_accuracy": 0.19444707185029983,
      "num_tokens": 76219316.0,
      "step": 37620
    },
    {
      "entropy": 5.835630130767822,
      "epoch": 3.288036354102945,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003938216613306138,
      "loss": 5.3697,
      "mean_token_accuracy": 0.19310953766107558,
      "num_tokens": 76230707.0,
      "step": 37625
    },
    {
      "entropy": 5.836760663986206,
      "epoch": 3.2884733024556496,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00039379519866952463,
      "loss": 5.4586,
      "mean_token_accuracy": 0.1850483685731888,
      "num_tokens": 76240197.0,
      "step": 37630
    },
    {
      "entropy": 5.900100183486939,
      "epoch": 3.2889102508083545,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039376873372988123,
      "loss": 5.4449,
      "mean_token_accuracy": 0.18837497532367706,
      "num_tokens": 76249292.0,
      "step": 37635
    },
    {
      "entropy": 5.760368967056275,
      "epoch": 3.289347199161059,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039374226651219137,
      "loss": 5.3338,
      "mean_token_accuracy": 0.19620702266693116,
      "num_tokens": 76259598.0,
      "step": 37640
    },
    {
      "entropy": 5.7789013385772705,
      "epoch": 3.289784147513764,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003937157970169626,
      "loss": 5.4304,
      "mean_token_accuracy": 0.19527169764041902,
      "num_tokens": 76269624.0,
      "step": 37645
    },
    {
      "entropy": 5.873786544799804,
      "epoch": 3.2902210958664684,
      "grad_norm": 1.6875,
      "learning_rate": 0.0003936893252447026,
      "loss": 5.4367,
      "mean_token_accuracy": 0.1848479375243187,
      "num_tokens": 76280662.0,
      "step": 37650
    },
    {
      "entropy": 5.849452638626099,
      "epoch": 3.2906580442191733,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003936628511959193,
      "loss": 5.3858,
      "mean_token_accuracy": 0.19217920750379563,
      "num_tokens": 76291194.0,
      "step": 37655
    },
    {
      "entropy": 5.796678304672241,
      "epoch": 3.291094992571878,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003936363748711202,
      "loss": 5.3843,
      "mean_token_accuracy": 0.19907109290361405,
      "num_tokens": 76301926.0,
      "step": 37660
    },
    {
      "entropy": 5.866828870773316,
      "epoch": 3.2915319409245827,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003936098962708132,
      "loss": 5.4675,
      "mean_token_accuracy": 0.18707575500011445,
      "num_tokens": 76313126.0,
      "step": 37665
    },
    {
      "entropy": 5.830357456207276,
      "epoch": 3.2919688892772876,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003935834153955064,
      "loss": 5.3502,
      "mean_token_accuracy": 0.19749904721975325,
      "num_tokens": 76322133.0,
      "step": 37670
    },
    {
      "entropy": 5.770975828170776,
      "epoch": 3.292405837629992,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003935569322457075,
      "loss": 5.3289,
      "mean_token_accuracy": 0.20193633288145066,
      "num_tokens": 76332459.0,
      "step": 37675
    },
    {
      "entropy": 5.806227397918701,
      "epoch": 3.292842785982697,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003935304468219245,
      "loss": 5.4162,
      "mean_token_accuracy": 0.19008435308933258,
      "num_tokens": 76342510.0,
      "step": 37680
    },
    {
      "entropy": 5.832158517837525,
      "epoch": 3.2932797343354014,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003935039591246654,
      "loss": 5.4009,
      "mean_token_accuracy": 0.19066881239414216,
      "num_tokens": 76352164.0,
      "step": 37685
    },
    {
      "entropy": 5.834787607192993,
      "epoch": 3.2937166826881064,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003934774691544383,
      "loss": 5.3869,
      "mean_token_accuracy": 0.19817162156105042,
      "num_tokens": 76362141.0,
      "step": 37690
    },
    {
      "entropy": 5.850611734390259,
      "epoch": 3.294153631040811,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003934509769117513,
      "loss": 5.4656,
      "mean_token_accuracy": 0.1894437402486801,
      "num_tokens": 76372152.0,
      "step": 37695
    },
    {
      "entropy": 5.846781063079834,
      "epoch": 3.2945905793935157,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003934244823971124,
      "loss": 5.4122,
      "mean_token_accuracy": 0.19122074395418168,
      "num_tokens": 76383033.0,
      "step": 37700
    },
    {
      "entropy": 5.900230026245117,
      "epoch": 3.29502752774622,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039339798561102994,
      "loss": 5.391,
      "mean_token_accuracy": 0.19138945043087005,
      "num_tokens": 76393552.0,
      "step": 37705
    },
    {
      "entropy": 5.749441003799438,
      "epoch": 3.295464476098925,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003933714865540121,
      "loss": 5.3281,
      "mean_token_accuracy": 0.19590044915676116,
      "num_tokens": 76404363.0,
      "step": 37710
    },
    {
      "entropy": 5.8249640464782715,
      "epoch": 3.29590142445163,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039334498522656706,
      "loss": 5.3714,
      "mean_token_accuracy": 0.18949464112520217,
      "num_tokens": 76413483.0,
      "step": 37715
    },
    {
      "entropy": 5.731119537353516,
      "epoch": 3.2963383728043345,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003933184816292032,
      "loss": 5.4202,
      "mean_token_accuracy": 0.1878756269812584,
      "num_tokens": 76424786.0,
      "step": 37720
    },
    {
      "entropy": 5.764379835128784,
      "epoch": 3.2967753211570394,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003932919757624289,
      "loss": 5.3147,
      "mean_token_accuracy": 0.20139551162719727,
      "num_tokens": 76434923.0,
      "step": 37725
    },
    {
      "entropy": 5.853986072540283,
      "epoch": 3.297212269509744,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039326546762675254,
      "loss": 5.3919,
      "mean_token_accuracy": 0.192108453810215,
      "num_tokens": 76444999.0,
      "step": 37730
    },
    {
      "entropy": 5.8164793968200685,
      "epoch": 3.297649217862449,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039323895722268243,
      "loss": 5.3792,
      "mean_token_accuracy": 0.1958403468132019,
      "num_tokens": 76454869.0,
      "step": 37735
    },
    {
      "entropy": 5.7770552158355715,
      "epoch": 3.2980861662151533,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003932124445507273,
      "loss": 5.3931,
      "mean_token_accuracy": 0.19038794487714766,
      "num_tokens": 76465336.0,
      "step": 37740
    },
    {
      "entropy": 5.8585728168487545,
      "epoch": 3.298523114567858,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039318592961139545,
      "loss": 5.4087,
      "mean_token_accuracy": 0.19274099171161652,
      "num_tokens": 76476142.0,
      "step": 37745
    },
    {
      "entropy": 5.827232599258423,
      "epoch": 3.2989600629205627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039315941240519557,
      "loss": 5.4319,
      "mean_token_accuracy": 0.19770665168762208,
      "num_tokens": 76486194.0,
      "step": 37750
    },
    {
      "entropy": 5.877488327026367,
      "epoch": 3.2993970112732676,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003931328929326363,
      "loss": 5.4968,
      "mean_token_accuracy": 0.18145326822996138,
      "num_tokens": 76496243.0,
      "step": 37755
    },
    {
      "entropy": 5.825432872772216,
      "epoch": 3.299833959625972,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039310637119422607,
      "loss": 5.377,
      "mean_token_accuracy": 0.19604537189006804,
      "num_tokens": 76506661.0,
      "step": 37760
    },
    {
      "entropy": 5.837115859985351,
      "epoch": 3.300270907978677,
      "grad_norm": 1.0625,
      "learning_rate": 0.000393079847190474,
      "loss": 5.4148,
      "mean_token_accuracy": 0.19494773000478743,
      "num_tokens": 76516812.0,
      "step": 37765
    },
    {
      "entropy": 5.774717235565186,
      "epoch": 3.300707856331382,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003930533209218884,
      "loss": 5.4489,
      "mean_token_accuracy": 0.193337844312191,
      "num_tokens": 76527280.0,
      "step": 37770
    },
    {
      "entropy": 5.752525472640992,
      "epoch": 3.3011448046840863,
      "grad_norm": 1.125,
      "learning_rate": 0.00039302679238897826,
      "loss": 5.2921,
      "mean_token_accuracy": 0.20377057045698166,
      "num_tokens": 76536706.0,
      "step": 37775
    },
    {
      "entropy": 5.774086141586304,
      "epoch": 3.3015817530367912,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003930002615922525,
      "loss": 5.3087,
      "mean_token_accuracy": 0.1941764086484909,
      "num_tokens": 76546574.0,
      "step": 37780
    },
    {
      "entropy": 5.834968948364258,
      "epoch": 3.3020187013894957,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003929737285322198,
      "loss": 5.3942,
      "mean_token_accuracy": 0.19614170640707015,
      "num_tokens": 76556315.0,
      "step": 37785
    },
    {
      "entropy": 5.821174955368042,
      "epoch": 3.3024556497422006,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039294719320938923,
      "loss": 5.4694,
      "mean_token_accuracy": 0.1878700762987137,
      "num_tokens": 76565721.0,
      "step": 37790
    },
    {
      "entropy": 5.847973918914795,
      "epoch": 3.302892598094905,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039292065562426966,
      "loss": 5.4553,
      "mean_token_accuracy": 0.18588947355747223,
      "num_tokens": 76575524.0,
      "step": 37795
    },
    {
      "entropy": 5.903319215774536,
      "epoch": 3.30332954644761,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00039289411577737014,
      "loss": 5.4377,
      "mean_token_accuracy": 0.1867754802107811,
      "num_tokens": 76586653.0,
      "step": 37800
    },
    {
      "entropy": 5.8221625804901125,
      "epoch": 3.3037664948003145,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039286757366919963,
      "loss": 5.3957,
      "mean_token_accuracy": 0.192162224650383,
      "num_tokens": 76597176.0,
      "step": 37805
    },
    {
      "entropy": 5.782625532150268,
      "epoch": 3.3042034431530194,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003928410293002674,
      "loss": 5.4166,
      "mean_token_accuracy": 0.19203439205884934,
      "num_tokens": 76608010.0,
      "step": 37810
    },
    {
      "entropy": 5.7694450378417965,
      "epoch": 3.304640391505724,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003928144826710824,
      "loss": 5.3034,
      "mean_token_accuracy": 0.19880235493183135,
      "num_tokens": 76618029.0,
      "step": 37815
    },
    {
      "entropy": 5.873829412460327,
      "epoch": 3.3050773398584288,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039278793378215394,
      "loss": 5.4378,
      "mean_token_accuracy": 0.18887365907430648,
      "num_tokens": 76629529.0,
      "step": 37820
    },
    {
      "entropy": 5.862482261657715,
      "epoch": 3.3055142882111337,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003927613826339912,
      "loss": 5.4691,
      "mean_token_accuracy": 0.1873701199889183,
      "num_tokens": 76640488.0,
      "step": 37825
    },
    {
      "entropy": 5.767457771301269,
      "epoch": 3.305951236563838,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00039273482922710343,
      "loss": 5.2672,
      "mean_token_accuracy": 0.20418388098478318,
      "num_tokens": 76650271.0,
      "step": 37830
    },
    {
      "entropy": 5.7622950077056885,
      "epoch": 3.3063881849165426,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003927082735619999,
      "loss": 5.446,
      "mean_token_accuracy": 0.1869499132037163,
      "num_tokens": 76661413.0,
      "step": 37835
    },
    {
      "entropy": 5.820785284042358,
      "epoch": 3.3068251332692475,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00039268171563919,
      "loss": 5.4096,
      "mean_token_accuracy": 0.1892259895801544,
      "num_tokens": 76671055.0,
      "step": 37840
    },
    {
      "entropy": 5.954837417602539,
      "epoch": 3.3072620816219525,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003926551554591832,
      "loss": 5.5541,
      "mean_token_accuracy": 0.1805234432220459,
      "num_tokens": 76683472.0,
      "step": 37845
    },
    {
      "entropy": 5.839903163909912,
      "epoch": 3.307699029974657,
      "grad_norm": 0.9375,
      "learning_rate": 0.00039262859302248876,
      "loss": 5.3818,
      "mean_token_accuracy": 0.19021742790937424,
      "num_tokens": 76694641.0,
      "step": 37850
    },
    {
      "entropy": 5.827463817596436,
      "epoch": 3.308135978327362,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003926020283296164,
      "loss": 5.402,
      "mean_token_accuracy": 0.19066319465637208,
      "num_tokens": 76704737.0,
      "step": 37855
    },
    {
      "entropy": 5.7582704544067385,
      "epoch": 3.3085729266800663,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003925754613810754,
      "loss": 5.3815,
      "mean_token_accuracy": 0.19527734220027923,
      "num_tokens": 76714157.0,
      "step": 37860
    },
    {
      "entropy": 5.802470779418945,
      "epoch": 3.309009875032771,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039254889217737554,
      "loss": 5.4093,
      "mean_token_accuracy": 0.19230413287878037,
      "num_tokens": 76724488.0,
      "step": 37865
    },
    {
      "entropy": 5.869930982589722,
      "epoch": 3.3094468233854757,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003925223207190263,
      "loss": 5.4902,
      "mean_token_accuracy": 0.1908029943704605,
      "num_tokens": 76734700.0,
      "step": 37870
    },
    {
      "entropy": 5.825002193450928,
      "epoch": 3.3098837717381806,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003924957470065373,
      "loss": 5.3947,
      "mean_token_accuracy": 0.19156905710697175,
      "num_tokens": 76744578.0,
      "step": 37875
    },
    {
      "entropy": 5.835206699371338,
      "epoch": 3.310320720090885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039246917104041843,
      "loss": 5.3702,
      "mean_token_accuracy": 0.19757355004549026,
      "num_tokens": 76754690.0,
      "step": 37880
    },
    {
      "entropy": 5.955978631973267,
      "epoch": 3.31075766844359,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003924425928211792,
      "loss": 5.4499,
      "mean_token_accuracy": 0.18340236246585845,
      "num_tokens": 76765197.0,
      "step": 37885
    },
    {
      "entropy": 5.814328575134278,
      "epoch": 3.3111946167962945,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039241601234932954,
      "loss": 5.3899,
      "mean_token_accuracy": 0.19030979573726653,
      "num_tokens": 76774985.0,
      "step": 37890
    },
    {
      "entropy": 5.807517862319946,
      "epoch": 3.3116315651489994,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003923894296253792,
      "loss": 5.4362,
      "mean_token_accuracy": 0.18593111783266067,
      "num_tokens": 76784033.0,
      "step": 37895
    },
    {
      "entropy": 5.851450061798095,
      "epoch": 3.3120685135017043,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003923628446498382,
      "loss": 5.419,
      "mean_token_accuracy": 0.19001539349555968,
      "num_tokens": 76794826.0,
      "step": 37900
    },
    {
      "entropy": 5.78839225769043,
      "epoch": 3.3125054618544088,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003923362574232163,
      "loss": 5.3353,
      "mean_token_accuracy": 0.19825979471206664,
      "num_tokens": 76805429.0,
      "step": 37905
    },
    {
      "entropy": 5.830336332321167,
      "epoch": 3.3129424102071137,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003923096679460234,
      "loss": 5.412,
      "mean_token_accuracy": 0.1847053661942482,
      "num_tokens": 76814880.0,
      "step": 37910
    },
    {
      "entropy": 5.811465549468994,
      "epoch": 3.313379358559818,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003922830762187697,
      "loss": 5.4254,
      "mean_token_accuracy": 0.19269710332155227,
      "num_tokens": 76824828.0,
      "step": 37915
    },
    {
      "entropy": 5.8928686618804935,
      "epoch": 3.313816306912523,
      "grad_norm": 1.171875,
      "learning_rate": 0.000392256482241965,
      "loss": 5.5389,
      "mean_token_accuracy": 0.18377918004989624,
      "num_tokens": 76835140.0,
      "step": 37920
    },
    {
      "entropy": 5.839574337005615,
      "epoch": 3.3142532552652275,
      "grad_norm": 1.0,
      "learning_rate": 0.0003922298860161197,
      "loss": 5.4268,
      "mean_token_accuracy": 0.19162242412567138,
      "num_tokens": 76844718.0,
      "step": 37925
    },
    {
      "entropy": 5.911635541915894,
      "epoch": 3.3146902036179324,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039220328754174376,
      "loss": 5.4982,
      "mean_token_accuracy": 0.187483486533165,
      "num_tokens": 76854738.0,
      "step": 37930
    },
    {
      "entropy": 5.85977144241333,
      "epoch": 3.315127151970637,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039217668681934727,
      "loss": 5.4527,
      "mean_token_accuracy": 0.19086002707481384,
      "num_tokens": 76864939.0,
      "step": 37935
    },
    {
      "entropy": 5.839663934707642,
      "epoch": 3.315564100323342,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003921500838494405,
      "loss": 5.4589,
      "mean_token_accuracy": 0.18719289749860762,
      "num_tokens": 76875895.0,
      "step": 37940
    },
    {
      "entropy": 5.830728435516358,
      "epoch": 3.3160010486760463,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039212347863253386,
      "loss": 5.3934,
      "mean_token_accuracy": 0.18982948660850524,
      "num_tokens": 76885415.0,
      "step": 37945
    },
    {
      "entropy": 5.803454780578614,
      "epoch": 3.316437997028751,
      "grad_norm": 1.125,
      "learning_rate": 0.0003920968711691374,
      "loss": 5.4266,
      "mean_token_accuracy": 0.18697019815444946,
      "num_tokens": 76896961.0,
      "step": 37950
    },
    {
      "entropy": 5.880494546890259,
      "epoch": 3.316874945381456,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00039207026145976167,
      "loss": 5.4073,
      "mean_token_accuracy": 0.19443342089653015,
      "num_tokens": 76906521.0,
      "step": 37955
    },
    {
      "entropy": 5.812441349029541,
      "epoch": 3.3173118937341606,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039204364950491693,
      "loss": 5.3844,
      "mean_token_accuracy": 0.19008879810571672,
      "num_tokens": 76916971.0,
      "step": 37960
    },
    {
      "entropy": 5.8207722187042235,
      "epoch": 3.3177488420868655,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003920170353051137,
      "loss": 5.4317,
      "mean_token_accuracy": 0.19163972288370132,
      "num_tokens": 76928070.0,
      "step": 37965
    },
    {
      "entropy": 5.853847694396973,
      "epoch": 3.31818579043957,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00039199041886086233,
      "loss": 5.4504,
      "mean_token_accuracy": 0.19253736287355422,
      "num_tokens": 76938020.0,
      "step": 37970
    },
    {
      "entropy": 5.845793724060059,
      "epoch": 3.318622738792275,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003919638001726735,
      "loss": 5.427,
      "mean_token_accuracy": 0.19096343964338303,
      "num_tokens": 76949371.0,
      "step": 37975
    },
    {
      "entropy": 5.865645551681519,
      "epoch": 3.3190596871449793,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003919371792410578,
      "loss": 5.342,
      "mean_token_accuracy": 0.19490147083997728,
      "num_tokens": 76959120.0,
      "step": 37980
    },
    {
      "entropy": 5.759266567230225,
      "epoch": 3.3194966354976843,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039191055606652554,
      "loss": 5.3745,
      "mean_token_accuracy": 0.1947824165225029,
      "num_tokens": 76969493.0,
      "step": 37985
    },
    {
      "entropy": 5.824531221389771,
      "epoch": 3.3199335838503887,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039188393064958766,
      "loss": 5.402,
      "mean_token_accuracy": 0.18918275833129883,
      "num_tokens": 76979732.0,
      "step": 37990
    },
    {
      "entropy": 5.857993459701538,
      "epoch": 3.3203705322030936,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039185730299075473,
      "loss": 5.4529,
      "mean_token_accuracy": 0.18914936631917953,
      "num_tokens": 76990354.0,
      "step": 37995
    },
    {
      "entropy": 5.842125701904297,
      "epoch": 3.320807480555798,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039183067309053743,
      "loss": 5.3605,
      "mean_token_accuracy": 0.194862399995327,
      "num_tokens": 76999469.0,
      "step": 38000
    },
    {
      "entropy": 5.784500932693481,
      "epoch": 3.321244428908503,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003918040409494467,
      "loss": 5.3818,
      "mean_token_accuracy": 0.19419819861650467,
      "num_tokens": 77009499.0,
      "step": 38005
    },
    {
      "entropy": 5.816696643829346,
      "epoch": 3.321681377261208,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00039177740656799314,
      "loss": 5.4486,
      "mean_token_accuracy": 0.1758212625980377,
      "num_tokens": 77019426.0,
      "step": 38010
    },
    {
      "entropy": 5.841699838638306,
      "epoch": 3.3221183256139124,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003917507699466879,
      "loss": 5.3933,
      "mean_token_accuracy": 0.1879703015089035,
      "num_tokens": 77028274.0,
      "step": 38015
    },
    {
      "entropy": 5.780853366851806,
      "epoch": 3.3225552739666173,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003917241310860416,
      "loss": 5.3597,
      "mean_token_accuracy": 0.19534945785999297,
      "num_tokens": 77038790.0,
      "step": 38020
    },
    {
      "entropy": 5.835748338699341,
      "epoch": 3.322992222319322,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039169748998656536,
      "loss": 5.38,
      "mean_token_accuracy": 0.19594810903072357,
      "num_tokens": 77047799.0,
      "step": 38025
    },
    {
      "entropy": 5.797893905639649,
      "epoch": 3.3234291706720267,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003916708466487701,
      "loss": 5.4725,
      "mean_token_accuracy": 0.18277254700660706,
      "num_tokens": 77057080.0,
      "step": 38030
    },
    {
      "entropy": 5.900370693206787,
      "epoch": 3.323866119024731,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039164420107316676,
      "loss": 5.4264,
      "mean_token_accuracy": 0.18628084361553193,
      "num_tokens": 77066689.0,
      "step": 38035
    },
    {
      "entropy": 5.862688684463501,
      "epoch": 3.324303067377436,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003916175532602666,
      "loss": 5.436,
      "mean_token_accuracy": 0.1875259518623352,
      "num_tokens": 77076366.0,
      "step": 38040
    },
    {
      "entropy": 5.760222673416138,
      "epoch": 3.3247400157301406,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003915909032105808,
      "loss": 5.3613,
      "mean_token_accuracy": 0.1974795788526535,
      "num_tokens": 77085825.0,
      "step": 38045
    },
    {
      "entropy": 5.889794921875,
      "epoch": 3.3251769640828455,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003915642509246202,
      "loss": 5.3424,
      "mean_token_accuracy": 0.2006537899374962,
      "num_tokens": 77095249.0,
      "step": 38050
    },
    {
      "entropy": 5.75334792137146,
      "epoch": 3.32561391243555,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00039153759640289627,
      "loss": 5.3227,
      "mean_token_accuracy": 0.19405142813920975,
      "num_tokens": 77105323.0,
      "step": 38055
    },
    {
      "entropy": 5.8065680980682375,
      "epoch": 3.326050860788255,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039151093964592007,
      "loss": 5.4468,
      "mean_token_accuracy": 0.18461512625217438,
      "num_tokens": 77116055.0,
      "step": 38060
    },
    {
      "entropy": 5.8370603084564205,
      "epoch": 3.3264878091409598,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003914842806542031,
      "loss": 5.4062,
      "mean_token_accuracy": 0.19418004006147385,
      "num_tokens": 77127293.0,
      "step": 38065
    },
    {
      "entropy": 5.867822122573853,
      "epoch": 3.3269247574936642,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003914576194282566,
      "loss": 5.3993,
      "mean_token_accuracy": 0.18463489264249802,
      "num_tokens": 77136872.0,
      "step": 38070
    },
    {
      "entropy": 5.8947803497314455,
      "epoch": 3.327361705846369,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039143095596859184,
      "loss": 5.4278,
      "mean_token_accuracy": 0.18771193772554398,
      "num_tokens": 77146561.0,
      "step": 38075
    },
    {
      "entropy": 5.790988779067993,
      "epoch": 3.3277986541990736,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039140429027572043,
      "loss": 5.3845,
      "mean_token_accuracy": 0.19778178483247758,
      "num_tokens": 77156043.0,
      "step": 38080
    },
    {
      "entropy": 5.794365072250367,
      "epoch": 3.3282356025517785,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003913776223501537,
      "loss": 5.3961,
      "mean_token_accuracy": 0.19250958263874055,
      "num_tokens": 77165578.0,
      "step": 38085
    },
    {
      "entropy": 5.836132478713989,
      "epoch": 3.328672550904483,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003913509521924032,
      "loss": 5.4242,
      "mean_token_accuracy": 0.18804809749126433,
      "num_tokens": 77175105.0,
      "step": 38090
    },
    {
      "entropy": 5.813952445983887,
      "epoch": 3.329109499257188,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003913242798029805,
      "loss": 5.4305,
      "mean_token_accuracy": 0.18344497233629226,
      "num_tokens": 77185821.0,
      "step": 38095
    },
    {
      "entropy": 5.7765685558319095,
      "epoch": 3.3295464476098924,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003912976051823971,
      "loss": 5.3463,
      "mean_token_accuracy": 0.1930749922990799,
      "num_tokens": 77195028.0,
      "step": 38100
    },
    {
      "entropy": 5.839330101013184,
      "epoch": 3.3299833959625973,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003912709283311648,
      "loss": 5.3653,
      "mean_token_accuracy": 0.18736871629953383,
      "num_tokens": 77206317.0,
      "step": 38105
    },
    {
      "entropy": 5.791133546829224,
      "epoch": 3.3304203443153018,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003912442492497951,
      "loss": 5.339,
      "mean_token_accuracy": 0.19514214098453522,
      "num_tokens": 77216485.0,
      "step": 38110
    },
    {
      "entropy": 5.801997518539428,
      "epoch": 3.3308572926680067,
      "grad_norm": 1.125,
      "learning_rate": 0.0003912175679387997,
      "loss": 5.3789,
      "mean_token_accuracy": 0.19757622182369233,
      "num_tokens": 77226457.0,
      "step": 38115
    },
    {
      "entropy": 5.745598554611206,
      "epoch": 3.3312942410207116,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003911908843986906,
      "loss": 5.3549,
      "mean_token_accuracy": 0.19862720370292664,
      "num_tokens": 77237673.0,
      "step": 38120
    },
    {
      "entropy": 5.814712715148926,
      "epoch": 3.331731189373416,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003911641986299793,
      "loss": 5.3439,
      "mean_token_accuracy": 0.19288630187511444,
      "num_tokens": 77247620.0,
      "step": 38125
    },
    {
      "entropy": 5.850339126586914,
      "epoch": 3.3321681377261205,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039113751063317795,
      "loss": 5.3779,
      "mean_token_accuracy": 0.19039501696825029,
      "num_tokens": 77257739.0,
      "step": 38130
    },
    {
      "entropy": 5.785992765426636,
      "epoch": 3.3326050860788254,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003911108204087982,
      "loss": 5.4328,
      "mean_token_accuracy": 0.18852743953466417,
      "num_tokens": 77268917.0,
      "step": 38135
    },
    {
      "entropy": 5.810041713714599,
      "epoch": 3.3330420344315304,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003910841279573521,
      "loss": 5.4355,
      "mean_token_accuracy": 0.1925508886575699,
      "num_tokens": 77279225.0,
      "step": 38140
    },
    {
      "entropy": 5.821598196029663,
      "epoch": 3.333478982784235,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003910574332793516,
      "loss": 5.367,
      "mean_token_accuracy": 0.1950889050960541,
      "num_tokens": 77289504.0,
      "step": 38145
    },
    {
      "entropy": 5.823986577987671,
      "epoch": 3.3339159311369397,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039103073637530877,
      "loss": 5.4263,
      "mean_token_accuracy": 0.19353846758604049,
      "num_tokens": 77299663.0,
      "step": 38150
    },
    {
      "entropy": 5.816971921920777,
      "epoch": 3.334352879489644,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039100403724573555,
      "loss": 5.3826,
      "mean_token_accuracy": 0.18908989131450654,
      "num_tokens": 77308755.0,
      "step": 38155
    },
    {
      "entropy": 5.804389333724975,
      "epoch": 3.334789827842349,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003909773358911441,
      "loss": 5.3463,
      "mean_token_accuracy": 0.19262565821409225,
      "num_tokens": 77318351.0,
      "step": 38160
    },
    {
      "entropy": 5.78879566192627,
      "epoch": 3.3352267761950536,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039095063231204657,
      "loss": 5.258,
      "mean_token_accuracy": 0.20058357268571853,
      "num_tokens": 77328091.0,
      "step": 38165
    },
    {
      "entropy": 5.733743047714233,
      "epoch": 3.3356637245477585,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003909239265089552,
      "loss": 5.3739,
      "mean_token_accuracy": 0.1950946033000946,
      "num_tokens": 77338209.0,
      "step": 38170
    },
    {
      "entropy": 5.827268075942993,
      "epoch": 3.336100672900463,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039089721848238216,
      "loss": 5.3163,
      "mean_token_accuracy": 0.1949080392718315,
      "num_tokens": 77349811.0,
      "step": 38175
    },
    {
      "entropy": 5.870585298538208,
      "epoch": 3.336537621253168,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003908705082328397,
      "loss": 5.4816,
      "mean_token_accuracy": 0.18459260165691377,
      "num_tokens": 77359995.0,
      "step": 38180
    },
    {
      "entropy": 5.776777696609497,
      "epoch": 3.3369745696058724,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003908437957608403,
      "loss": 5.3251,
      "mean_token_accuracy": 0.1983663558959961,
      "num_tokens": 77371149.0,
      "step": 38185
    },
    {
      "entropy": 5.792235946655273,
      "epoch": 3.3374115179585773,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003908170810668961,
      "loss": 5.386,
      "mean_token_accuracy": 0.18913600891828536,
      "num_tokens": 77381107.0,
      "step": 38190
    },
    {
      "entropy": 5.8136559009552,
      "epoch": 3.337848466311282,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039079036415151967,
      "loss": 5.4772,
      "mean_token_accuracy": 0.17879953980445862,
      "num_tokens": 77391982.0,
      "step": 38195
    },
    {
      "entropy": 5.833231973648071,
      "epoch": 3.3382854146639866,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003907636450152232,
      "loss": 5.3856,
      "mean_token_accuracy": 0.1921495407819748,
      "num_tokens": 77403129.0,
      "step": 38200
    },
    {
      "entropy": 5.836671209335327,
      "epoch": 3.3387223630166916,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003907369236585195,
      "loss": 5.4528,
      "mean_token_accuracy": 0.18895686119794847,
      "num_tokens": 77412562.0,
      "step": 38205
    },
    {
      "entropy": 5.838816833496094,
      "epoch": 3.339159311369396,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000390710200081921,
      "loss": 5.4192,
      "mean_token_accuracy": 0.18941676020622253,
      "num_tokens": 77422693.0,
      "step": 38210
    },
    {
      "entropy": 5.827454996109009,
      "epoch": 3.339596259722101,
      "grad_norm": 1.0,
      "learning_rate": 0.0003906834742859402,
      "loss": 5.3938,
      "mean_token_accuracy": 0.1921449437737465,
      "num_tokens": 77432455.0,
      "step": 38215
    },
    {
      "entropy": 5.804687261581421,
      "epoch": 3.3400332080748054,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003906567462710897,
      "loss": 5.412,
      "mean_token_accuracy": 0.1888372629880905,
      "num_tokens": 77442785.0,
      "step": 38220
    },
    {
      "entropy": 5.8363807678222654,
      "epoch": 3.3404701564275103,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039063001603788217,
      "loss": 5.3848,
      "mean_token_accuracy": 0.19037923365831375,
      "num_tokens": 77453307.0,
      "step": 38225
    },
    {
      "entropy": 5.860900831222534,
      "epoch": 3.340907104780215,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003906032835868304,
      "loss": 5.5143,
      "mean_token_accuracy": 0.18816453367471694,
      "num_tokens": 77463851.0,
      "step": 38230
    },
    {
      "entropy": 5.862775278091431,
      "epoch": 3.3413440531329197,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039057654891844716,
      "loss": 5.4311,
      "mean_token_accuracy": 0.18537207394838334,
      "num_tokens": 77473391.0,
      "step": 38235
    },
    {
      "entropy": 5.841175127029419,
      "epoch": 3.341781001485624,
      "grad_norm": 1.109375,
      "learning_rate": 0.000390549812033245,
      "loss": 5.404,
      "mean_token_accuracy": 0.18838237524032592,
      "num_tokens": 77482750.0,
      "step": 38240
    },
    {
      "entropy": 5.789202785491943,
      "epoch": 3.342217949838329,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000390523072931737,
      "loss": 5.3588,
      "mean_token_accuracy": 0.188697911798954,
      "num_tokens": 77492637.0,
      "step": 38245
    },
    {
      "entropy": 5.860478353500366,
      "epoch": 3.342654898191034,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003904963316144359,
      "loss": 5.4587,
      "mean_token_accuracy": 0.18356575667858124,
      "num_tokens": 77502877.0,
      "step": 38250
    },
    {
      "entropy": 5.811665630340576,
      "epoch": 3.3430918465437385,
      "grad_norm": 1.125,
      "learning_rate": 0.00039046958808185454,
      "loss": 5.401,
      "mean_token_accuracy": 0.1965809717774391,
      "num_tokens": 77513630.0,
      "step": 38255
    },
    {
      "entropy": 5.796649932861328,
      "epoch": 3.3435287948964434,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039044284233450614,
      "loss": 5.3451,
      "mean_token_accuracy": 0.19444118440151215,
      "num_tokens": 77523801.0,
      "step": 38260
    },
    {
      "entropy": 5.802445363998413,
      "epoch": 3.343965743249148,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039041609437290346,
      "loss": 5.3986,
      "mean_token_accuracy": 0.19640681147575378,
      "num_tokens": 77533985.0,
      "step": 38265
    },
    {
      "entropy": 5.8045014381408695,
      "epoch": 3.3444026916018528,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003903893441975596,
      "loss": 5.3335,
      "mean_token_accuracy": 0.20106881558895112,
      "num_tokens": 77543713.0,
      "step": 38270
    },
    {
      "entropy": 5.748989057540894,
      "epoch": 3.3448396399545572,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0003903625918089877,
      "loss": 5.4115,
      "mean_token_accuracy": 0.19179010391235352,
      "num_tokens": 77553945.0,
      "step": 38275
    },
    {
      "entropy": 5.739352083206176,
      "epoch": 3.345276588307262,
      "grad_norm": 1.734375,
      "learning_rate": 0.0003903358372077008,
      "loss": 5.3411,
      "mean_token_accuracy": 0.19319756478071212,
      "num_tokens": 77563615.0,
      "step": 38280
    },
    {
      "entropy": 5.845353841781616,
      "epoch": 3.3457135366599666,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003903090803942121,
      "loss": 5.3527,
      "mean_token_accuracy": 0.1965511918067932,
      "num_tokens": 77573886.0,
      "step": 38285
    },
    {
      "entropy": 5.821716260910034,
      "epoch": 3.3461504850126715,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039028232136903483,
      "loss": 5.4234,
      "mean_token_accuracy": 0.18695738166570663,
      "num_tokens": 77585017.0,
      "step": 38290
    },
    {
      "entropy": 5.797801923751831,
      "epoch": 3.346587433365376,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00039025556013268226,
      "loss": 5.3927,
      "mean_token_accuracy": 0.19356644451618193,
      "num_tokens": 77594605.0,
      "step": 38295
    },
    {
      "entropy": 5.86253080368042,
      "epoch": 3.347024381718081,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039022879668566764,
      "loss": 5.3346,
      "mean_token_accuracy": 0.19507890492677687,
      "num_tokens": 77603262.0,
      "step": 38300
    },
    {
      "entropy": 5.747863626480102,
      "epoch": 3.347461330070786,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003902020310285043,
      "loss": 5.36,
      "mean_token_accuracy": 0.18832706212997435,
      "num_tokens": 77612963.0,
      "step": 38305
    },
    {
      "entropy": 5.802910089492798,
      "epoch": 3.3478982784234903,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003901752631617056,
      "loss": 5.3873,
      "mean_token_accuracy": 0.18641684353351592,
      "num_tokens": 77623237.0,
      "step": 38310
    },
    {
      "entropy": 5.820637130737305,
      "epoch": 3.348335226776195,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003901484930857851,
      "loss": 5.3446,
      "mean_token_accuracy": 0.19495671689510347,
      "num_tokens": 77632513.0,
      "step": 38315
    },
    {
      "entropy": 5.753305721282959,
      "epoch": 3.3487721751288997,
      "grad_norm": 1.125,
      "learning_rate": 0.0003901217208012561,
      "loss": 5.3214,
      "mean_token_accuracy": 0.197629177570343,
      "num_tokens": 77643160.0,
      "step": 38320
    },
    {
      "entropy": 5.789045095443726,
      "epoch": 3.3492091234816046,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039009494630863216,
      "loss": 5.4036,
      "mean_token_accuracy": 0.19495292901992797,
      "num_tokens": 77653311.0,
      "step": 38325
    },
    {
      "entropy": 5.848870038986206,
      "epoch": 3.349646071834309,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039006816960842695,
      "loss": 5.3847,
      "mean_token_accuracy": 0.1946505919098854,
      "num_tokens": 77663142.0,
      "step": 38330
    },
    {
      "entropy": 5.849627733230591,
      "epoch": 3.350083020187014,
      "grad_norm": 1.734375,
      "learning_rate": 0.0003900413907011538,
      "loss": 5.4494,
      "mean_token_accuracy": 0.1828645721077919,
      "num_tokens": 77675143.0,
      "step": 38335
    },
    {
      "entropy": 5.7859972476959225,
      "epoch": 3.3505199685397185,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003900146095873266,
      "loss": 5.39,
      "mean_token_accuracy": 0.19525964707136154,
      "num_tokens": 77685464.0,
      "step": 38340
    },
    {
      "entropy": 5.85359148979187,
      "epoch": 3.3509569168924234,
      "grad_norm": 1.0,
      "learning_rate": 0.0003899878262674588,
      "loss": 5.4081,
      "mean_token_accuracy": 0.19081859290599823,
      "num_tokens": 77695749.0,
      "step": 38345
    },
    {
      "entropy": 5.819074487686157,
      "epoch": 3.351393865245128,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003899610407420643,
      "loss": 5.3434,
      "mean_token_accuracy": 0.18737497180700302,
      "num_tokens": 77705678.0,
      "step": 38350
    },
    {
      "entropy": 5.867503833770752,
      "epoch": 3.3518308135978327,
      "grad_norm": 2.015625,
      "learning_rate": 0.0003899342530116569,
      "loss": 5.4211,
      "mean_token_accuracy": 0.18772393763065337,
      "num_tokens": 77716481.0,
      "step": 38355
    },
    {
      "entropy": 5.806518650054931,
      "epoch": 3.3522677619505377,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003899074630767502,
      "loss": 5.4483,
      "mean_token_accuracy": 0.1865553081035614,
      "num_tokens": 77726977.0,
      "step": 38360
    },
    {
      "entropy": 5.760276842117309,
      "epoch": 3.352704710303242,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038988067093785817,
      "loss": 5.3138,
      "mean_token_accuracy": 0.202323217689991,
      "num_tokens": 77736624.0,
      "step": 38365
    },
    {
      "entropy": 5.814867305755615,
      "epoch": 3.353141658655947,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038985387659549466,
      "loss": 5.3799,
      "mean_token_accuracy": 0.2018822729587555,
      "num_tokens": 77745964.0,
      "step": 38370
    },
    {
      "entropy": 5.842219972610474,
      "epoch": 3.3535786070086515,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038982708005017366,
      "loss": 5.4065,
      "mean_token_accuracy": 0.1895868092775345,
      "num_tokens": 77755801.0,
      "step": 38375
    },
    {
      "entropy": 5.8221269130706785,
      "epoch": 3.3540155553613564,
      "grad_norm": 1.125,
      "learning_rate": 0.0003898002813024091,
      "loss": 5.4328,
      "mean_token_accuracy": 0.18398652821779252,
      "num_tokens": 77765604.0,
      "step": 38380
    },
    {
      "entropy": 5.878724050521851,
      "epoch": 3.354452503714061,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003897734803527149,
      "loss": 5.4365,
      "mean_token_accuracy": 0.19272028803825378,
      "num_tokens": 77775754.0,
      "step": 38385
    },
    {
      "entropy": 5.927569341659546,
      "epoch": 3.354889452066766,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038974667720160534,
      "loss": 5.5554,
      "mean_token_accuracy": 0.17910337150096894,
      "num_tokens": 77786063.0,
      "step": 38390
    },
    {
      "entropy": 5.770317697525025,
      "epoch": 3.3553264004194703,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038971987184959427,
      "loss": 5.333,
      "mean_token_accuracy": 0.20139125883579254,
      "num_tokens": 77795217.0,
      "step": 38395
    },
    {
      "entropy": 5.767702674865722,
      "epoch": 3.355763348772175,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000389693064297196,
      "loss": 5.4403,
      "mean_token_accuracy": 0.18967166990041734,
      "num_tokens": 77804637.0,
      "step": 38400
    },
    {
      "entropy": 5.817670202255249,
      "epoch": 3.3562002971248797,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003896662545449247,
      "loss": 5.426,
      "mean_token_accuracy": 0.18842926025390624,
      "num_tokens": 77816048.0,
      "step": 38405
    },
    {
      "entropy": 5.839239072799683,
      "epoch": 3.3566372454775846,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003896394425932944,
      "loss": 5.3579,
      "mean_token_accuracy": 0.19074534475803376,
      "num_tokens": 77826385.0,
      "step": 38410
    },
    {
      "entropy": 5.784382009506226,
      "epoch": 3.3570741938302895,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038961262844281966,
      "loss": 5.4401,
      "mean_token_accuracy": 0.1842538982629776,
      "num_tokens": 77835744.0,
      "step": 38415
    },
    {
      "entropy": 5.824497556686401,
      "epoch": 3.357511142182994,
      "grad_norm": 1.0,
      "learning_rate": 0.0003895858120940146,
      "loss": 5.4008,
      "mean_token_accuracy": 0.19505189955234528,
      "num_tokens": 77845883.0,
      "step": 38420
    },
    {
      "entropy": 5.8332531452178955,
      "epoch": 3.357948090535699,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038955899354739364,
      "loss": 5.3739,
      "mean_token_accuracy": 0.1895967721939087,
      "num_tokens": 77856213.0,
      "step": 38425
    },
    {
      "entropy": 5.833643865585327,
      "epoch": 3.3583850388884033,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038953217280347113,
      "loss": 5.4522,
      "mean_token_accuracy": 0.19021186232566833,
      "num_tokens": 77867266.0,
      "step": 38430
    },
    {
      "entropy": 5.887501049041748,
      "epoch": 3.3588219872411083,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003895053498627616,
      "loss": 5.4848,
      "mean_token_accuracy": 0.19080362915992738,
      "num_tokens": 77878358.0,
      "step": 38435
    },
    {
      "entropy": 5.894864702224732,
      "epoch": 3.3592589355938127,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003894785247257794,
      "loss": 5.4921,
      "mean_token_accuracy": 0.18704378455877305,
      "num_tokens": 77889191.0,
      "step": 38440
    },
    {
      "entropy": 5.816532039642334,
      "epoch": 3.3596958839465176,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003894516973930392,
      "loss": 5.4258,
      "mean_token_accuracy": 0.1937419965863228,
      "num_tokens": 77898937.0,
      "step": 38445
    },
    {
      "entropy": 5.832320308685302,
      "epoch": 3.360132832299222,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003894248678650553,
      "loss": 5.3323,
      "mean_token_accuracy": 0.19270540475845338,
      "num_tokens": 77907440.0,
      "step": 38450
    },
    {
      "entropy": 5.823440456390381,
      "epoch": 3.360569780651927,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003893980361423426,
      "loss": 5.3993,
      "mean_token_accuracy": 0.19406728595495223,
      "num_tokens": 77917471.0,
      "step": 38455
    },
    {
      "entropy": 5.853096771240234,
      "epoch": 3.3610067290046315,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003893712022254156,
      "loss": 5.351,
      "mean_token_accuracy": 0.1962318792939186,
      "num_tokens": 77927292.0,
      "step": 38460
    },
    {
      "entropy": 5.850928544998169,
      "epoch": 3.3614436773573364,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038934436611478895,
      "loss": 5.4577,
      "mean_token_accuracy": 0.18931830823421478,
      "num_tokens": 77937870.0,
      "step": 38465
    },
    {
      "entropy": 5.8316985130310055,
      "epoch": 3.3618806257100413,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038931752781097757,
      "loss": 5.3753,
      "mean_token_accuracy": 0.19456178396940232,
      "num_tokens": 77947369.0,
      "step": 38470
    },
    {
      "entropy": 5.847871112823486,
      "epoch": 3.362317574062746,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038929068731449604,
      "loss": 5.4509,
      "mean_token_accuracy": 0.1868853434920311,
      "num_tokens": 77958314.0,
      "step": 38475
    },
    {
      "entropy": 5.870884704589844,
      "epoch": 3.3627545224154503,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003892638446258592,
      "loss": 5.5111,
      "mean_token_accuracy": 0.18142468780279158,
      "num_tokens": 77968734.0,
      "step": 38480
    },
    {
      "entropy": 5.868155574798584,
      "epoch": 3.363191470768155,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000389236999745582,
      "loss": 5.4206,
      "mean_token_accuracy": 0.18624064028263093,
      "num_tokens": 77978964.0,
      "step": 38485
    },
    {
      "entropy": 5.837673377990723,
      "epoch": 3.36362841912086,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003892101526741792,
      "loss": 5.3597,
      "mean_token_accuracy": 0.19742391854524613,
      "num_tokens": 77989530.0,
      "step": 38490
    },
    {
      "entropy": 5.810611009597778,
      "epoch": 3.3640653674735645,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000389183303412166,
      "loss": 5.3838,
      "mean_token_accuracy": 0.19674962610006333,
      "num_tokens": 77999870.0,
      "step": 38495
    },
    {
      "entropy": 5.809145784378051,
      "epoch": 3.3645023158262695,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038915645196005704,
      "loss": 5.407,
      "mean_token_accuracy": 0.19118477255105973,
      "num_tokens": 78010047.0,
      "step": 38500
    },
    {
      "entropy": 5.822990131378174,
      "epoch": 3.364939264178974,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003891295983183676,
      "loss": 5.3366,
      "mean_token_accuracy": 0.19647634625434876,
      "num_tokens": 78019834.0,
      "step": 38505
    },
    {
      "entropy": 5.829148244857788,
      "epoch": 3.365376212531679,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038910274248761263,
      "loss": 5.3817,
      "mean_token_accuracy": 0.1913413107395172,
      "num_tokens": 78029639.0,
      "step": 38510
    },
    {
      "entropy": 5.807842874526978,
      "epoch": 3.3658131608843833,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003890758844683072,
      "loss": 5.3538,
      "mean_token_accuracy": 0.1944003537297249,
      "num_tokens": 78039638.0,
      "step": 38515
    },
    {
      "entropy": 5.777235794067383,
      "epoch": 3.3662501092370882,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00038904902426096664,
      "loss": 5.3787,
      "mean_token_accuracy": 0.1929403141140938,
      "num_tokens": 78050666.0,
      "step": 38520
    },
    {
      "entropy": 5.785884714126587,
      "epoch": 3.3666870575897927,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000389022161866106,
      "loss": 5.3111,
      "mean_token_accuracy": 0.20303037017583847,
      "num_tokens": 78060292.0,
      "step": 38525
    },
    {
      "entropy": 5.804590845108033,
      "epoch": 3.3671240059424976,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038899529728424057,
      "loss": 5.4118,
      "mean_token_accuracy": 0.19162963181734086,
      "num_tokens": 78069900.0,
      "step": 38530
    },
    {
      "entropy": 5.766816091537476,
      "epoch": 3.367560954295202,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003889684305158856,
      "loss": 5.3377,
      "mean_token_accuracy": 0.19103807657957078,
      "num_tokens": 78078788.0,
      "step": 38535
    },
    {
      "entropy": 5.815062522888184,
      "epoch": 3.367997902647907,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003889415615615564,
      "loss": 5.4171,
      "mean_token_accuracy": 0.19146147668361663,
      "num_tokens": 78087612.0,
      "step": 38540
    },
    {
      "entropy": 5.810218620300293,
      "epoch": 3.368434851000612,
      "grad_norm": 1.125,
      "learning_rate": 0.00038891469042176834,
      "loss": 5.3595,
      "mean_token_accuracy": 0.1932126209139824,
      "num_tokens": 78098399.0,
      "step": 38545
    },
    {
      "entropy": 5.843989133834839,
      "epoch": 3.3688717993533164,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003888878170970369,
      "loss": 5.4096,
      "mean_token_accuracy": 0.18290835916996,
      "num_tokens": 78107956.0,
      "step": 38550
    },
    {
      "entropy": 5.858592510223389,
      "epoch": 3.3693087477060213,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038886094158787726,
      "loss": 5.4467,
      "mean_token_accuracy": 0.18720406740903855,
      "num_tokens": 78119613.0,
      "step": 38555
    },
    {
      "entropy": 5.862892198562622,
      "epoch": 3.3697456960587258,
      "grad_norm": 1.25,
      "learning_rate": 0.0003888340638948052,
      "loss": 5.4008,
      "mean_token_accuracy": 0.19444627761840821,
      "num_tokens": 78128986.0,
      "step": 38560
    },
    {
      "entropy": 5.794154691696167,
      "epoch": 3.3701826444114307,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038880718401833623,
      "loss": 5.4317,
      "mean_token_accuracy": 0.19286461025476456,
      "num_tokens": 78139085.0,
      "step": 38565
    },
    {
      "entropy": 5.755275726318359,
      "epoch": 3.370619592764135,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003887803019589858,
      "loss": 5.3024,
      "mean_token_accuracy": 0.20014226734638213,
      "num_tokens": 78148780.0,
      "step": 38570
    },
    {
      "entropy": 5.835796642303467,
      "epoch": 3.37105654111684,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003887534177172695,
      "loss": 5.4115,
      "mean_token_accuracy": 0.1904605895280838,
      "num_tokens": 78159021.0,
      "step": 38575
    },
    {
      "entropy": 5.818333721160888,
      "epoch": 3.3714934894695445,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038872653129370303,
      "loss": 5.3956,
      "mean_token_accuracy": 0.1984315648674965,
      "num_tokens": 78169553.0,
      "step": 38580
    },
    {
      "entropy": 5.7742475986480715,
      "epoch": 3.3719304378222494,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038869964268880224,
      "loss": 5.3086,
      "mean_token_accuracy": 0.19637836664915084,
      "num_tokens": 78179631.0,
      "step": 38585
    },
    {
      "entropy": 5.824590015411377,
      "epoch": 3.372367386174954,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003886727519030826,
      "loss": 5.3462,
      "mean_token_accuracy": 0.18911227136850356,
      "num_tokens": 78189497.0,
      "step": 38590
    },
    {
      "entropy": 5.726376295089722,
      "epoch": 3.372804334527659,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00038864585893706006,
      "loss": 5.3726,
      "mean_token_accuracy": 0.19380147755146027,
      "num_tokens": 78200786.0,
      "step": 38595
    },
    {
      "entropy": 5.81823034286499,
      "epoch": 3.3732412828803637,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038861896379125034,
      "loss": 5.3786,
      "mean_token_accuracy": 0.19356072843074798,
      "num_tokens": 78209783.0,
      "step": 38600
    },
    {
      "entropy": 5.772899866104126,
      "epoch": 3.373678231233068,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003885920664661695,
      "loss": 5.3867,
      "mean_token_accuracy": 0.19432875961065293,
      "num_tokens": 78218988.0,
      "step": 38605
    },
    {
      "entropy": 5.799957895278931,
      "epoch": 3.374115179585773,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003885651669623331,
      "loss": 5.38,
      "mean_token_accuracy": 0.19438841491937636,
      "num_tokens": 78228545.0,
      "step": 38610
    },
    {
      "entropy": 5.817229986190796,
      "epoch": 3.3745521279384776,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003885382652802574,
      "loss": 5.3371,
      "mean_token_accuracy": 0.20289178043603898,
      "num_tokens": 78238167.0,
      "step": 38615
    },
    {
      "entropy": 5.7609992027282715,
      "epoch": 3.3749890762911825,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003885113614204583,
      "loss": 5.339,
      "mean_token_accuracy": 0.20096780955791474,
      "num_tokens": 78248257.0,
      "step": 38620
    },
    {
      "entropy": 5.822871112823487,
      "epoch": 3.375426024643887,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038848445538345173,
      "loss": 5.4112,
      "mean_token_accuracy": 0.191854764521122,
      "num_tokens": 78257406.0,
      "step": 38625
    },
    {
      "entropy": 5.790321683883667,
      "epoch": 3.375862972996592,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003884575471697539,
      "loss": 5.2877,
      "mean_token_accuracy": 0.1969161331653595,
      "num_tokens": 78267022.0,
      "step": 38630
    },
    {
      "entropy": 5.829236030578613,
      "epoch": 3.3762999213492964,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038843063677988093,
      "loss": 5.416,
      "mean_token_accuracy": 0.1892804205417633,
      "num_tokens": 78277584.0,
      "step": 38635
    },
    {
      "entropy": 5.816058015823364,
      "epoch": 3.3767368697020013,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003884037242143488,
      "loss": 5.4008,
      "mean_token_accuracy": 0.1942979171872139,
      "num_tokens": 78288218.0,
      "step": 38640
    },
    {
      "entropy": 5.8232310771942135,
      "epoch": 3.3771738180547057,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003883768094736738,
      "loss": 5.3762,
      "mean_token_accuracy": 0.19676986187696457,
      "num_tokens": 78297585.0,
      "step": 38645
    },
    {
      "entropy": 5.781335735321045,
      "epoch": 3.3776107664074106,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003883498925583723,
      "loss": 5.3875,
      "mean_token_accuracy": 0.1892029270529747,
      "num_tokens": 78308406.0,
      "step": 38650
    },
    {
      "entropy": 5.80248384475708,
      "epoch": 3.3780477147601156,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003883229734689604,
      "loss": 5.3448,
      "mean_token_accuracy": 0.1970770999789238,
      "num_tokens": 78318151.0,
      "step": 38655
    },
    {
      "entropy": 5.8196173191070555,
      "epoch": 3.37848466311282,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003882960522059545,
      "loss": 5.3425,
      "mean_token_accuracy": 0.19485474079847337,
      "num_tokens": 78327316.0,
      "step": 38660
    },
    {
      "entropy": 5.752933979034424,
      "epoch": 3.378921611465525,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003882691287698709,
      "loss": 5.3331,
      "mean_token_accuracy": 0.19770449846982957,
      "num_tokens": 78337564.0,
      "step": 38665
    },
    {
      "entropy": 5.823856353759766,
      "epoch": 3.3793585598182294,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003882422031612261,
      "loss": 5.4044,
      "mean_token_accuracy": 0.1928711161017418,
      "num_tokens": 78347796.0,
      "step": 38670
    },
    {
      "entropy": 5.894328212738037,
      "epoch": 3.3797955081709343,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003882152753805366,
      "loss": 5.4675,
      "mean_token_accuracy": 0.1810987710952759,
      "num_tokens": 78357269.0,
      "step": 38675
    },
    {
      "entropy": 5.847375726699829,
      "epoch": 3.380232456523639,
      "grad_norm": 0.9375,
      "learning_rate": 0.00038818834542831863,
      "loss": 5.3556,
      "mean_token_accuracy": 0.19622703939676284,
      "num_tokens": 78368001.0,
      "step": 38680
    },
    {
      "entropy": 5.746448040008545,
      "epoch": 3.3806694048763437,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000388161413305089,
      "loss": 5.2961,
      "mean_token_accuracy": 0.19669006317853927,
      "num_tokens": 78377908.0,
      "step": 38685
    },
    {
      "entropy": 5.784519100189209,
      "epoch": 3.381106353229048,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003881344790113641,
      "loss": 5.4612,
      "mean_token_accuracy": 0.1867654874920845,
      "num_tokens": 78389450.0,
      "step": 38690
    },
    {
      "entropy": 5.8492724895477295,
      "epoch": 3.381543301581753,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038810754254766066,
      "loss": 5.3871,
      "mean_token_accuracy": 0.1921858862042427,
      "num_tokens": 78400397.0,
      "step": 38695
    },
    {
      "entropy": 5.783969163894653,
      "epoch": 3.3819802499344576,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038808060391449525,
      "loss": 5.3178,
      "mean_token_accuracy": 0.1952036902308464,
      "num_tokens": 78410322.0,
      "step": 38700
    },
    {
      "entropy": 5.814543342590332,
      "epoch": 3.3824171982871625,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003880536631123847,
      "loss": 5.4185,
      "mean_token_accuracy": 0.1839195415377617,
      "num_tokens": 78420816.0,
      "step": 38705
    },
    {
      "entropy": 5.761245918273926,
      "epoch": 3.3828541466398674,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003880267201418455,
      "loss": 5.3999,
      "mean_token_accuracy": 0.19557118713855742,
      "num_tokens": 78430217.0,
      "step": 38710
    },
    {
      "entropy": 5.873152446746826,
      "epoch": 3.383291094992572,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003879997750033946,
      "loss": 5.3849,
      "mean_token_accuracy": 0.19430710822343827,
      "num_tokens": 78439938.0,
      "step": 38715
    },
    {
      "entropy": 5.828181171417237,
      "epoch": 3.3837280433452768,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003879728276975489,
      "loss": 5.374,
      "mean_token_accuracy": 0.19113788306713103,
      "num_tokens": 78450805.0,
      "step": 38720
    },
    {
      "entropy": 5.77551679611206,
      "epoch": 3.3841649916979812,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00038794587822482506,
      "loss": 5.3687,
      "mean_token_accuracy": 0.19577304720878602,
      "num_tokens": 78460653.0,
      "step": 38725
    },
    {
      "entropy": 5.820439052581787,
      "epoch": 3.384601940050686,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038791892658574016,
      "loss": 5.3378,
      "mean_token_accuracy": 0.1928214654326439,
      "num_tokens": 78471457.0,
      "step": 38730
    },
    {
      "entropy": 5.860285663604737,
      "epoch": 3.3850388884033906,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003878919727808111,
      "loss": 5.4108,
      "mean_token_accuracy": 0.18701445758342744,
      "num_tokens": 78480834.0,
      "step": 38735
    },
    {
      "entropy": 5.792324829101562,
      "epoch": 3.3854758367560955,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038786501681055483,
      "loss": 5.4117,
      "mean_token_accuracy": 0.19760375767946242,
      "num_tokens": 78491509.0,
      "step": 38740
    },
    {
      "entropy": 5.867590808868409,
      "epoch": 3.3859127851088,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003878380586754884,
      "loss": 5.4913,
      "mean_token_accuracy": 0.1874448150396347,
      "num_tokens": 78502137.0,
      "step": 38745
    },
    {
      "entropy": 5.789913940429687,
      "epoch": 3.386349733461505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038781109837612886,
      "loss": 5.2992,
      "mean_token_accuracy": 0.20076699256896974,
      "num_tokens": 78511888.0,
      "step": 38750
    },
    {
      "entropy": 5.796478796005249,
      "epoch": 3.3867866818142094,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003877841359129934,
      "loss": 5.4085,
      "mean_token_accuracy": 0.18366410285234452,
      "num_tokens": 78521553.0,
      "step": 38755
    },
    {
      "entropy": 5.715404987335205,
      "epoch": 3.3872236301669143,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000387757171286599,
      "loss": 5.2828,
      "mean_token_accuracy": 0.20308609008789064,
      "num_tokens": 78531290.0,
      "step": 38760
    },
    {
      "entropy": 5.7535913467407225,
      "epoch": 3.387660578519619,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003877302044974631,
      "loss": 5.3636,
      "mean_token_accuracy": 0.19817270636558532,
      "num_tokens": 78541200.0,
      "step": 38765
    },
    {
      "entropy": 5.768282890319824,
      "epoch": 3.3880975268723237,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003877032355461027,
      "loss": 5.3736,
      "mean_token_accuracy": 0.1936270236968994,
      "num_tokens": 78551075.0,
      "step": 38770
    },
    {
      "entropy": 5.819861936569214,
      "epoch": 3.388534475225028,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038767626443303525,
      "loss": 5.4035,
      "mean_token_accuracy": 0.19407716095447541,
      "num_tokens": 78560935.0,
      "step": 38775
    },
    {
      "entropy": 5.783627843856811,
      "epoch": 3.388971423577733,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000387649291158778,
      "loss": 5.3853,
      "mean_token_accuracy": 0.19349320530891417,
      "num_tokens": 78570856.0,
      "step": 38780
    },
    {
      "entropy": 5.781821393966675,
      "epoch": 3.389408371930438,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038762231572384833,
      "loss": 5.4423,
      "mean_token_accuracy": 0.1882506862282753,
      "num_tokens": 78581128.0,
      "step": 38785
    },
    {
      "entropy": 5.823237419128418,
      "epoch": 3.3898453202831424,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003875953381287636,
      "loss": 5.3764,
      "mean_token_accuracy": 0.1909894973039627,
      "num_tokens": 78591131.0,
      "step": 38790
    },
    {
      "entropy": 5.848174905776977,
      "epoch": 3.3902822686358474,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003875683583740413,
      "loss": 5.4343,
      "mean_token_accuracy": 0.2017843633890152,
      "num_tokens": 78600470.0,
      "step": 38795
    },
    {
      "entropy": 5.800705051422119,
      "epoch": 3.390719216988552,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003875413764601989,
      "loss": 5.4451,
      "mean_token_accuracy": 0.18581296801567077,
      "num_tokens": 78610916.0,
      "step": 38800
    },
    {
      "entropy": 5.7597301483154295,
      "epoch": 3.3911561653412567,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003875143923877539,
      "loss": 5.317,
      "mean_token_accuracy": 0.19729986935853958,
      "num_tokens": 78620199.0,
      "step": 38805
    },
    {
      "entropy": 5.884800720214844,
      "epoch": 3.391593113693961,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00038748740615722395,
      "loss": 5.4182,
      "mean_token_accuracy": 0.1974984496831894,
      "num_tokens": 78629186.0,
      "step": 38810
    },
    {
      "entropy": 5.791781234741211,
      "epoch": 3.392030062046666,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038746041776912654,
      "loss": 5.3594,
      "mean_token_accuracy": 0.1927062377333641,
      "num_tokens": 78638786.0,
      "step": 38815
    },
    {
      "entropy": 5.849562549591065,
      "epoch": 3.392467010399371,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00038743342722397946,
      "loss": 5.4346,
      "mean_token_accuracy": 0.18737295567989348,
      "num_tokens": 78647904.0,
      "step": 38820
    },
    {
      "entropy": 5.810832738876343,
      "epoch": 3.3929039587520755,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003874064345223003,
      "loss": 5.3892,
      "mean_token_accuracy": 0.18890935629606248,
      "num_tokens": 78657765.0,
      "step": 38825
    },
    {
      "entropy": 5.828126668930054,
      "epoch": 3.39334090710478,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00038737943966460684,
      "loss": 5.3734,
      "mean_token_accuracy": 0.19290507286787034,
      "num_tokens": 78667871.0,
      "step": 38830
    },
    {
      "entropy": 5.730621814727783,
      "epoch": 3.393777855457485,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038735244265141676,
      "loss": 5.3452,
      "mean_token_accuracy": 0.1936816766858101,
      "num_tokens": 78677824.0,
      "step": 38835
    },
    {
      "entropy": 5.842223405838013,
      "epoch": 3.39421480381019,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003873254434832481,
      "loss": 5.4452,
      "mean_token_accuracy": 0.18859403133392333,
      "num_tokens": 78687498.0,
      "step": 38840
    },
    {
      "entropy": 5.862747383117676,
      "epoch": 3.3946517521628943,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00038729844216061845,
      "loss": 5.448,
      "mean_token_accuracy": 0.19053904116153716,
      "num_tokens": 78698645.0,
      "step": 38845
    },
    {
      "entropy": 5.789618968963623,
      "epoch": 3.395088700515599,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038727143868404597,
      "loss": 5.2781,
      "mean_token_accuracy": 0.20088742822408676,
      "num_tokens": 78707796.0,
      "step": 38850
    },
    {
      "entropy": 5.744783210754394,
      "epoch": 3.3955256488683037,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003872444330540483,
      "loss": 5.3846,
      "mean_token_accuracy": 0.19455288350582123,
      "num_tokens": 78717838.0,
      "step": 38855
    },
    {
      "entropy": 5.847500610351562,
      "epoch": 3.3959625972210086,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003872174252711437,
      "loss": 5.3648,
      "mean_token_accuracy": 0.19235617369413377,
      "num_tokens": 78727318.0,
      "step": 38860
    },
    {
      "entropy": 5.860805559158325,
      "epoch": 3.396399545573713,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038719041533585,
      "loss": 5.4348,
      "mean_token_accuracy": 0.18216488361358643,
      "num_tokens": 78737446.0,
      "step": 38865
    },
    {
      "entropy": 5.828619146347046,
      "epoch": 3.396836493926418,
      "grad_norm": 1.0,
      "learning_rate": 0.0003871634032486854,
      "loss": 5.4592,
      "mean_token_accuracy": 0.18481800109148025,
      "num_tokens": 78748557.0,
      "step": 38870
    },
    {
      "entropy": 5.924154281616211,
      "epoch": 3.3972734422791224,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000387136389010168,
      "loss": 5.5275,
      "mean_token_accuracy": 0.1806876078248024,
      "num_tokens": 78759482.0,
      "step": 38875
    },
    {
      "entropy": 5.7754264831542965,
      "epoch": 3.3977103906318273,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038710937262081583,
      "loss": 5.2603,
      "mean_token_accuracy": 0.20580047070980073,
      "num_tokens": 78768959.0,
      "step": 38880
    },
    {
      "entropy": 5.722011756896973,
      "epoch": 3.398147338984532,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003870823540811472,
      "loss": 5.3494,
      "mean_token_accuracy": 0.20157893002033234,
      "num_tokens": 78779631.0,
      "step": 38885
    },
    {
      "entropy": 5.778551149368286,
      "epoch": 3.3985842873372367,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038705533339168023,
      "loss": 5.3993,
      "mean_token_accuracy": 0.19852608442306519,
      "num_tokens": 78789291.0,
      "step": 38890
    },
    {
      "entropy": 5.77886209487915,
      "epoch": 3.3990212356899416,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003870283105529333,
      "loss": 5.4238,
      "mean_token_accuracy": 0.19515773355960847,
      "num_tokens": 78799452.0,
      "step": 38895
    },
    {
      "entropy": 5.837583398818969,
      "epoch": 3.399458184042646,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038700128556542465,
      "loss": 5.3934,
      "mean_token_accuracy": 0.189585579931736,
      "num_tokens": 78809390.0,
      "step": 38900
    },
    {
      "entropy": 5.82257490158081,
      "epoch": 3.399895132395351,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003869742584296727,
      "loss": 5.3414,
      "mean_token_accuracy": 0.19707971662282944,
      "num_tokens": 78818979.0,
      "step": 38905
    },
    {
      "entropy": 5.8553791522979735,
      "epoch": 3.4003320807480555,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003869472291461957,
      "loss": 5.4801,
      "mean_token_accuracy": 0.18688323944807053,
      "num_tokens": 78829773.0,
      "step": 38910
    },
    {
      "entropy": 5.845514726638794,
      "epoch": 3.4007690291007604,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038692019771551227,
      "loss": 5.3938,
      "mean_token_accuracy": 0.18958209156990052,
      "num_tokens": 78839664.0,
      "step": 38915
    },
    {
      "entropy": 5.815493154525757,
      "epoch": 3.401205977453465,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003868931641381408,
      "loss": 5.3449,
      "mean_token_accuracy": 0.19672079533338546,
      "num_tokens": 78849941.0,
      "step": 38920
    },
    {
      "entropy": 5.852480220794678,
      "epoch": 3.40164292580617,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038686612841459973,
      "loss": 5.4148,
      "mean_token_accuracy": 0.1889297142624855,
      "num_tokens": 78860709.0,
      "step": 38925
    },
    {
      "entropy": 5.816280126571655,
      "epoch": 3.4020798741588743,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003868390905454078,
      "loss": 5.3845,
      "mean_token_accuracy": 0.18878885209560395,
      "num_tokens": 78871001.0,
      "step": 38930
    },
    {
      "entropy": 5.819601440429688,
      "epoch": 3.402516822511579,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003868120505310835,
      "loss": 5.4188,
      "mean_token_accuracy": 0.19072214066982268,
      "num_tokens": 78881033.0,
      "step": 38935
    },
    {
      "entropy": 5.76335391998291,
      "epoch": 3.4029537708642836,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038678500837214543,
      "loss": 5.3468,
      "mean_token_accuracy": 0.2020101621747017,
      "num_tokens": 78890078.0,
      "step": 38940
    },
    {
      "entropy": 5.827701282501221,
      "epoch": 3.4033907192169885,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00038675796406911235,
      "loss": 5.3856,
      "mean_token_accuracy": 0.19114421010017396,
      "num_tokens": 78900209.0,
      "step": 38945
    },
    {
      "entropy": 5.860567045211792,
      "epoch": 3.4038276675696935,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038673091762250295,
      "loss": 5.4183,
      "mean_token_accuracy": 0.19118718504905702,
      "num_tokens": 78911431.0,
      "step": 38950
    },
    {
      "entropy": 5.859270334243774,
      "epoch": 3.404264615922398,
      "grad_norm": 1.140625,
      "learning_rate": 0.000386703869032836,
      "loss": 5.4402,
      "mean_token_accuracy": 0.19270741641521455,
      "num_tokens": 78922516.0,
      "step": 38955
    },
    {
      "entropy": 5.750797462463379,
      "epoch": 3.404701564275103,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0003866768183006303,
      "loss": 5.2745,
      "mean_token_accuracy": 0.19656687080860138,
      "num_tokens": 78933186.0,
      "step": 38960
    },
    {
      "entropy": 5.717288541793823,
      "epoch": 3.4051385126278073,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003866497654264048,
      "loss": 5.2577,
      "mean_token_accuracy": 0.2058498352766037,
      "num_tokens": 78944602.0,
      "step": 38965
    },
    {
      "entropy": 5.749498701095581,
      "epoch": 3.4055754609805122,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003866227104106782,
      "loss": 5.3164,
      "mean_token_accuracy": 0.2037012755870819,
      "num_tokens": 78954336.0,
      "step": 38970
    },
    {
      "entropy": 5.794346189498901,
      "epoch": 3.4060124093332167,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003865956532539695,
      "loss": 5.3846,
      "mean_token_accuracy": 0.19500399231910706,
      "num_tokens": 78965306.0,
      "step": 38975
    },
    {
      "entropy": 5.792839193344117,
      "epoch": 3.4064493576859216,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003865685939567977,
      "loss": 5.3906,
      "mean_token_accuracy": 0.1906792163848877,
      "num_tokens": 78975528.0,
      "step": 38980
    },
    {
      "entropy": 5.894146585464478,
      "epoch": 3.406886306038626,
      "grad_norm": 1.0,
      "learning_rate": 0.0003865415325196819,
      "loss": 5.4781,
      "mean_token_accuracy": 0.18190753012895583,
      "num_tokens": 78985402.0,
      "step": 38985
    },
    {
      "entropy": 5.809747362136841,
      "epoch": 3.407323254391331,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038651446894314107,
      "loss": 5.3935,
      "mean_token_accuracy": 0.19180953800678252,
      "num_tokens": 78996053.0,
      "step": 38990
    },
    {
      "entropy": 5.876932382583618,
      "epoch": 3.4077602027440355,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003864874032276942,
      "loss": 5.4428,
      "mean_token_accuracy": 0.18614659160375596,
      "num_tokens": 79007645.0,
      "step": 38995
    },
    {
      "entropy": 5.835749530792237,
      "epoch": 3.4081971510967404,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003864603353738605,
      "loss": 5.3288,
      "mean_token_accuracy": 0.19901180863380433,
      "num_tokens": 79018649.0,
      "step": 39000
    },
    {
      "epoch": 3.4081971510967404,
      "eval_entropy": 5.567773506660453,
      "eval_loss": 5.45458984375,
      "eval_mean_token_accuracy": 0.19806989202340686,
      "eval_num_tokens": 79018649.0,
      "eval_runtime": 28.2531,
      "eval_samples_per_second": 1310.085,
      "eval_steps_per_second": 163.769,
      "step": 39000
    },
    {
      "entropy": 5.715877962112427,
      "epoch": 3.4086340994494453,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003864332653821592,
      "loss": 5.296,
      "mean_token_accuracy": 0.19730603843927383,
      "num_tokens": 79028549.0,
      "step": 39005
    },
    {
      "entropy": 5.764383172988891,
      "epoch": 3.4090710478021498,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003864061932531095,
      "loss": 5.3568,
      "mean_token_accuracy": 0.19028745591640472,
      "num_tokens": 79037788.0,
      "step": 39010
    },
    {
      "entropy": 5.845075988769532,
      "epoch": 3.4095079961548547,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038637911898723067,
      "loss": 5.4273,
      "mean_token_accuracy": 0.1872292473912239,
      "num_tokens": 79048205.0,
      "step": 39015
    },
    {
      "entropy": 5.753201723098755,
      "epoch": 3.409944944507559,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003863520425850419,
      "loss": 5.3046,
      "mean_token_accuracy": 0.19760358780622483,
      "num_tokens": 79057594.0,
      "step": 39020
    },
    {
      "entropy": 5.765127325057984,
      "epoch": 3.410381892860264,
      "grad_norm": 1.125,
      "learning_rate": 0.0003863249640470627,
      "loss": 5.3996,
      "mean_token_accuracy": 0.1944640576839447,
      "num_tokens": 79066815.0,
      "step": 39025
    },
    {
      "entropy": 5.837144756317139,
      "epoch": 3.4108188412129685,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038629788337381227,
      "loss": 5.3369,
      "mean_token_accuracy": 0.19760765582323075,
      "num_tokens": 79077191.0,
      "step": 39030
    },
    {
      "entropy": 5.7807981967926025,
      "epoch": 3.4112557895656734,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003862708005658102,
      "loss": 5.3826,
      "mean_token_accuracy": 0.19483171701431273,
      "num_tokens": 79086958.0,
      "step": 39035
    },
    {
      "entropy": 5.799136972427368,
      "epoch": 3.411692737918378,
      "grad_norm": 1.0,
      "learning_rate": 0.00038624371562357573,
      "loss": 5.4123,
      "mean_token_accuracy": 0.19084291607141496,
      "num_tokens": 79098298.0,
      "step": 39040
    },
    {
      "entropy": 5.784835290908814,
      "epoch": 3.412129686271083,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038621662854762865,
      "loss": 5.3574,
      "mean_token_accuracy": 0.19245072156190873,
      "num_tokens": 79107954.0,
      "step": 39045
    },
    {
      "entropy": 5.788955211639404,
      "epoch": 3.4125666346237873,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003861895393384882,
      "loss": 5.3787,
      "mean_token_accuracy": 0.19187183529138566,
      "num_tokens": 79118062.0,
      "step": 39050
    },
    {
      "entropy": 5.854259014129639,
      "epoch": 3.413003582976492,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038616244799667427,
      "loss": 5.4272,
      "mean_token_accuracy": 0.18842597901821137,
      "num_tokens": 79128358.0,
      "step": 39055
    },
    {
      "entropy": 5.842365026473999,
      "epoch": 3.413440531329197,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003861353545227063,
      "loss": 5.4151,
      "mean_token_accuracy": 0.19431231170892715,
      "num_tokens": 79139882.0,
      "step": 39060
    },
    {
      "entropy": 5.84859299659729,
      "epoch": 3.4138774796819016,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003861082589171039,
      "loss": 5.4128,
      "mean_token_accuracy": 0.19006162285804748,
      "num_tokens": 79150748.0,
      "step": 39065
    },
    {
      "entropy": 5.764991521835327,
      "epoch": 3.4143144280346065,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038608116118038705,
      "loss": 5.3624,
      "mean_token_accuracy": 0.19543280601501464,
      "num_tokens": 79160088.0,
      "step": 39070
    },
    {
      "entropy": 5.731704139709473,
      "epoch": 3.414751376387311,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003860540613130752,
      "loss": 5.296,
      "mean_token_accuracy": 0.20659617334604263,
      "num_tokens": 79169539.0,
      "step": 39075
    },
    {
      "entropy": 5.866506958007813,
      "epoch": 3.415188324740016,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038602695931568834,
      "loss": 5.4469,
      "mean_token_accuracy": 0.1925524204969406,
      "num_tokens": 79179505.0,
      "step": 39080
    },
    {
      "entropy": 5.822225141525268,
      "epoch": 3.4156252730927203,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00038599985518874614,
      "loss": 5.3421,
      "mean_token_accuracy": 0.19346646666526796,
      "num_tokens": 79189791.0,
      "step": 39085
    },
    {
      "entropy": 5.8661902904510494,
      "epoch": 3.4160622214454253,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038597274893276855,
      "loss": 5.3865,
      "mean_token_accuracy": 0.19231331646442412,
      "num_tokens": 79199708.0,
      "step": 39090
    },
    {
      "entropy": 5.784337043762207,
      "epoch": 3.4164991697981297,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003859456405482755,
      "loss": 5.3953,
      "mean_token_accuracy": 0.19334367662668228,
      "num_tokens": 79209624.0,
      "step": 39095
    },
    {
      "entropy": 5.780905294418335,
      "epoch": 3.4169361181508346,
      "grad_norm": 1.03125,
      "learning_rate": 0.000385918530035787,
      "loss": 5.3478,
      "mean_token_accuracy": 0.19391023367643356,
      "num_tokens": 79218873.0,
      "step": 39100
    },
    {
      "entropy": 5.813691568374634,
      "epoch": 3.417373066503539,
      "grad_norm": 0.9375,
      "learning_rate": 0.00038589141739582295,
      "loss": 5.3859,
      "mean_token_accuracy": 0.19196268022060395,
      "num_tokens": 79228838.0,
      "step": 39105
    },
    {
      "entropy": 5.819198846817017,
      "epoch": 3.417810014856244,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003858643026289034,
      "loss": 5.353,
      "mean_token_accuracy": 0.19535655677318572,
      "num_tokens": 79238185.0,
      "step": 39110
    },
    {
      "entropy": 5.807522821426391,
      "epoch": 3.418246963208949,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038583718573554845,
      "loss": 5.3863,
      "mean_token_accuracy": 0.18896746933460234,
      "num_tokens": 79248777.0,
      "step": 39115
    },
    {
      "entropy": 5.749645376205445,
      "epoch": 3.4186839115616534,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003858100667162781,
      "loss": 5.2826,
      "mean_token_accuracy": 0.20325341522693635,
      "num_tokens": 79258471.0,
      "step": 39120
    },
    {
      "entropy": 5.708644056320191,
      "epoch": 3.419120859914358,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003857829455716127,
      "loss": 5.3,
      "mean_token_accuracy": 0.1985971137881279,
      "num_tokens": 79268483.0,
      "step": 39125
    },
    {
      "entropy": 5.774135112762451,
      "epoch": 3.419557808267063,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003857558223020724,
      "loss": 5.3668,
      "mean_token_accuracy": 0.1959307610988617,
      "num_tokens": 79278497.0,
      "step": 39130
    },
    {
      "entropy": 5.805976009368896,
      "epoch": 3.4199947566197677,
      "grad_norm": 1.0,
      "learning_rate": 0.0003857286969081773,
      "loss": 5.3311,
      "mean_token_accuracy": 0.20178561955690383,
      "num_tokens": 79288010.0,
      "step": 39135
    },
    {
      "entropy": 5.777049207687378,
      "epoch": 3.420431704972472,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038570156939044784,
      "loss": 5.4073,
      "mean_token_accuracy": 0.1928394541144371,
      "num_tokens": 79298092.0,
      "step": 39140
    },
    {
      "entropy": 5.8094323635101315,
      "epoch": 3.420868653325177,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003856744397494042,
      "loss": 5.4346,
      "mean_token_accuracy": 0.18820183724164963,
      "num_tokens": 79308458.0,
      "step": 39145
    },
    {
      "entropy": 5.863060569763183,
      "epoch": 3.4213056016778816,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003856473079855668,
      "loss": 5.4601,
      "mean_token_accuracy": 0.18298359960317612,
      "num_tokens": 79318740.0,
      "step": 39150
    },
    {
      "entropy": 5.819454860687256,
      "epoch": 3.4217425500305865,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038562017409945614,
      "loss": 5.4262,
      "mean_token_accuracy": 0.20075150728225707,
      "num_tokens": 79329738.0,
      "step": 39155
    },
    {
      "entropy": 5.82776780128479,
      "epoch": 3.422179498383291,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038559303809159254,
      "loss": 5.4082,
      "mean_token_accuracy": 0.19641637653112412,
      "num_tokens": 79341137.0,
      "step": 39160
    },
    {
      "entropy": 5.830035161972046,
      "epoch": 3.422616446735996,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003855658999624965,
      "loss": 5.3966,
      "mean_token_accuracy": 0.19620237052440642,
      "num_tokens": 79351601.0,
      "step": 39165
    },
    {
      "entropy": 5.83033013343811,
      "epoch": 3.4230533950887003,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003855387597126886,
      "loss": 5.3797,
      "mean_token_accuracy": 0.1908641502261162,
      "num_tokens": 79360875.0,
      "step": 39170
    },
    {
      "entropy": 5.837616205215454,
      "epoch": 3.4234903434414052,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038551161734268934,
      "loss": 5.4149,
      "mean_token_accuracy": 0.19204048067331314,
      "num_tokens": 79371033.0,
      "step": 39175
    },
    {
      "entropy": 5.837235355377198,
      "epoch": 3.4239272917941097,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038548447285301934,
      "loss": 5.4267,
      "mean_token_accuracy": 0.19642667323350907,
      "num_tokens": 79381237.0,
      "step": 39180
    },
    {
      "entropy": 5.826984643936157,
      "epoch": 3.4243642401468146,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003854573262441993,
      "loss": 5.4387,
      "mean_token_accuracy": 0.18681686371564865,
      "num_tokens": 79392279.0,
      "step": 39185
    },
    {
      "entropy": 5.812698650360107,
      "epoch": 3.4248011884995195,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038543017751674986,
      "loss": 5.3894,
      "mean_token_accuracy": 0.19255284667015077,
      "num_tokens": 79401723.0,
      "step": 39190
    },
    {
      "entropy": 5.733371067047119,
      "epoch": 3.425238136852224,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003854030266711917,
      "loss": 5.3114,
      "mean_token_accuracy": 0.20598590821027757,
      "num_tokens": 79413436.0,
      "step": 39195
    },
    {
      "entropy": 5.804712963104248,
      "epoch": 3.425675085204929,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00038537587370804567,
      "loss": 5.4014,
      "mean_token_accuracy": 0.19691072702407836,
      "num_tokens": 79423923.0,
      "step": 39200
    },
    {
      "entropy": 5.828931903839111,
      "epoch": 3.4261120335576334,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003853487186278326,
      "loss": 5.3772,
      "mean_token_accuracy": 0.18844411820173262,
      "num_tokens": 79433992.0,
      "step": 39205
    },
    {
      "entropy": 5.833505439758301,
      "epoch": 3.4265489819103383,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038532156143107323,
      "loss": 5.4275,
      "mean_token_accuracy": 0.19290297329425812,
      "num_tokens": 79444564.0,
      "step": 39210
    },
    {
      "entropy": 5.849134683609009,
      "epoch": 3.4269859302630428,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003852944021182885,
      "loss": 5.3905,
      "mean_token_accuracy": 0.1954462543129921,
      "num_tokens": 79454985.0,
      "step": 39215
    },
    {
      "entropy": 5.812239456176758,
      "epoch": 3.4274228786157477,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003852672406899993,
      "loss": 5.4052,
      "mean_token_accuracy": 0.18819452226161956,
      "num_tokens": 79464959.0,
      "step": 39220
    },
    {
      "entropy": 5.802862024307251,
      "epoch": 3.427859826968452,
      "grad_norm": 0.96875,
      "learning_rate": 0.00038524007714672674,
      "loss": 5.3736,
      "mean_token_accuracy": 0.18733367323875427,
      "num_tokens": 79474679.0,
      "step": 39225
    },
    {
      "entropy": 5.819068861007691,
      "epoch": 3.428296775321157,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003852129114889916,
      "loss": 5.4116,
      "mean_token_accuracy": 0.19166318327188492,
      "num_tokens": 79485908.0,
      "step": 39230
    },
    {
      "entropy": 5.797839689254761,
      "epoch": 3.4287337236738615,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003851857437173152,
      "loss": 5.4149,
      "mean_token_accuracy": 0.19097733348608018,
      "num_tokens": 79496209.0,
      "step": 39235
    },
    {
      "entropy": 5.869043064117432,
      "epoch": 3.4291706720265664,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003851585738322184,
      "loss": 5.361,
      "mean_token_accuracy": 0.19883085042238235,
      "num_tokens": 79507351.0,
      "step": 39240
    },
    {
      "entropy": 5.724036598205567,
      "epoch": 3.4296076203792714,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038513140183422256,
      "loss": 5.3576,
      "mean_token_accuracy": 0.19477999955415726,
      "num_tokens": 79517517.0,
      "step": 39245
    },
    {
      "entropy": 5.699533700942993,
      "epoch": 3.430044568731976,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003851042277238486,
      "loss": 5.3301,
      "mean_token_accuracy": 0.19939940571784973,
      "num_tokens": 79528616.0,
      "step": 39250
    },
    {
      "entropy": 5.820388126373291,
      "epoch": 3.4304815170846807,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003850770515016179,
      "loss": 5.3775,
      "mean_token_accuracy": 0.19204545617103577,
      "num_tokens": 79538424.0,
      "step": 39255
    },
    {
      "entropy": 5.882828187942505,
      "epoch": 3.430918465437385,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0003850498731680517,
      "loss": 5.4422,
      "mean_token_accuracy": 0.18377741873264314,
      "num_tokens": 79548717.0,
      "step": 39260
    },
    {
      "entropy": 5.8221300601959225,
      "epoch": 3.43135541379009,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038502269272367123,
      "loss": 5.3049,
      "mean_token_accuracy": 0.20287707895040513,
      "num_tokens": 79558998.0,
      "step": 39265
    },
    {
      "entropy": 5.833305835723877,
      "epoch": 3.4317923621427946,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038499551016899793,
      "loss": 5.4585,
      "mean_token_accuracy": 0.18436232060194016,
      "num_tokens": 79569424.0,
      "step": 39270
    },
    {
      "entropy": 5.8135158061981205,
      "epoch": 3.4322293104954995,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038496832550455307,
      "loss": 5.375,
      "mean_token_accuracy": 0.19413819015026093,
      "num_tokens": 79578670.0,
      "step": 39275
    },
    {
      "entropy": 5.829845428466797,
      "epoch": 3.432666258848204,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038494113873085805,
      "loss": 5.4595,
      "mean_token_accuracy": 0.1893165737390518,
      "num_tokens": 79588883.0,
      "step": 39280
    },
    {
      "entropy": 5.79682674407959,
      "epoch": 3.433103207200909,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038491394984843446,
      "loss": 5.3429,
      "mean_token_accuracy": 0.19774559885263443,
      "num_tokens": 79598158.0,
      "step": 39285
    },
    {
      "entropy": 5.762523984909057,
      "epoch": 3.4335401555536134,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038488675885780366,
      "loss": 5.3302,
      "mean_token_accuracy": 0.19862671345472335,
      "num_tokens": 79608699.0,
      "step": 39290
    },
    {
      "entropy": 5.838223695755005,
      "epoch": 3.4339771039063183,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038485956575948726,
      "loss": 5.3414,
      "mean_token_accuracy": 0.20125005543231964,
      "num_tokens": 79618628.0,
      "step": 39295
    },
    {
      "entropy": 5.831130027770996,
      "epoch": 3.434414052259023,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003848323705540069,
      "loss": 5.3722,
      "mean_token_accuracy": 0.19023431241512298,
      "num_tokens": 79628836.0,
      "step": 39300
    },
    {
      "entropy": 5.792210912704467,
      "epoch": 3.4348510006117277,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000384805173241884,
      "loss": 5.3318,
      "mean_token_accuracy": 0.19373501539230348,
      "num_tokens": 79638306.0,
      "step": 39305
    },
    {
      "entropy": 5.7843516826629635,
      "epoch": 3.4352879489644326,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038477797382364037,
      "loss": 5.4278,
      "mean_token_accuracy": 0.1921383947134018,
      "num_tokens": 79647955.0,
      "step": 39310
    },
    {
      "entropy": 5.872236919403076,
      "epoch": 3.435724897317137,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038475077229979774,
      "loss": 5.4023,
      "mean_token_accuracy": 0.19115757942199707,
      "num_tokens": 79658701.0,
      "step": 39315
    },
    {
      "entropy": 5.816879463195801,
      "epoch": 3.436161845669842,
      "grad_norm": 1.375,
      "learning_rate": 0.00038472356867087776,
      "loss": 5.4275,
      "mean_token_accuracy": 0.18858869671821593,
      "num_tokens": 79669436.0,
      "step": 39320
    },
    {
      "entropy": 5.806035757064819,
      "epoch": 3.4365987940225464,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003846963629374022,
      "loss": 5.4144,
      "mean_token_accuracy": 0.18756455928087234,
      "num_tokens": 79679350.0,
      "step": 39325
    },
    {
      "entropy": 5.822504091262817,
      "epoch": 3.4370357423752513,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038466915509989297,
      "loss": 5.323,
      "mean_token_accuracy": 0.19894784837961196,
      "num_tokens": 79689762.0,
      "step": 39330
    },
    {
      "entropy": 5.815987014770508,
      "epoch": 3.437472690727956,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038464194515887187,
      "loss": 5.3197,
      "mean_token_accuracy": 0.19714833945035934,
      "num_tokens": 79699594.0,
      "step": 39335
    },
    {
      "entropy": 5.771500444412231,
      "epoch": 3.4379096390806607,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003846147331148608,
      "loss": 5.4364,
      "mean_token_accuracy": 0.19502972811460495,
      "num_tokens": 79710403.0,
      "step": 39340
    },
    {
      "entropy": 5.690166091918945,
      "epoch": 3.438346587433365,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00038458751896838176,
      "loss": 5.2384,
      "mean_token_accuracy": 0.20763132423162461,
      "num_tokens": 79721291.0,
      "step": 39345
    },
    {
      "entropy": 5.823867464065552,
      "epoch": 3.43878353578607,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003845603027199566,
      "loss": 5.4241,
      "mean_token_accuracy": 0.18580671399831772,
      "num_tokens": 79732505.0,
      "step": 39350
    },
    {
      "entropy": 5.896340942382812,
      "epoch": 3.439220484138775,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038453308437010747,
      "loss": 5.4507,
      "mean_token_accuracy": 0.191101336479187,
      "num_tokens": 79743707.0,
      "step": 39355
    },
    {
      "entropy": 5.910034465789795,
      "epoch": 3.4396574324914795,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038450586391935635,
      "loss": 5.4077,
      "mean_token_accuracy": 0.19088027626276016,
      "num_tokens": 79754669.0,
      "step": 39360
    },
    {
      "entropy": 5.815587425231934,
      "epoch": 3.4400943808441844,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003844786413682254,
      "loss": 5.3831,
      "mean_token_accuracy": 0.19930624067783356,
      "num_tokens": 79764444.0,
      "step": 39365
    },
    {
      "entropy": 5.8064628601074215,
      "epoch": 3.440531329196889,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003844514167172368,
      "loss": 5.4311,
      "mean_token_accuracy": 0.1807194322347641,
      "num_tokens": 79774942.0,
      "step": 39370
    },
    {
      "entropy": 5.8583191394805905,
      "epoch": 3.440968277549594,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003844241899669126,
      "loss": 5.4107,
      "mean_token_accuracy": 0.18228285908699035,
      "num_tokens": 79785710.0,
      "step": 39375
    },
    {
      "entropy": 5.889687204360962,
      "epoch": 3.4414052259022982,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003843969611177751,
      "loss": 5.4556,
      "mean_token_accuracy": 0.18526993840932846,
      "num_tokens": 79795841.0,
      "step": 39380
    },
    {
      "entropy": 5.771808290481568,
      "epoch": 3.441842174255003,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003843697301703466,
      "loss": 5.3187,
      "mean_token_accuracy": 0.19362842440605163,
      "num_tokens": 79805400.0,
      "step": 39385
    },
    {
      "entropy": 5.762027025222778,
      "epoch": 3.4422791226077076,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003843424971251494,
      "loss": 5.4067,
      "mean_token_accuracy": 0.1903933420777321,
      "num_tokens": 79816212.0,
      "step": 39390
    },
    {
      "entropy": 5.790729522705078,
      "epoch": 3.4427160709604125,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003843152619827058,
      "loss": 5.3638,
      "mean_token_accuracy": 0.1917254313826561,
      "num_tokens": 79827125.0,
      "step": 39395
    },
    {
      "entropy": 5.891231060028076,
      "epoch": 3.443153019313117,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038428802474353816,
      "loss": 5.4129,
      "mean_token_accuracy": 0.1916695773601532,
      "num_tokens": 79837436.0,
      "step": 39400
    },
    {
      "entropy": 5.889025115966797,
      "epoch": 3.443589967665822,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003842607854081688,
      "loss": 5.4368,
      "mean_token_accuracy": 0.19400036484003066,
      "num_tokens": 79847297.0,
      "step": 39405
    },
    {
      "entropy": 5.783824729919433,
      "epoch": 3.444026916018527,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003842335439771205,
      "loss": 5.3044,
      "mean_token_accuracy": 0.19578584283590317,
      "num_tokens": 79856321.0,
      "step": 39410
    },
    {
      "entropy": 5.769756031036377,
      "epoch": 3.4444638643712313,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038420630045091556,
      "loss": 5.3279,
      "mean_token_accuracy": 0.1973993092775345,
      "num_tokens": 79866244.0,
      "step": 39415
    },
    {
      "entropy": 5.86252727508545,
      "epoch": 3.4449008127239362,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003841790548300765,
      "loss": 5.4536,
      "mean_token_accuracy": 0.18452873826026917,
      "num_tokens": 79876900.0,
      "step": 39420
    },
    {
      "entropy": 5.884059810638428,
      "epoch": 3.4453377610766407,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038415180711512604,
      "loss": 5.4141,
      "mean_token_accuracy": 0.19130707681179046,
      "num_tokens": 79887864.0,
      "step": 39425
    },
    {
      "entropy": 5.830131959915161,
      "epoch": 3.4457747094293456,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003841245573065867,
      "loss": 5.3667,
      "mean_token_accuracy": 0.19047940373420716,
      "num_tokens": 79897625.0,
      "step": 39430
    },
    {
      "entropy": 5.805777931213379,
      "epoch": 3.44621165778205,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003840973054049812,
      "loss": 5.405,
      "mean_token_accuracy": 0.19354272037744522,
      "num_tokens": 79907742.0,
      "step": 39435
    },
    {
      "entropy": 5.787826108932495,
      "epoch": 3.446648606134755,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038407005141083216,
      "loss": 5.3473,
      "mean_token_accuracy": 0.1958503782749176,
      "num_tokens": 79917974.0,
      "step": 39440
    },
    {
      "entropy": 5.7897437572479244,
      "epoch": 3.4470855544874595,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003840427953246624,
      "loss": 5.3632,
      "mean_token_accuracy": 0.19613091349601747,
      "num_tokens": 79928089.0,
      "step": 39445
    },
    {
      "entropy": 5.709189319610596,
      "epoch": 3.4475225028401644,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00038401553714699466,
      "loss": 5.3084,
      "mean_token_accuracy": 0.202835051715374,
      "num_tokens": 79938396.0,
      "step": 39450
    },
    {
      "entropy": 5.863973617553711,
      "epoch": 3.447959451192869,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003839882768783518,
      "loss": 5.3823,
      "mean_token_accuracy": 0.19097906798124314,
      "num_tokens": 79948157.0,
      "step": 39455
    },
    {
      "entropy": 5.879318809509277,
      "epoch": 3.4483963995455738,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003839610145192567,
      "loss": 5.4063,
      "mean_token_accuracy": 0.18683266043663024,
      "num_tokens": 79958653.0,
      "step": 39460
    },
    {
      "entropy": 5.718146276473999,
      "epoch": 3.4488333478982787,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003839337500702323,
      "loss": 5.388,
      "mean_token_accuracy": 0.18874556720256805,
      "num_tokens": 79969180.0,
      "step": 39465
    },
    {
      "entropy": 5.809643411636353,
      "epoch": 3.449270296250983,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038390648353180145,
      "loss": 5.4481,
      "mean_token_accuracy": 0.19233476519584655,
      "num_tokens": 79978467.0,
      "step": 39470
    },
    {
      "entropy": 5.826445627212524,
      "epoch": 3.4497072446036876,
      "grad_norm": 1.0,
      "learning_rate": 0.0003838792149044871,
      "loss": 5.3521,
      "mean_token_accuracy": 0.1933510944247246,
      "num_tokens": 79989071.0,
      "step": 39475
    },
    {
      "entropy": 5.784649085998535,
      "epoch": 3.4501441929563925,
      "grad_norm": 1.125,
      "learning_rate": 0.00038385194418881247,
      "loss": 5.3619,
      "mean_token_accuracy": 0.1939457654953003,
      "num_tokens": 79999476.0,
      "step": 39480
    },
    {
      "entropy": 5.731386137008667,
      "epoch": 3.4505811413090974,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038382467138530055,
      "loss": 5.2943,
      "mean_token_accuracy": 0.20461076200008393,
      "num_tokens": 80009986.0,
      "step": 39485
    },
    {
      "entropy": 5.802434396743775,
      "epoch": 3.451018089661802,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038379739649447425,
      "loss": 5.4738,
      "mean_token_accuracy": 0.18844135105609894,
      "num_tokens": 80020916.0,
      "step": 39490
    },
    {
      "entropy": 5.790802955627441,
      "epoch": 3.451455038014507,
      "grad_norm": 1.015625,
      "learning_rate": 0.000383770119516857,
      "loss": 5.3271,
      "mean_token_accuracy": 0.1981397271156311,
      "num_tokens": 80030769.0,
      "step": 39495
    },
    {
      "entropy": 5.877876091003418,
      "epoch": 3.4518919863672113,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003837428404529718,
      "loss": 5.3761,
      "mean_token_accuracy": 0.19260393381118773,
      "num_tokens": 80040838.0,
      "step": 39500
    },
    {
      "entropy": 5.845251655578613,
      "epoch": 3.452328934719916,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00038371555930334194,
      "loss": 5.3795,
      "mean_token_accuracy": 0.19285198748111726,
      "num_tokens": 80050600.0,
      "step": 39505
    },
    {
      "entropy": 5.778702116012573,
      "epoch": 3.4527658830726207,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003836882760684907,
      "loss": 5.3785,
      "mean_token_accuracy": 0.19369707852602006,
      "num_tokens": 80060826.0,
      "step": 39510
    },
    {
      "entropy": 5.872387456893921,
      "epoch": 3.4532028314253256,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003836609907489414,
      "loss": 5.4374,
      "mean_token_accuracy": 0.19023407846689225,
      "num_tokens": 80070213.0,
      "step": 39515
    },
    {
      "entropy": 5.808171796798706,
      "epoch": 3.45363977977803,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003836337033452173,
      "loss": 5.3119,
      "mean_token_accuracy": 0.1947313129901886,
      "num_tokens": 80080197.0,
      "step": 39520
    },
    {
      "entropy": 5.876462125778199,
      "epoch": 3.454076728130735,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003836064138578419,
      "loss": 5.4121,
      "mean_token_accuracy": 0.19072862416505815,
      "num_tokens": 80089504.0,
      "step": 39525
    },
    {
      "entropy": 5.815023136138916,
      "epoch": 3.4545136764834394,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003835791222873386,
      "loss": 5.3568,
      "mean_token_accuracy": 0.19810165464878082,
      "num_tokens": 80099621.0,
      "step": 39530
    },
    {
      "entropy": 5.778339147567749,
      "epoch": 3.4549506248361443,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038355182863423077,
      "loss": 5.3507,
      "mean_token_accuracy": 0.19575644731521608,
      "num_tokens": 80109880.0,
      "step": 39535
    },
    {
      "entropy": 5.787299299240113,
      "epoch": 3.4553875731888493,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038352453289904206,
      "loss": 5.3532,
      "mean_token_accuracy": 0.19987080246210098,
      "num_tokens": 80119236.0,
      "step": 39540
    },
    {
      "entropy": 5.754014873504639,
      "epoch": 3.4558245215415537,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038349723508229587,
      "loss": 5.3572,
      "mean_token_accuracy": 0.19451094567775726,
      "num_tokens": 80129598.0,
      "step": 39545
    },
    {
      "entropy": 5.867105960845947,
      "epoch": 3.4562614698942586,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038346993518451587,
      "loss": 5.3696,
      "mean_token_accuracy": 0.18424701690673828,
      "num_tokens": 80140050.0,
      "step": 39550
    },
    {
      "entropy": 5.805595827102661,
      "epoch": 3.456698418246963,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038344263320622574,
      "loss": 5.3341,
      "mean_token_accuracy": 0.2007899671792984,
      "num_tokens": 80150101.0,
      "step": 39555
    },
    {
      "entropy": 5.834888362884522,
      "epoch": 3.457135366599668,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038341532914794904,
      "loss": 5.3812,
      "mean_token_accuracy": 0.18501625657081605,
      "num_tokens": 80159416.0,
      "step": 39560
    },
    {
      "entropy": 5.741442728042602,
      "epoch": 3.4575723149523725,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003833880230102095,
      "loss": 5.3569,
      "mean_token_accuracy": 0.19590692669153215,
      "num_tokens": 80168979.0,
      "step": 39565
    },
    {
      "entropy": 5.775707769393921,
      "epoch": 3.4580092633050774,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038336071479353086,
      "loss": 5.3534,
      "mean_token_accuracy": 0.1979563683271408,
      "num_tokens": 80179328.0,
      "step": 39570
    },
    {
      "entropy": 5.843149662017822,
      "epoch": 3.458446211657782,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038333340449843694,
      "loss": 5.3488,
      "mean_token_accuracy": 0.19640517085790635,
      "num_tokens": 80189151.0,
      "step": 39575
    },
    {
      "entropy": 5.809050273895264,
      "epoch": 3.458883160010487,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003833060921254517,
      "loss": 5.3998,
      "mean_token_accuracy": 0.19568121880292894,
      "num_tokens": 80199622.0,
      "step": 39580
    },
    {
      "entropy": 5.841926431655883,
      "epoch": 3.4593201083631913,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038327877767509866,
      "loss": 5.3986,
      "mean_token_accuracy": 0.19060468226671218,
      "num_tokens": 80209335.0,
      "step": 39585
    },
    {
      "entropy": 5.821146440505982,
      "epoch": 3.459757056715896,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000383251461147902,
      "loss": 5.4055,
      "mean_token_accuracy": 0.1931871071457863,
      "num_tokens": 80220336.0,
      "step": 39590
    },
    {
      "entropy": 5.835360431671143,
      "epoch": 3.460194005068601,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038322414254438557,
      "loss": 5.3992,
      "mean_token_accuracy": 0.18820488303899766,
      "num_tokens": 80231459.0,
      "step": 39595
    },
    {
      "entropy": 5.83115520477295,
      "epoch": 3.4606309534213056,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003831968218650735,
      "loss": 5.4212,
      "mean_token_accuracy": 0.19331736415624617,
      "num_tokens": 80240997.0,
      "step": 39600
    },
    {
      "entropy": 5.8194653511047365,
      "epoch": 3.4610679017740105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038316949911048953,
      "loss": 5.3772,
      "mean_token_accuracy": 0.1935667648911476,
      "num_tokens": 80251663.0,
      "step": 39605
    },
    {
      "entropy": 5.813055372238159,
      "epoch": 3.461504850126715,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00038314217428115797,
      "loss": 5.4344,
      "mean_token_accuracy": 0.19031124860048293,
      "num_tokens": 80261059.0,
      "step": 39610
    },
    {
      "entropy": 5.756773233413696,
      "epoch": 3.46194179847942,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003831148473776028,
      "loss": 5.3147,
      "mean_token_accuracy": 0.20613828897476197,
      "num_tokens": 80271781.0,
      "step": 39615
    },
    {
      "entropy": 5.785203170776367,
      "epoch": 3.4623787468321243,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038308751840034826,
      "loss": 5.4169,
      "mean_token_accuracy": 0.19275887459516525,
      "num_tokens": 80281996.0,
      "step": 39620
    },
    {
      "entropy": 5.825659942626953,
      "epoch": 3.4628156951848292,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003830601873499185,
      "loss": 5.3715,
      "mean_token_accuracy": 0.19797059744596482,
      "num_tokens": 80293181.0,
      "step": 39625
    },
    {
      "entropy": 5.831800746917724,
      "epoch": 3.4632526435375337,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003830328542268377,
      "loss": 5.3964,
      "mean_token_accuracy": 0.1914390057325363,
      "num_tokens": 80303697.0,
      "step": 39630
    },
    {
      "entropy": 5.777182102203369,
      "epoch": 3.4636895918902386,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038300551903163013,
      "loss": 5.2936,
      "mean_token_accuracy": 0.2041643425822258,
      "num_tokens": 80313104.0,
      "step": 39635
    },
    {
      "entropy": 5.768961000442505,
      "epoch": 3.464126540242943,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038297818176482006,
      "loss": 5.3035,
      "mean_token_accuracy": 0.20453406423330306,
      "num_tokens": 80323168.0,
      "step": 39640
    },
    {
      "entropy": 5.790028858184814,
      "epoch": 3.464563488595648,
      "grad_norm": 1.125,
      "learning_rate": 0.000382950842426932,
      "loss": 5.3848,
      "mean_token_accuracy": 0.19274291694164275,
      "num_tokens": 80333064.0,
      "step": 39645
    },
    {
      "entropy": 5.765751457214355,
      "epoch": 3.465000436948353,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038292350101849017,
      "loss": 5.292,
      "mean_token_accuracy": 0.19989534169435502,
      "num_tokens": 80342827.0,
      "step": 39650
    },
    {
      "entropy": 5.796984720230102,
      "epoch": 3.4654373853010574,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000382896157540019,
      "loss": 5.416,
      "mean_token_accuracy": 0.19096724092960357,
      "num_tokens": 80352732.0,
      "step": 39655
    },
    {
      "entropy": 5.896733093261719,
      "epoch": 3.4658743336537623,
      "grad_norm": 1.078125,
      "learning_rate": 0.000382868811992043,
      "loss": 5.4483,
      "mean_token_accuracy": 0.18618662655353546,
      "num_tokens": 80362427.0,
      "step": 39660
    },
    {
      "entropy": 5.90294041633606,
      "epoch": 3.4663112820064668,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003828414643750867,
      "loss": 5.3485,
      "mean_token_accuracy": 0.1959345057606697,
      "num_tokens": 80372719.0,
      "step": 39665
    },
    {
      "entropy": 5.775144433975219,
      "epoch": 3.4667482303591717,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038281411468967457,
      "loss": 5.3395,
      "mean_token_accuracy": 0.19291892796754836,
      "num_tokens": 80381753.0,
      "step": 39670
    },
    {
      "entropy": 5.7116780281066895,
      "epoch": 3.467185178711876,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003827867629363312,
      "loss": 5.3871,
      "mean_token_accuracy": 0.19839372485876083,
      "num_tokens": 80393401.0,
      "step": 39675
    },
    {
      "entropy": 5.872656202316284,
      "epoch": 3.467622127064581,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038275940911558135,
      "loss": 5.411,
      "mean_token_accuracy": 0.1885163053870201,
      "num_tokens": 80403194.0,
      "step": 39680
    },
    {
      "entropy": 5.854885482788086,
      "epoch": 3.4680590754172855,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003827320532279495,
      "loss": 5.3672,
      "mean_token_accuracy": 0.1897878959774971,
      "num_tokens": 80413283.0,
      "step": 39685
    },
    {
      "entropy": 5.83965277671814,
      "epoch": 3.4684960237699904,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038270469527396035,
      "loss": 5.3905,
      "mean_token_accuracy": 0.19092378318309783,
      "num_tokens": 80423976.0,
      "step": 39690
    },
    {
      "entropy": 5.762296390533447,
      "epoch": 3.468932972122695,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038267733525413884,
      "loss": 5.3612,
      "mean_token_accuracy": 0.20068747997283937,
      "num_tokens": 80434017.0,
      "step": 39695
    },
    {
      "entropy": 5.802002239227295,
      "epoch": 3.4693699204754,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003826499731690095,
      "loss": 5.3475,
      "mean_token_accuracy": 0.19578684270381927,
      "num_tokens": 80444289.0,
      "step": 39700
    },
    {
      "entropy": 5.862638139724732,
      "epoch": 3.4698068688281047,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003826226090190974,
      "loss": 5.3976,
      "mean_token_accuracy": 0.19347311109304427,
      "num_tokens": 80453420.0,
      "step": 39705
    },
    {
      "entropy": 5.873456621170044,
      "epoch": 3.470243817180809,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038259524280492716,
      "loss": 5.4028,
      "mean_token_accuracy": 0.18869862854480743,
      "num_tokens": 80462977.0,
      "step": 39710
    },
    {
      "entropy": 5.7429815292358395,
      "epoch": 3.470680765533514,
      "grad_norm": 1.0,
      "learning_rate": 0.00038256787452702377,
      "loss": 5.2644,
      "mean_token_accuracy": 0.2008230596780777,
      "num_tokens": 80471806.0,
      "step": 39715
    },
    {
      "entropy": 5.732673788070679,
      "epoch": 3.4711177138862186,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038254050418591227,
      "loss": 5.3446,
      "mean_token_accuracy": 0.19350268095731735,
      "num_tokens": 80481510.0,
      "step": 39720
    },
    {
      "entropy": 5.8461644649505615,
      "epoch": 3.4715546622389235,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003825131317821175,
      "loss": 5.3767,
      "mean_token_accuracy": 0.18849007338285445,
      "num_tokens": 80491670.0,
      "step": 39725
    },
    {
      "entropy": 5.888040256500244,
      "epoch": 3.471991610591628,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038248575731616454,
      "loss": 5.4693,
      "mean_token_accuracy": 0.18876772671937941,
      "num_tokens": 80501518.0,
      "step": 39730
    },
    {
      "entropy": 5.7845557689666744,
      "epoch": 3.472428558944333,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00038245838078857837,
      "loss": 5.3641,
      "mean_token_accuracy": 0.19422477930784227,
      "num_tokens": 80511943.0,
      "step": 39735
    },
    {
      "entropy": 5.759305238723755,
      "epoch": 3.4728655072970374,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003824310021998841,
      "loss": 5.3305,
      "mean_token_accuracy": 0.1928809851408005,
      "num_tokens": 80523114.0,
      "step": 39740
    },
    {
      "entropy": 5.8880716323852536,
      "epoch": 3.4733024556497423,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003824036215506071,
      "loss": 5.4447,
      "mean_token_accuracy": 0.1838892564177513,
      "num_tokens": 80534802.0,
      "step": 39745
    },
    {
      "entropy": 5.862362241744995,
      "epoch": 3.4737394040024467,
      "grad_norm": 0.921875,
      "learning_rate": 0.00038237623884127225,
      "loss": 5.3652,
      "mean_token_accuracy": 0.1929283320903778,
      "num_tokens": 80544367.0,
      "step": 39750
    },
    {
      "entropy": 5.784653091430664,
      "epoch": 3.4741763523551517,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038234885407240494,
      "loss": 5.4345,
      "mean_token_accuracy": 0.19255390614271164,
      "num_tokens": 80554841.0,
      "step": 39755
    },
    {
      "entropy": 5.856284809112549,
      "epoch": 3.4746133007078566,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003823214672445302,
      "loss": 5.4135,
      "mean_token_accuracy": 0.18550460040569305,
      "num_tokens": 80565076.0,
      "step": 39760
    },
    {
      "entropy": 5.836421918869019,
      "epoch": 3.475050249060561,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038229407835817366,
      "loss": 5.4471,
      "mean_token_accuracy": 0.1899029478430748,
      "num_tokens": 80575410.0,
      "step": 39765
    },
    {
      "entropy": 5.753887557983399,
      "epoch": 3.4754871974132655,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038226668741386046,
      "loss": 5.3742,
      "mean_token_accuracy": 0.19443319290876387,
      "num_tokens": 80585400.0,
      "step": 39770
    },
    {
      "entropy": 5.87325530052185,
      "epoch": 3.4759241457659704,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003822392944121159,
      "loss": 5.3653,
      "mean_token_accuracy": 0.19200910329818727,
      "num_tokens": 80595429.0,
      "step": 39775
    },
    {
      "entropy": 5.799306106567383,
      "epoch": 3.4763610941186753,
      "grad_norm": 1.125,
      "learning_rate": 0.0003822118993534655,
      "loss": 5.3599,
      "mean_token_accuracy": 0.19509782642126083,
      "num_tokens": 80605060.0,
      "step": 39780
    },
    {
      "entropy": 5.765074300765991,
      "epoch": 3.47679804247138,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003821845022384348,
      "loss": 5.2805,
      "mean_token_accuracy": 0.20259252190589905,
      "num_tokens": 80614927.0,
      "step": 39785
    },
    {
      "entropy": 5.757108068466186,
      "epoch": 3.4772349908240847,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038215710306754906,
      "loss": 5.3494,
      "mean_token_accuracy": 0.20056080371141433,
      "num_tokens": 80624748.0,
      "step": 39790
    },
    {
      "entropy": 5.84643874168396,
      "epoch": 3.477671939176789,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038212970184133395,
      "loss": 5.4635,
      "mean_token_accuracy": 0.186288483440876,
      "num_tokens": 80635221.0,
      "step": 39795
    },
    {
      "entropy": 5.819649028778076,
      "epoch": 3.478108887529494,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038210229856031503,
      "loss": 5.4216,
      "mean_token_accuracy": 0.1951184630393982,
      "num_tokens": 80645268.0,
      "step": 39800
    },
    {
      "entropy": 5.976134157180786,
      "epoch": 3.4785458358821986,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003820748932250179,
      "loss": 5.4531,
      "mean_token_accuracy": 0.18154942244291306,
      "num_tokens": 80655209.0,
      "step": 39805
    },
    {
      "entropy": 5.827305459976197,
      "epoch": 3.4789827842349035,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038204748583596835,
      "loss": 5.4277,
      "mean_token_accuracy": 0.18922246843576432,
      "num_tokens": 80665652.0,
      "step": 39810
    },
    {
      "entropy": 5.730616569519043,
      "epoch": 3.4794197325876084,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003820200763936917,
      "loss": 5.4023,
      "mean_token_accuracy": 0.19621917605400085,
      "num_tokens": 80675430.0,
      "step": 39815
    },
    {
      "entropy": 5.8540181636810305,
      "epoch": 3.479856680940313,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003819926648987141,
      "loss": 5.3646,
      "mean_token_accuracy": 0.19368818551301956,
      "num_tokens": 80685997.0,
      "step": 39820
    },
    {
      "entropy": 5.8474620342254635,
      "epoch": 3.4802936292930173,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038196525135156094,
      "loss": 5.4214,
      "mean_token_accuracy": 0.1878158152103424,
      "num_tokens": 80697470.0,
      "step": 39825
    },
    {
      "entropy": 5.811095952987671,
      "epoch": 3.4807305776457222,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003819378357527583,
      "loss": 5.349,
      "mean_token_accuracy": 0.1932387873530388,
      "num_tokens": 80708810.0,
      "step": 39830
    },
    {
      "entropy": 5.742403984069824,
      "epoch": 3.481167525998427,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000381910418102832,
      "loss": 5.2895,
      "mean_token_accuracy": 0.20346793830394744,
      "num_tokens": 80718036.0,
      "step": 39835
    },
    {
      "entropy": 5.830156373977661,
      "epoch": 3.4816044743511316,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003818829984023078,
      "loss": 5.4142,
      "mean_token_accuracy": 0.19037824124097824,
      "num_tokens": 80728247.0,
      "step": 39840
    },
    {
      "entropy": 5.77269139289856,
      "epoch": 3.4820414227038365,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038185557665171174,
      "loss": 5.3888,
      "mean_token_accuracy": 0.19426493793725969,
      "num_tokens": 80738873.0,
      "step": 39845
    },
    {
      "entropy": 5.807488298416137,
      "epoch": 3.482478371056541,
      "grad_norm": 0.9375,
      "learning_rate": 0.00038182815285156964,
      "loss": 5.4013,
      "mean_token_accuracy": 0.184089694917202,
      "num_tokens": 80748734.0,
      "step": 39850
    },
    {
      "entropy": 5.830950736999512,
      "epoch": 3.482915319409246,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003818007270024076,
      "loss": 5.3655,
      "mean_token_accuracy": 0.20034806430339813,
      "num_tokens": 80759256.0,
      "step": 39855
    },
    {
      "entropy": 5.807183647155762,
      "epoch": 3.4833522677619504,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003817732991047518,
      "loss": 5.353,
      "mean_token_accuracy": 0.2003925174474716,
      "num_tokens": 80769670.0,
      "step": 39860
    },
    {
      "entropy": 5.752134466171265,
      "epoch": 3.4837892161146553,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038174586915912804,
      "loss": 5.3113,
      "mean_token_accuracy": 0.199858857691288,
      "num_tokens": 80778937.0,
      "step": 39865
    },
    {
      "entropy": 5.763743448257446,
      "epoch": 3.4842261644673598,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00038171843716606254,
      "loss": 5.3438,
      "mean_token_accuracy": 0.19498790800571442,
      "num_tokens": 80789329.0,
      "step": 39870
    },
    {
      "entropy": 5.764565134048462,
      "epoch": 3.4846631128200647,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038169100312608156,
      "loss": 5.3413,
      "mean_token_accuracy": 0.19898552745580672,
      "num_tokens": 80798943.0,
      "step": 39875
    },
    {
      "entropy": 5.854193687438965,
      "epoch": 3.485100061172769,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003816635670397112,
      "loss": 5.4421,
      "mean_token_accuracy": 0.18637582957744597,
      "num_tokens": 80808915.0,
      "step": 39880
    },
    {
      "entropy": 5.80160174369812,
      "epoch": 3.485537009525474,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038163612890747777,
      "loss": 5.3599,
      "mean_token_accuracy": 0.20070130079984666,
      "num_tokens": 80820162.0,
      "step": 39885
    },
    {
      "entropy": 5.714869356155395,
      "epoch": 3.485973957878179,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003816086887299075,
      "loss": 5.2967,
      "mean_token_accuracy": 0.21040673106908797,
      "num_tokens": 80830179.0,
      "step": 39890
    },
    {
      "entropy": 5.822159671783448,
      "epoch": 3.4864109062308835,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003815812465075267,
      "loss": 5.3465,
      "mean_token_accuracy": 0.19106957018375398,
      "num_tokens": 80839924.0,
      "step": 39895
    },
    {
      "entropy": 5.797197723388672,
      "epoch": 3.4868478545835884,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038155380224086177,
      "loss": 5.3673,
      "mean_token_accuracy": 0.1997702181339264,
      "num_tokens": 80849594.0,
      "step": 39900
    },
    {
      "entropy": 5.749416542053223,
      "epoch": 3.487284802936293,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003815263559304391,
      "loss": 5.3257,
      "mean_token_accuracy": 0.1956765204668045,
      "num_tokens": 80859706.0,
      "step": 39905
    },
    {
      "entropy": 5.845856237411499,
      "epoch": 3.4877217512889978,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038149890757678507,
      "loss": 5.4429,
      "mean_token_accuracy": 0.18783395141363143,
      "num_tokens": 80869121.0,
      "step": 39910
    },
    {
      "entropy": 5.812019729614258,
      "epoch": 3.488158699641702,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00038147145718042626,
      "loss": 5.4089,
      "mean_token_accuracy": 0.19547529220581056,
      "num_tokens": 80879347.0,
      "step": 39915
    },
    {
      "entropy": 5.81161060333252,
      "epoch": 3.488595647994407,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000381444004741889,
      "loss": 5.343,
      "mean_token_accuracy": 0.19725535959005355,
      "num_tokens": 80889819.0,
      "step": 39920
    },
    {
      "entropy": 5.841306161880493,
      "epoch": 3.4890325963471116,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003814165502617,
      "loss": 5.4165,
      "mean_token_accuracy": 0.19382905960083008,
      "num_tokens": 80900576.0,
      "step": 39925
    },
    {
      "entropy": 5.817193031311035,
      "epoch": 3.4894695446998165,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003813890937403858,
      "loss": 5.3628,
      "mean_token_accuracy": 0.19989619255065919,
      "num_tokens": 80910966.0,
      "step": 39930
    },
    {
      "entropy": 5.749672698974609,
      "epoch": 3.489906493052521,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038136163517847306,
      "loss": 5.3488,
      "mean_token_accuracy": 0.189080049097538,
      "num_tokens": 80920533.0,
      "step": 39935
    },
    {
      "entropy": 5.82741003036499,
      "epoch": 3.490343441405226,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038133417457648835,
      "loss": 5.4057,
      "mean_token_accuracy": 0.19401743710041047,
      "num_tokens": 80930989.0,
      "step": 39940
    },
    {
      "entropy": 5.8985175609588625,
      "epoch": 3.490780389757931,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038130671193495845,
      "loss": 5.4095,
      "mean_token_accuracy": 0.18585214018821716,
      "num_tokens": 80942283.0,
      "step": 39945
    },
    {
      "entropy": 5.836645317077637,
      "epoch": 3.4912173381106353,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038127924725441024,
      "loss": 5.3553,
      "mean_token_accuracy": 0.19212764948606492,
      "num_tokens": 80953258.0,
      "step": 39950
    },
    {
      "entropy": 5.726585960388183,
      "epoch": 3.49165428646334,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003812517805353703,
      "loss": 5.3286,
      "mean_token_accuracy": 0.2008584752678871,
      "num_tokens": 80963748.0,
      "step": 39955
    },
    {
      "entropy": 5.755689716339111,
      "epoch": 3.4920912348160447,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003812243117783656,
      "loss": 5.3815,
      "mean_token_accuracy": 0.19354016035795213,
      "num_tokens": 80973677.0,
      "step": 39960
    },
    {
      "entropy": 5.816460657119751,
      "epoch": 3.4925281831687496,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038119684098392284,
      "loss": 5.3869,
      "mean_token_accuracy": 0.19360256046056748,
      "num_tokens": 80984021.0,
      "step": 39965
    },
    {
      "entropy": 5.765130090713501,
      "epoch": 3.492965131521454,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000381169368152569,
      "loss": 5.2063,
      "mean_token_accuracy": 0.20964002311229707,
      "num_tokens": 80993498.0,
      "step": 39970
    },
    {
      "entropy": 5.802444219589233,
      "epoch": 3.493402079874159,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038114189328483124,
      "loss": 5.4162,
      "mean_token_accuracy": 0.1899924248456955,
      "num_tokens": 81002693.0,
      "step": 39975
    },
    {
      "entropy": 5.803858947753906,
      "epoch": 3.4938390282268634,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003811144163812362,
      "loss": 5.4393,
      "mean_token_accuracy": 0.1892872244119644,
      "num_tokens": 81012715.0,
      "step": 39980
    },
    {
      "entropy": 5.837252473831176,
      "epoch": 3.4942759765795683,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038108693744231105,
      "loss": 5.3715,
      "mean_token_accuracy": 0.19435057938098907,
      "num_tokens": 81023779.0,
      "step": 39985
    },
    {
      "entropy": 5.76839542388916,
      "epoch": 3.494712924932273,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003810594564685829,
      "loss": 5.3345,
      "mean_token_accuracy": 0.21034308075904845,
      "num_tokens": 81035103.0,
      "step": 39990
    },
    {
      "entropy": 5.78112301826477,
      "epoch": 3.4951498732849777,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003810319734605787,
      "loss": 5.37,
      "mean_token_accuracy": 0.19115860462188722,
      "num_tokens": 81045011.0,
      "step": 39995
    },
    {
      "entropy": 5.797879028320312,
      "epoch": 3.4955868216376826,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003810044884188258,
      "loss": 5.3845,
      "mean_token_accuracy": 0.1930662527680397,
      "num_tokens": 81055841.0,
      "step": 40000
    },
    {
      "entropy": 5.798837804794312,
      "epoch": 3.496023769990387,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003809770013438512,
      "loss": 5.4083,
      "mean_token_accuracy": 0.1914048194885254,
      "num_tokens": 81067270.0,
      "step": 40005
    },
    {
      "entropy": 5.840986633300782,
      "epoch": 3.496460718343092,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038094951223618224,
      "loss": 5.3754,
      "mean_token_accuracy": 0.19607421457767488,
      "num_tokens": 81076963.0,
      "step": 40010
    },
    {
      "entropy": 5.804752254486084,
      "epoch": 3.4968976666957965,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038092202109634603,
      "loss": 5.3477,
      "mean_token_accuracy": 0.19404055327177047,
      "num_tokens": 81086146.0,
      "step": 40015
    },
    {
      "entropy": 5.788507318496704,
      "epoch": 3.4973346150485014,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038089452792487005,
      "loss": 5.354,
      "mean_token_accuracy": 0.19728409945964814,
      "num_tokens": 81095924.0,
      "step": 40020
    },
    {
      "entropy": 5.8224263191223145,
      "epoch": 3.497771563401206,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003808670327222815,
      "loss": 5.371,
      "mean_token_accuracy": 0.1906808838248253,
      "num_tokens": 81106396.0,
      "step": 40025
    },
    {
      "entropy": 5.815728187561035,
      "epoch": 3.498208511753911,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038083953548910776,
      "loss": 5.3227,
      "mean_token_accuracy": 0.19352494776248932,
      "num_tokens": 81117098.0,
      "step": 40030
    },
    {
      "entropy": 5.704619646072388,
      "epoch": 3.4986454601066153,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00038081203622587637,
      "loss": 5.3202,
      "mean_token_accuracy": 0.19209887236356735,
      "num_tokens": 81127335.0,
      "step": 40035
    },
    {
      "entropy": 5.795662689208984,
      "epoch": 3.49908240845932,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038078453493311463,
      "loss": 5.3487,
      "mean_token_accuracy": 0.19170138239860535,
      "num_tokens": 81136608.0,
      "step": 40040
    },
    {
      "entropy": 5.889249181747436,
      "epoch": 3.4995193568120246,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038075703161135,
      "loss": 5.4542,
      "mean_token_accuracy": 0.18486583083868027,
      "num_tokens": 81146690.0,
      "step": 40045
    },
    {
      "entropy": 5.756820201873779,
      "epoch": 3.4999563051647296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038072952626111016,
      "loss": 5.2877,
      "mean_token_accuracy": 0.19876244515180588,
      "num_tokens": 81155914.0,
      "step": 40050
    },
    {
      "entropy": 5.7488080024719235,
      "epoch": 3.5003932535174345,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003807020188829226,
      "loss": 5.3323,
      "mean_token_accuracy": 0.19787991493940355,
      "num_tokens": 81166537.0,
      "step": 40055
    },
    {
      "entropy": 5.80820894241333,
      "epoch": 3.500830201870139,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003806745094773149,
      "loss": 5.3331,
      "mean_token_accuracy": 0.19646632820367813,
      "num_tokens": 81176328.0,
      "step": 40060
    },
    {
      "entropy": 5.791860866546631,
      "epoch": 3.5012671502228434,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038064699804481476,
      "loss": 5.3902,
      "mean_token_accuracy": 0.19276347309350966,
      "num_tokens": 81186345.0,
      "step": 40065
    },
    {
      "entropy": 5.781443405151367,
      "epoch": 3.5017040985755483,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003806194845859499,
      "loss": 5.3713,
      "mean_token_accuracy": 0.19225073009729385,
      "num_tokens": 81196683.0,
      "step": 40070
    },
    {
      "entropy": 5.7932640552520756,
      "epoch": 3.5021410469282532,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003805919691012479,
      "loss": 5.3526,
      "mean_token_accuracy": 0.19804300367832184,
      "num_tokens": 81206381.0,
      "step": 40075
    },
    {
      "entropy": 5.836909914016724,
      "epoch": 3.5025779952809577,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038056445159123664,
      "loss": 5.446,
      "mean_token_accuracy": 0.1911589041352272,
      "num_tokens": 81217056.0,
      "step": 40080
    },
    {
      "entropy": 5.816232061386108,
      "epoch": 3.5030149436336626,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003805369320564438,
      "loss": 5.356,
      "mean_token_accuracy": 0.19727783948183059,
      "num_tokens": 81227803.0,
      "step": 40085
    },
    {
      "entropy": 5.811060190200806,
      "epoch": 3.503451891986367,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003805094104973974,
      "loss": 5.3337,
      "mean_token_accuracy": 0.19668003916740417,
      "num_tokens": 81237680.0,
      "step": 40090
    },
    {
      "entropy": 5.823747348785401,
      "epoch": 3.503888840339072,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003804818869146252,
      "loss": 5.3595,
      "mean_token_accuracy": 0.19548023641109466,
      "num_tokens": 81247624.0,
      "step": 40095
    },
    {
      "entropy": 5.840116548538208,
      "epoch": 3.5043257886917765,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038045436130865515,
      "loss": 5.3464,
      "mean_token_accuracy": 0.19610957354307174,
      "num_tokens": 81257932.0,
      "step": 40100
    },
    {
      "entropy": 5.842609548568726,
      "epoch": 3.5047627370444814,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003804268336800151,
      "loss": 5.3966,
      "mean_token_accuracy": 0.19319830387830733,
      "num_tokens": 81268595.0,
      "step": 40105
    },
    {
      "entropy": 5.804691314697266,
      "epoch": 3.5051996853971863,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038039930402923323,
      "loss": 5.3291,
      "mean_token_accuracy": 0.19845246970653535,
      "num_tokens": 81279252.0,
      "step": 40110
    },
    {
      "entropy": 5.748768949508667,
      "epoch": 3.5056366337498908,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038037177235683746,
      "loss": 5.3221,
      "mean_token_accuracy": 0.19933997839689255,
      "num_tokens": 81289897.0,
      "step": 40115
    },
    {
      "entropy": 5.780804777145386,
      "epoch": 3.5060735821025952,
      "grad_norm": 1.125,
      "learning_rate": 0.00038034423866335586,
      "loss": 5.3626,
      "mean_token_accuracy": 0.19363826960325242,
      "num_tokens": 81300297.0,
      "step": 40120
    },
    {
      "entropy": 5.860300827026367,
      "epoch": 3.5065105304553,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003803167029493166,
      "loss": 5.385,
      "mean_token_accuracy": 0.18868109583854675,
      "num_tokens": 81309628.0,
      "step": 40125
    },
    {
      "entropy": 5.77282452583313,
      "epoch": 3.506947478808005,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038028916521524775,
      "loss": 5.3733,
      "mean_token_accuracy": 0.19337102323770522,
      "num_tokens": 81320511.0,
      "step": 40130
    },
    {
      "entropy": 5.780802297592163,
      "epoch": 3.5073844271607095,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003802616254616776,
      "loss": 5.3673,
      "mean_token_accuracy": 0.19318604916334153,
      "num_tokens": 81330982.0,
      "step": 40135
    },
    {
      "entropy": 5.821206283569336,
      "epoch": 3.5078213755134144,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003802340836891343,
      "loss": 5.366,
      "mean_token_accuracy": 0.19465139657258987,
      "num_tokens": 81339955.0,
      "step": 40140
    },
    {
      "entropy": 5.803774118423462,
      "epoch": 3.508258323866119,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038020653989814606,
      "loss": 5.3492,
      "mean_token_accuracy": 0.19849766939878463,
      "num_tokens": 81349151.0,
      "step": 40145
    },
    {
      "entropy": 5.8200767993927,
      "epoch": 3.508695272218824,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003801789940892413,
      "loss": 5.2797,
      "mean_token_accuracy": 0.19879917353391646,
      "num_tokens": 81359563.0,
      "step": 40150
    },
    {
      "entropy": 5.805410718917846,
      "epoch": 3.5091322205715283,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003801514462629483,
      "loss": 5.4072,
      "mean_token_accuracy": 0.1915351152420044,
      "num_tokens": 81369669.0,
      "step": 40155
    },
    {
      "entropy": 5.782684278488159,
      "epoch": 3.509569168924233,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038012389641979555,
      "loss": 5.4359,
      "mean_token_accuracy": 0.18259939253330232,
      "num_tokens": 81380322.0,
      "step": 40160
    },
    {
      "entropy": 5.867336082458496,
      "epoch": 3.510006117276938,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003800963445603114,
      "loss": 5.3959,
      "mean_token_accuracy": 0.19502140134572982,
      "num_tokens": 81389871.0,
      "step": 40165
    },
    {
      "entropy": 5.738810205459595,
      "epoch": 3.5104430656296426,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003800687906850242,
      "loss": 5.2389,
      "mean_token_accuracy": 0.20725541114807128,
      "num_tokens": 81399684.0,
      "step": 40170
    },
    {
      "entropy": 5.800620985031128,
      "epoch": 3.510880013982347,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003800412347944626,
      "loss": 5.3596,
      "mean_token_accuracy": 0.19105800837278367,
      "num_tokens": 81409638.0,
      "step": 40175
    },
    {
      "entropy": 5.852757310867309,
      "epoch": 3.511316962335052,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003800136768891549,
      "loss": 5.378,
      "mean_token_accuracy": 0.19695538878440857,
      "num_tokens": 81419963.0,
      "step": 40180
    },
    {
      "entropy": 5.783743190765381,
      "epoch": 3.511753910687757,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037998611696963,
      "loss": 5.2984,
      "mean_token_accuracy": 0.19553535729646682,
      "num_tokens": 81429747.0,
      "step": 40185
    },
    {
      "entropy": 5.770343017578125,
      "epoch": 3.5121908590404614,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003799585550364164,
      "loss": 5.2999,
      "mean_token_accuracy": 0.1961883544921875,
      "num_tokens": 81439012.0,
      "step": 40190
    },
    {
      "entropy": 5.826728391647339,
      "epoch": 3.5126278073931663,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003799309910900427,
      "loss": 5.4234,
      "mean_token_accuracy": 0.1918175607919693,
      "num_tokens": 81449713.0,
      "step": 40195
    },
    {
      "entropy": 5.764848136901856,
      "epoch": 3.5130647557458707,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003799034251310376,
      "loss": 5.3054,
      "mean_token_accuracy": 0.20182280987501144,
      "num_tokens": 81459326.0,
      "step": 40200
    },
    {
      "entropy": 5.815541839599609,
      "epoch": 3.5135017040985757,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037987585715992974,
      "loss": 5.3331,
      "mean_token_accuracy": 0.2016705945134163,
      "num_tokens": 81469419.0,
      "step": 40205
    },
    {
      "entropy": 5.738886308670044,
      "epoch": 3.51393865245128,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037984828717724814,
      "loss": 5.3423,
      "mean_token_accuracy": 0.2030295968055725,
      "num_tokens": 81478494.0,
      "step": 40210
    },
    {
      "entropy": 5.742088270187378,
      "epoch": 3.514375600803985,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037982071518352135,
      "loss": 5.3458,
      "mean_token_accuracy": 0.19342704713344575,
      "num_tokens": 81488080.0,
      "step": 40215
    },
    {
      "entropy": 5.83881025314331,
      "epoch": 3.51481254915669,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003797931411792783,
      "loss": 5.3909,
      "mean_token_accuracy": 0.19532307386398315,
      "num_tokens": 81498075.0,
      "step": 40220
    },
    {
      "entropy": 5.846643400192261,
      "epoch": 3.5152494975093944,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037976556516504796,
      "loss": 5.3471,
      "mean_token_accuracy": 0.1931694656610489,
      "num_tokens": 81508407.0,
      "step": 40225
    },
    {
      "entropy": 5.837780904769898,
      "epoch": 3.515686445862099,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037973798714135916,
      "loss": 5.3599,
      "mean_token_accuracy": 0.18299242705106736,
      "num_tokens": 81518009.0,
      "step": 40230
    },
    {
      "entropy": 5.743351030349731,
      "epoch": 3.516123394214804,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003797104071087409,
      "loss": 5.3554,
      "mean_token_accuracy": 0.20457558631896972,
      "num_tokens": 81527468.0,
      "step": 40235
    },
    {
      "entropy": 5.795500230789185,
      "epoch": 3.5165603425675087,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003796828250677221,
      "loss": 5.3748,
      "mean_token_accuracy": 0.19317141622304917,
      "num_tokens": 81538133.0,
      "step": 40240
    },
    {
      "entropy": 5.770478248596191,
      "epoch": 3.516997290920213,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037965524101883196,
      "loss": 5.2981,
      "mean_token_accuracy": 0.20059166550636293,
      "num_tokens": 81548051.0,
      "step": 40245
    },
    {
      "entropy": 5.802758693695068,
      "epoch": 3.517434239272918,
      "grad_norm": 1.0,
      "learning_rate": 0.0003796276549625994,
      "loss": 5.4186,
      "mean_token_accuracy": 0.19085308015346528,
      "num_tokens": 81557984.0,
      "step": 40250
    },
    {
      "entropy": 5.76617579460144,
      "epoch": 3.5178711876256226,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037960006689955356,
      "loss": 5.3745,
      "mean_token_accuracy": 0.20084408968687056,
      "num_tokens": 81568587.0,
      "step": 40255
    },
    {
      "entropy": 5.788554239273071,
      "epoch": 3.5183081359783275,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037957247683022375,
      "loss": 5.3422,
      "mean_token_accuracy": 0.1927005246281624,
      "num_tokens": 81578085.0,
      "step": 40260
    },
    {
      "entropy": 5.85494818687439,
      "epoch": 3.518745084331032,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003795448847551389,
      "loss": 5.4634,
      "mean_token_accuracy": 0.18501547873020172,
      "num_tokens": 81588037.0,
      "step": 40265
    },
    {
      "entropy": 5.809961557388306,
      "epoch": 3.519182032683737,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037951729067482845,
      "loss": 5.3702,
      "mean_token_accuracy": 0.1978350266814232,
      "num_tokens": 81598364.0,
      "step": 40270
    },
    {
      "entropy": 5.849759960174561,
      "epoch": 3.5196189810364413,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037948969458982153,
      "loss": 5.3341,
      "mean_token_accuracy": 0.1903867617249489,
      "num_tokens": 81609428.0,
      "step": 40275
    },
    {
      "entropy": 5.735621786117553,
      "epoch": 3.5200559293891462,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037946209650064754,
      "loss": 5.3881,
      "mean_token_accuracy": 0.18566359281539918,
      "num_tokens": 81619605.0,
      "step": 40280
    },
    {
      "entropy": 5.826530694961548,
      "epoch": 3.5204928777418507,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003794344964078359,
      "loss": 5.4023,
      "mean_token_accuracy": 0.18695327937602996,
      "num_tokens": 81628987.0,
      "step": 40285
    },
    {
      "entropy": 5.801140022277832,
      "epoch": 3.5209298260945556,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003794068943119159,
      "loss": 5.3245,
      "mean_token_accuracy": 0.19911540448665618,
      "num_tokens": 81637628.0,
      "step": 40290
    },
    {
      "entropy": 5.7126696586608885,
      "epoch": 3.5213667744472605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003793792902134168,
      "loss": 5.2912,
      "mean_token_accuracy": 0.20323529243469238,
      "num_tokens": 81647470.0,
      "step": 40295
    },
    {
      "entropy": 5.787311267852783,
      "epoch": 3.521803722799965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003793516841128683,
      "loss": 5.3421,
      "mean_token_accuracy": 0.19437724351882935,
      "num_tokens": 81657882.0,
      "step": 40300
    },
    {
      "entropy": 5.819318246841431,
      "epoch": 3.5222406711526695,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037932407601079987,
      "loss": 5.3889,
      "mean_token_accuracy": 0.19301040470600128,
      "num_tokens": 81668506.0,
      "step": 40305
    },
    {
      "entropy": 5.792092370986938,
      "epoch": 3.5226776195053744,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037929646590774093,
      "loss": 5.3721,
      "mean_token_accuracy": 0.18976324796676636,
      "num_tokens": 81678590.0,
      "step": 40310
    },
    {
      "entropy": 5.763207912445068,
      "epoch": 3.5231145678580793,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003792688538042211,
      "loss": 5.381,
      "mean_token_accuracy": 0.20366569310426713,
      "num_tokens": 81689625.0,
      "step": 40315
    },
    {
      "entropy": 5.820023536682129,
      "epoch": 3.5235515162107838,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003792412397007701,
      "loss": 5.4114,
      "mean_token_accuracy": 0.1948063537478447,
      "num_tokens": 81699251.0,
      "step": 40320
    },
    {
      "entropy": 5.836707639694214,
      "epoch": 3.5239884645634887,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037921362359791743,
      "loss": 5.3073,
      "mean_token_accuracy": 0.20456915646791457,
      "num_tokens": 81708851.0,
      "step": 40325
    },
    {
      "entropy": 5.796651220321655,
      "epoch": 3.524425412916193,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003791860054961929,
      "loss": 5.367,
      "mean_token_accuracy": 0.1891581282019615,
      "num_tokens": 81718399.0,
      "step": 40330
    },
    {
      "entropy": 5.827752494812012,
      "epoch": 3.524862361268898,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037915838539612624,
      "loss": 5.4059,
      "mean_token_accuracy": 0.19907266795635223,
      "num_tokens": 81728546.0,
      "step": 40335
    },
    {
      "entropy": 5.887000560760498,
      "epoch": 3.5252993096216025,
      "grad_norm": 1.125,
      "learning_rate": 0.000379130763298247,
      "loss": 5.3506,
      "mean_token_accuracy": 0.1945217341184616,
      "num_tokens": 81737830.0,
      "step": 40340
    },
    {
      "entropy": 5.730906581878662,
      "epoch": 3.5257362579743075,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003791031392030853,
      "loss": 5.2996,
      "mean_token_accuracy": 0.1996949389576912,
      "num_tokens": 81747534.0,
      "step": 40345
    },
    {
      "entropy": 5.75040979385376,
      "epoch": 3.5261732063270124,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037907551311117083,
      "loss": 5.3826,
      "mean_token_accuracy": 0.19604501128196716,
      "num_tokens": 81758411.0,
      "step": 40350
    },
    {
      "entropy": 5.805878591537476,
      "epoch": 3.526610154679717,
      "grad_norm": 1.0,
      "learning_rate": 0.0003790478850230335,
      "loss": 5.3399,
      "mean_token_accuracy": 0.1927989974617958,
      "num_tokens": 81768441.0,
      "step": 40355
    },
    {
      "entropy": 5.847728824615478,
      "epoch": 3.5270471030324213,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037902025493920314,
      "loss": 5.4096,
      "mean_token_accuracy": 0.1970725417137146,
      "num_tokens": 81778899.0,
      "step": 40360
    },
    {
      "entropy": 5.847040700912475,
      "epoch": 3.527484051385126,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037899262286020984,
      "loss": 5.3596,
      "mean_token_accuracy": 0.19548549503087997,
      "num_tokens": 81788643.0,
      "step": 40365
    },
    {
      "entropy": 5.78129563331604,
      "epoch": 3.527920999737831,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037896498878658345,
      "loss": 5.3625,
      "mean_token_accuracy": 0.19318372905254363,
      "num_tokens": 81798781.0,
      "step": 40370
    },
    {
      "entropy": 5.800914764404297,
      "epoch": 3.5283579480905356,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003789373527188542,
      "loss": 5.3338,
      "mean_token_accuracy": 0.20056002736091613,
      "num_tokens": 81808402.0,
      "step": 40375
    },
    {
      "entropy": 5.816304779052734,
      "epoch": 3.5287948964432405,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003789097146575521,
      "loss": 5.3937,
      "mean_token_accuracy": 0.19228834211826323,
      "num_tokens": 81819285.0,
      "step": 40380
    },
    {
      "entropy": 5.788665390014648,
      "epoch": 3.529231844795945,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003788820746032071,
      "loss": 5.2993,
      "mean_token_accuracy": 0.19616834223270416,
      "num_tokens": 81829357.0,
      "step": 40385
    },
    {
      "entropy": 5.796677970886231,
      "epoch": 3.52966879314865,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037885443255634955,
      "loss": 5.2898,
      "mean_token_accuracy": 0.20263197869062424,
      "num_tokens": 81838547.0,
      "step": 40390
    },
    {
      "entropy": 5.816992855072021,
      "epoch": 3.5301057415013544,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003788267885175095,
      "loss": 5.3854,
      "mean_token_accuracy": 0.19192425459623336,
      "num_tokens": 81849286.0,
      "step": 40395
    },
    {
      "entropy": 5.913304328918457,
      "epoch": 3.5305426898540593,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037879914248721737,
      "loss": 5.4411,
      "mean_token_accuracy": 0.18614507019519805,
      "num_tokens": 81859599.0,
      "step": 40400
    },
    {
      "entropy": 5.791816854476929,
      "epoch": 3.530979638206764,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003787714944660032,
      "loss": 5.3421,
      "mean_token_accuracy": 0.20363471359014512,
      "num_tokens": 81869497.0,
      "step": 40405
    },
    {
      "entropy": 5.778452825546265,
      "epoch": 3.5314165865594687,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037874384445439743,
      "loss": 5.2959,
      "mean_token_accuracy": 0.20165136307477952,
      "num_tokens": 81878502.0,
      "step": 40410
    },
    {
      "entropy": 5.75891079902649,
      "epoch": 3.531853534912173,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003787161924529303,
      "loss": 5.2774,
      "mean_token_accuracy": 0.2029660865664482,
      "num_tokens": 81888748.0,
      "step": 40415
    },
    {
      "entropy": 5.756647062301636,
      "epoch": 3.532290483264878,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003786885384621323,
      "loss": 5.2759,
      "mean_token_accuracy": 0.20579937994480133,
      "num_tokens": 81897705.0,
      "step": 40420
    },
    {
      "entropy": 5.758359098434449,
      "epoch": 3.532727431617583,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003786608824825338,
      "loss": 5.4153,
      "mean_token_accuracy": 0.1959238037467003,
      "num_tokens": 81909294.0,
      "step": 40425
    },
    {
      "entropy": 5.91106915473938,
      "epoch": 3.5331643799702874,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037863322451466523,
      "loss": 5.4469,
      "mean_token_accuracy": 0.18615524768829345,
      "num_tokens": 81919191.0,
      "step": 40430
    },
    {
      "entropy": 5.809289598464966,
      "epoch": 3.5336013283229923,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003786055645590571,
      "loss": 5.2976,
      "mean_token_accuracy": 0.206255641579628,
      "num_tokens": 81928660.0,
      "step": 40435
    },
    {
      "entropy": 5.7555468559265135,
      "epoch": 3.534038276675697,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037857790261624006,
      "loss": 5.3512,
      "mean_token_accuracy": 0.2021782875061035,
      "num_tokens": 81938626.0,
      "step": 40440
    },
    {
      "entropy": 5.775583410263062,
      "epoch": 3.5344752250284017,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003785502386867445,
      "loss": 5.3541,
      "mean_token_accuracy": 0.20298111885786058,
      "num_tokens": 81948304.0,
      "step": 40445
    },
    {
      "entropy": 5.755040311813355,
      "epoch": 3.534912173381106,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037852257277110106,
      "loss": 5.3477,
      "mean_token_accuracy": 0.2008087858557701,
      "num_tokens": 81958455.0,
      "step": 40450
    },
    {
      "entropy": 5.736729717254638,
      "epoch": 3.535349121733811,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003784949048698405,
      "loss": 5.242,
      "mean_token_accuracy": 0.21347091048955918,
      "num_tokens": 81967809.0,
      "step": 40455
    },
    {
      "entropy": 5.790262460708618,
      "epoch": 3.535786070086516,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003784672349834934,
      "loss": 5.3147,
      "mean_token_accuracy": 0.20334353893995286,
      "num_tokens": 81979801.0,
      "step": 40460
    },
    {
      "entropy": 5.790998220443726,
      "epoch": 3.5362230184392205,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037843956311259047,
      "loss": 5.3581,
      "mean_token_accuracy": 0.19785282015800476,
      "num_tokens": 81989591.0,
      "step": 40465
    },
    {
      "entropy": 5.823132705688477,
      "epoch": 3.536659966791925,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003784118892576626,
      "loss": 5.3353,
      "mean_token_accuracy": 0.19585055708885193,
      "num_tokens": 81999470.0,
      "step": 40470
    },
    {
      "entropy": 5.719961881637573,
      "epoch": 3.53709691514463,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037838421341924045,
      "loss": 5.3289,
      "mean_token_accuracy": 0.199250990152359,
      "num_tokens": 82010307.0,
      "step": 40475
    },
    {
      "entropy": 5.74740424156189,
      "epoch": 3.537533863497335,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037835653559785496,
      "loss": 5.2696,
      "mean_token_accuracy": 0.20433515459299087,
      "num_tokens": 82019997.0,
      "step": 40480
    },
    {
      "entropy": 5.790434885025024,
      "epoch": 3.5379708118500393,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003783288557940369,
      "loss": 5.2668,
      "mean_token_accuracy": 0.20037078708410264,
      "num_tokens": 82030328.0,
      "step": 40485
    },
    {
      "entropy": 5.837762260437012,
      "epoch": 3.538407760202744,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003783011740083173,
      "loss": 5.41,
      "mean_token_accuracy": 0.18798351436853408,
      "num_tokens": 82041153.0,
      "step": 40490
    },
    {
      "entropy": 5.688846158981323,
      "epoch": 3.5388447085554486,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003782734902412271,
      "loss": 5.225,
      "mean_token_accuracy": 0.21661083102226258,
      "num_tokens": 82050962.0,
      "step": 40495
    },
    {
      "entropy": 5.768764543533325,
      "epoch": 3.5392816569081536,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037824580449329716,
      "loss": 5.3593,
      "mean_token_accuracy": 0.20080561339855194,
      "num_tokens": 82061033.0,
      "step": 40500
    },
    {
      "entropy": 5.827947044372559,
      "epoch": 3.539718605260858,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037821811676505855,
      "loss": 5.3347,
      "mean_token_accuracy": 0.19961430877447128,
      "num_tokens": 82070352.0,
      "step": 40505
    },
    {
      "entropy": 5.822429561614991,
      "epoch": 3.540155553613563,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037819042705704244,
      "loss": 5.3505,
      "mean_token_accuracy": 0.19846716225147248,
      "num_tokens": 82080608.0,
      "step": 40510
    },
    {
      "entropy": 5.8421577453613285,
      "epoch": 3.540592501966268,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037816273536977986,
      "loss": 5.3042,
      "mean_token_accuracy": 0.20197907984256744,
      "num_tokens": 82089935.0,
      "step": 40515
    },
    {
      "entropy": 5.805672740936279,
      "epoch": 3.5410294503189723,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000378135041703802,
      "loss": 5.3569,
      "mean_token_accuracy": 0.1979023441672325,
      "num_tokens": 82100983.0,
      "step": 40520
    },
    {
      "entropy": 5.850345230102539,
      "epoch": 3.541466398671677,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037810734605963996,
      "loss": 5.425,
      "mean_token_accuracy": 0.1862711638212204,
      "num_tokens": 82110791.0,
      "step": 40525
    },
    {
      "entropy": 5.812013292312622,
      "epoch": 3.5419033470243817,
      "grad_norm": 1.296875,
      "learning_rate": 0.000378079648437825,
      "loss": 5.3176,
      "mean_token_accuracy": 0.19842013716697693,
      "num_tokens": 82120330.0,
      "step": 40530
    },
    {
      "entropy": 5.8892735004425045,
      "epoch": 3.5423402953770866,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037805194883888834,
      "loss": 5.3879,
      "mean_token_accuracy": 0.1909844234585762,
      "num_tokens": 82130589.0,
      "step": 40535
    },
    {
      "entropy": 5.785531759262085,
      "epoch": 3.542777243729791,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003780242472633614,
      "loss": 5.3086,
      "mean_token_accuracy": 0.20633533149957656,
      "num_tokens": 82141421.0,
      "step": 40540
    },
    {
      "entropy": 5.759484767913818,
      "epoch": 3.543214192082496,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003779965437117753,
      "loss": 5.3555,
      "mean_token_accuracy": 0.19553417712450027,
      "num_tokens": 82150688.0,
      "step": 40545
    },
    {
      "entropy": 5.828740549087525,
      "epoch": 3.5436511404352005,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037796883818466155,
      "loss": 5.41,
      "mean_token_accuracy": 0.1881601020693779,
      "num_tokens": 82161850.0,
      "step": 40550
    },
    {
      "entropy": 5.863890361785889,
      "epoch": 3.5440880887879054,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037794113068255155,
      "loss": 5.3518,
      "mean_token_accuracy": 0.19365316927433013,
      "num_tokens": 82171492.0,
      "step": 40555
    },
    {
      "entropy": 5.836336469650268,
      "epoch": 3.54452503714061,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037791342120597674,
      "loss": 5.3967,
      "mean_token_accuracy": 0.1926010638475418,
      "num_tokens": 82182469.0,
      "step": 40560
    },
    {
      "entropy": 5.8065284252166744,
      "epoch": 3.5449619854933148,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00037788570975546854,
      "loss": 5.33,
      "mean_token_accuracy": 0.1960781767964363,
      "num_tokens": 82192395.0,
      "step": 40565
    },
    {
      "entropy": 5.797710609436035,
      "epoch": 3.5453989338460197,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037785799633155854,
      "loss": 5.3449,
      "mean_token_accuracy": 0.1940249428153038,
      "num_tokens": 82201555.0,
      "step": 40570
    },
    {
      "entropy": 5.767383432388305,
      "epoch": 3.545835882198724,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003778302809347782,
      "loss": 5.349,
      "mean_token_accuracy": 0.19619117975234984,
      "num_tokens": 82211013.0,
      "step": 40575
    },
    {
      "entropy": 5.831550550460816,
      "epoch": 3.5462728305514286,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003778025635656593,
      "loss": 5.3551,
      "mean_token_accuracy": 0.19234956800937653,
      "num_tokens": 82221317.0,
      "step": 40580
    },
    {
      "entropy": 5.78925895690918,
      "epoch": 3.5467097789041335,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003777748442247333,
      "loss": 5.3073,
      "mean_token_accuracy": 0.20555862039327621,
      "num_tokens": 82230837.0,
      "step": 40585
    },
    {
      "entropy": 5.7997393131256105,
      "epoch": 3.5471467272568384,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000377747122912532,
      "loss": 5.3865,
      "mean_token_accuracy": 0.18970917165279388,
      "num_tokens": 82241034.0,
      "step": 40590
    },
    {
      "entropy": 5.7779848098754885,
      "epoch": 3.547583675609543,
      "grad_norm": 1.125,
      "learning_rate": 0.000377719399629587,
      "loss": 5.3524,
      "mean_token_accuracy": 0.19471572190523148,
      "num_tokens": 82251047.0,
      "step": 40595
    },
    {
      "entropy": 5.899235677719116,
      "epoch": 3.548020623962248,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003776916743764301,
      "loss": 5.365,
      "mean_token_accuracy": 0.1951907083392143,
      "num_tokens": 82261215.0,
      "step": 40600
    },
    {
      "entropy": 5.816900825500488,
      "epoch": 3.5484575723149523,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003776639471535931,
      "loss": 5.3686,
      "mean_token_accuracy": 0.1956222414970398,
      "num_tokens": 82271527.0,
      "step": 40605
    },
    {
      "entropy": 5.848743104934693,
      "epoch": 3.548894520667657,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003776362179616077,
      "loss": 5.4712,
      "mean_token_accuracy": 0.18488092422485353,
      "num_tokens": 82282021.0,
      "step": 40610
    },
    {
      "entropy": 5.863064098358154,
      "epoch": 3.5493314690203617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037760848680100595,
      "loss": 5.4311,
      "mean_token_accuracy": 0.19014002680778502,
      "num_tokens": 82291279.0,
      "step": 40615
    },
    {
      "entropy": 5.775391006469727,
      "epoch": 3.5497684173730666,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003775807536723196,
      "loss": 5.2967,
      "mean_token_accuracy": 0.20049454271793365,
      "num_tokens": 82300756.0,
      "step": 40620
    },
    {
      "entropy": 5.8409511089324955,
      "epoch": 3.550205365725771,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003775530185760807,
      "loss": 5.37,
      "mean_token_accuracy": 0.19776571094989776,
      "num_tokens": 82309896.0,
      "step": 40625
    },
    {
      "entropy": 5.7276153564453125,
      "epoch": 3.550642314078476,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003775252815128211,
      "loss": 5.2285,
      "mean_token_accuracy": 0.20786117762327194,
      "num_tokens": 82319646.0,
      "step": 40630
    },
    {
      "entropy": 5.761891412734985,
      "epoch": 3.5510792624311804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003774975424830729,
      "loss": 5.3919,
      "mean_token_accuracy": 0.1933552846312523,
      "num_tokens": 82329492.0,
      "step": 40635
    },
    {
      "entropy": 5.778854846954346,
      "epoch": 3.5515162107838854,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003774698014873682,
      "loss": 5.3394,
      "mean_token_accuracy": 0.20143123865127563,
      "num_tokens": 82340339.0,
      "step": 40640
    },
    {
      "entropy": 5.827274179458618,
      "epoch": 3.5519531591365903,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037744205852623895,
      "loss": 5.3533,
      "mean_token_accuracy": 0.19541082829236983,
      "num_tokens": 82351337.0,
      "step": 40645
    },
    {
      "entropy": 5.786427545547485,
      "epoch": 3.5523901074892947,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003774143136002173,
      "loss": 5.3368,
      "mean_token_accuracy": 0.20521182119846343,
      "num_tokens": 82361616.0,
      "step": 40650
    },
    {
      "entropy": 5.749963569641113,
      "epoch": 3.552827055841999,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037738656670983553,
      "loss": 5.3044,
      "mean_token_accuracy": 0.18850114047527314,
      "num_tokens": 82371359.0,
      "step": 40655
    },
    {
      "entropy": 5.71950855255127,
      "epoch": 3.553264004194704,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003773588178556258,
      "loss": 5.247,
      "mean_token_accuracy": 0.19730072766542434,
      "num_tokens": 82381575.0,
      "step": 40660
    },
    {
      "entropy": 5.747772216796875,
      "epoch": 3.553700952547409,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003773310670381202,
      "loss": 5.3368,
      "mean_token_accuracy": 0.1973538503050804,
      "num_tokens": 82392175.0,
      "step": 40665
    },
    {
      "entropy": 5.844163799285889,
      "epoch": 3.5541379009001135,
      "grad_norm": 1.53125,
      "learning_rate": 0.00037730331425785114,
      "loss": 5.405,
      "mean_token_accuracy": 0.1920766443014145,
      "num_tokens": 82402570.0,
      "step": 40670
    },
    {
      "entropy": 5.867616796493531,
      "epoch": 3.5545748492528184,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00037727555951535093,
      "loss": 5.3689,
      "mean_token_accuracy": 0.19384675621986389,
      "num_tokens": 82413433.0,
      "step": 40675
    },
    {
      "entropy": 5.84561767578125,
      "epoch": 3.555011797605523,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003772478028111519,
      "loss": 5.376,
      "mean_token_accuracy": 0.18951232433319093,
      "num_tokens": 82423144.0,
      "step": 40680
    },
    {
      "entropy": 5.7510786056518555,
      "epoch": 3.555448745958228,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003772200441457864,
      "loss": 5.3045,
      "mean_token_accuracy": 0.19760207682847977,
      "num_tokens": 82433096.0,
      "step": 40685
    },
    {
      "entropy": 5.711239242553711,
      "epoch": 3.5558856943109323,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003771922835197869,
      "loss": 5.2835,
      "mean_token_accuracy": 0.20485470592975616,
      "num_tokens": 82444261.0,
      "step": 40690
    },
    {
      "entropy": 5.863011407852173,
      "epoch": 3.556322642663637,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003771645209336859,
      "loss": 5.4294,
      "mean_token_accuracy": 0.1878667026758194,
      "num_tokens": 82454442.0,
      "step": 40695
    },
    {
      "entropy": 5.8575855731964115,
      "epoch": 3.556759591016342,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003771367563880158,
      "loss": 5.3812,
      "mean_token_accuracy": 0.19504771381616592,
      "num_tokens": 82464589.0,
      "step": 40700
    },
    {
      "entropy": 5.806654548645019,
      "epoch": 3.5571965393690466,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037710898988330924,
      "loss": 5.3944,
      "mean_token_accuracy": 0.19478770792484285,
      "num_tokens": 82475624.0,
      "step": 40705
    },
    {
      "entropy": 5.831898641586304,
      "epoch": 3.557633487721751,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003770812214200987,
      "loss": 5.3731,
      "mean_token_accuracy": 0.19927532225847244,
      "num_tokens": 82485142.0,
      "step": 40710
    },
    {
      "entropy": 5.8681361198425295,
      "epoch": 3.558070436074456,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037705345099891684,
      "loss": 5.3932,
      "mean_token_accuracy": 0.1897299364209175,
      "num_tokens": 82495165.0,
      "step": 40715
    },
    {
      "entropy": 5.722726345062256,
      "epoch": 3.558507384427161,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003770256786202964,
      "loss": 5.277,
      "mean_token_accuracy": 0.1936880812048912,
      "num_tokens": 82505867.0,
      "step": 40720
    },
    {
      "entropy": 5.843424558639526,
      "epoch": 3.5589443327798653,
      "grad_norm": 0.9375,
      "learning_rate": 0.00037699790428476996,
      "loss": 5.4406,
      "mean_token_accuracy": 0.1904527261853218,
      "num_tokens": 82517416.0,
      "step": 40725
    },
    {
      "entropy": 5.866220617294312,
      "epoch": 3.5593812811325702,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003769701279928702,
      "loss": 5.3541,
      "mean_token_accuracy": 0.19927685111761093,
      "num_tokens": 82527192.0,
      "step": 40730
    },
    {
      "entropy": 5.801815938949585,
      "epoch": 3.5598182294852747,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037694234974512995,
      "loss": 5.3542,
      "mean_token_accuracy": 0.1998138651251793,
      "num_tokens": 82536600.0,
      "step": 40735
    },
    {
      "entropy": 5.824660682678223,
      "epoch": 3.5602551778379796,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037691456954208207,
      "loss": 5.4223,
      "mean_token_accuracy": 0.19810859262943267,
      "num_tokens": 82546899.0,
      "step": 40740
    },
    {
      "entropy": 5.806639623641968,
      "epoch": 3.560692126190684,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003768867873842593,
      "loss": 5.3903,
      "mean_token_accuracy": 0.19498620331287383,
      "num_tokens": 82557884.0,
      "step": 40745
    },
    {
      "entropy": 5.821110105514526,
      "epoch": 3.561129074543389,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003768590032721946,
      "loss": 5.2521,
      "mean_token_accuracy": 0.20320090502500535,
      "num_tokens": 82567726.0,
      "step": 40750
    },
    {
      "entropy": 5.833733558654785,
      "epoch": 3.561566022896094,
      "grad_norm": 1.0,
      "learning_rate": 0.00037683121720642087,
      "loss": 5.423,
      "mean_token_accuracy": 0.1929179161787033,
      "num_tokens": 82577741.0,
      "step": 40755
    },
    {
      "entropy": 5.8415467739105225,
      "epoch": 3.5620029712487984,
      "grad_norm": 1.0,
      "learning_rate": 0.00037680342918747103,
      "loss": 5.4226,
      "mean_token_accuracy": 0.18461191803216934,
      "num_tokens": 82588591.0,
      "step": 40760
    },
    {
      "entropy": 5.770736455917358,
      "epoch": 3.562439919601503,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000376775639215878,
      "loss": 5.3142,
      "mean_token_accuracy": 0.19577400684356688,
      "num_tokens": 82598276.0,
      "step": 40765
    },
    {
      "entropy": 5.732030916213989,
      "epoch": 3.5628768679542078,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037674784729217493,
      "loss": 5.3349,
      "mean_token_accuracy": 0.19901566505432128,
      "num_tokens": 82608715.0,
      "step": 40770
    },
    {
      "entropy": 5.803441953659058,
      "epoch": 3.5633138163069127,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037672005341689486,
      "loss": 5.3985,
      "mean_token_accuracy": 0.1960705727338791,
      "num_tokens": 82618854.0,
      "step": 40775
    },
    {
      "entropy": 5.8119621753692625,
      "epoch": 3.563750764659617,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003766922575905708,
      "loss": 5.3128,
      "mean_token_accuracy": 0.20170230269432068,
      "num_tokens": 82628391.0,
      "step": 40780
    },
    {
      "entropy": 5.839096546173096,
      "epoch": 3.564187713012322,
      "grad_norm": 1.046875,
      "learning_rate": 0.000376664459813736,
      "loss": 5.4577,
      "mean_token_accuracy": 0.1864958092570305,
      "num_tokens": 82639560.0,
      "step": 40785
    },
    {
      "entropy": 5.749555492401123,
      "epoch": 3.5646246613650265,
      "grad_norm": 1.0,
      "learning_rate": 0.00037663666008692356,
      "loss": 5.3275,
      "mean_token_accuracy": 0.19880484491586686,
      "num_tokens": 82650252.0,
      "step": 40790
    },
    {
      "entropy": 5.882618141174317,
      "epoch": 3.5650616097177315,
      "grad_norm": 0.921875,
      "learning_rate": 0.0003766088584106667,
      "loss": 5.3761,
      "mean_token_accuracy": 0.19570626616477965,
      "num_tokens": 82660035.0,
      "step": 40795
    },
    {
      "entropy": 5.739819860458374,
      "epoch": 3.565498558070436,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037658105478549875,
      "loss": 5.3101,
      "mean_token_accuracy": 0.19799587428569793,
      "num_tokens": 82669435.0,
      "step": 40800
    },
    {
      "entropy": 5.800925827026367,
      "epoch": 3.565935506423141,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037655324921195295,
      "loss": 5.3666,
      "mean_token_accuracy": 0.19507166296243666,
      "num_tokens": 82679314.0,
      "step": 40805
    },
    {
      "entropy": 5.747103548049926,
      "epoch": 3.5663724547758457,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003765254416905626,
      "loss": 5.3091,
      "mean_token_accuracy": 0.20438383817672728,
      "num_tokens": 82690140.0,
      "step": 40810
    },
    {
      "entropy": 5.786478710174561,
      "epoch": 3.56680940312855,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000376497632221861,
      "loss": 5.3523,
      "mean_token_accuracy": 0.19535846561193465,
      "num_tokens": 82699210.0,
      "step": 40815
    },
    {
      "entropy": 5.892255592346191,
      "epoch": 3.5672463514812547,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003764698208063818,
      "loss": 5.4795,
      "mean_token_accuracy": 0.18339427560567856,
      "num_tokens": 82710398.0,
      "step": 40820
    },
    {
      "entropy": 5.898912477493286,
      "epoch": 3.5676832998339596,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003764420074446581,
      "loss": 5.3522,
      "mean_token_accuracy": 0.1963021293282509,
      "num_tokens": 82721036.0,
      "step": 40825
    },
    {
      "entropy": 5.799274253845215,
      "epoch": 3.5681202481866645,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003764141921372236,
      "loss": 5.402,
      "mean_token_accuracy": 0.1967087060213089,
      "num_tokens": 82730884.0,
      "step": 40830
    },
    {
      "entropy": 5.7795415878295895,
      "epoch": 3.568557196539369,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003763863748846118,
      "loss": 5.3296,
      "mean_token_accuracy": 0.19889552444219588,
      "num_tokens": 82740561.0,
      "step": 40835
    },
    {
      "entropy": 5.8547478199005125,
      "epoch": 3.568994144892074,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037635855568735613,
      "loss": 5.4362,
      "mean_token_accuracy": 0.19131795018911363,
      "num_tokens": 82751064.0,
      "step": 40840
    },
    {
      "entropy": 5.816405010223389,
      "epoch": 3.5694310932447784,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003763307345459903,
      "loss": 5.2986,
      "mean_token_accuracy": 0.1986121878027916,
      "num_tokens": 82760502.0,
      "step": 40845
    },
    {
      "entropy": 5.738469505310059,
      "epoch": 3.5698680415974833,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003763029114610479,
      "loss": 5.3805,
      "mean_token_accuracy": 0.19146148711442948,
      "num_tokens": 82771477.0,
      "step": 40850
    },
    {
      "entropy": 5.7646812915802,
      "epoch": 3.5703049899501877,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003762750864330625,
      "loss": 5.3179,
      "mean_token_accuracy": 0.19842873364686966,
      "num_tokens": 82781572.0,
      "step": 40855
    },
    {
      "entropy": 5.8752227306365965,
      "epoch": 3.5707419383028927,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037624725946256793,
      "loss": 5.2985,
      "mean_token_accuracy": 0.194462351500988,
      "num_tokens": 82790217.0,
      "step": 40860
    },
    {
      "entropy": 5.82034330368042,
      "epoch": 3.5711788866555976,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00037621943055009786,
      "loss": 5.3671,
      "mean_token_accuracy": 0.19679739475250244,
      "num_tokens": 82800566.0,
      "step": 40865
    },
    {
      "entropy": 5.791106796264648,
      "epoch": 3.571615835008302,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003761915996961861,
      "loss": 5.3771,
      "mean_token_accuracy": 0.19584673941135405,
      "num_tokens": 82810188.0,
      "step": 40870
    },
    {
      "entropy": 5.817187547683716,
      "epoch": 3.5720527833610065,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037616376690136646,
      "loss": 5.3724,
      "mean_token_accuracy": 0.19182460755109787,
      "num_tokens": 82819509.0,
      "step": 40875
    },
    {
      "entropy": 5.8602348327636715,
      "epoch": 3.5724897317137114,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003761359321661727,
      "loss": 5.3633,
      "mean_token_accuracy": 0.1867164507508278,
      "num_tokens": 82830172.0,
      "step": 40880
    },
    {
      "entropy": 5.8266129970550535,
      "epoch": 3.5729266800664163,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037610809549113876,
      "loss": 5.3489,
      "mean_token_accuracy": 0.19689034968614577,
      "num_tokens": 82839245.0,
      "step": 40885
    },
    {
      "entropy": 5.78306303024292,
      "epoch": 3.573363628419121,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003760802568767986,
      "loss": 5.4104,
      "mean_token_accuracy": 0.18780374825000762,
      "num_tokens": 82850479.0,
      "step": 40890
    },
    {
      "entropy": 5.830339241027832,
      "epoch": 3.5738005767718257,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003760524163236862,
      "loss": 5.3605,
      "mean_token_accuracy": 0.19207275658845901,
      "num_tokens": 82859816.0,
      "step": 40895
    },
    {
      "entropy": 5.770054912567138,
      "epoch": 3.57423752512453,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003760245738323354,
      "loss": 5.3865,
      "mean_token_accuracy": 0.1897186443209648,
      "num_tokens": 82870124.0,
      "step": 40900
    },
    {
      "entropy": 5.817461109161377,
      "epoch": 3.574674473477235,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003759967294032803,
      "loss": 5.3593,
      "mean_token_accuracy": 0.19298612624406813,
      "num_tokens": 82880721.0,
      "step": 40905
    },
    {
      "entropy": 5.857951879501343,
      "epoch": 3.5751114218299396,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003759688830370551,
      "loss": 5.3942,
      "mean_token_accuracy": 0.19100387394428253,
      "num_tokens": 82890580.0,
      "step": 40910
    },
    {
      "entropy": 5.824918174743653,
      "epoch": 3.5755483701826445,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037594103473419384,
      "loss": 5.5163,
      "mean_token_accuracy": 0.18248146027326584,
      "num_tokens": 82902465.0,
      "step": 40915
    },
    {
      "entropy": 5.8276807308197025,
      "epoch": 3.575985318535349,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037591318449523063,
      "loss": 5.3211,
      "mean_token_accuracy": 0.19813536256551742,
      "num_tokens": 82912665.0,
      "step": 40920
    },
    {
      "entropy": 5.7767815589904785,
      "epoch": 3.576422266888054,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037588533232069967,
      "loss": 5.2646,
      "mean_token_accuracy": 0.19791983366012572,
      "num_tokens": 82921727.0,
      "step": 40925
    },
    {
      "entropy": 5.8361859798431395,
      "epoch": 3.5768592152407583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003758574782111351,
      "loss": 5.3394,
      "mean_token_accuracy": 0.1952318549156189,
      "num_tokens": 82932449.0,
      "step": 40930
    },
    {
      "entropy": 5.776624488830566,
      "epoch": 3.5772961635934633,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003758296221670713,
      "loss": 5.2537,
      "mean_token_accuracy": 0.2047355592250824,
      "num_tokens": 82941939.0,
      "step": 40935
    },
    {
      "entropy": 5.614080858230591,
      "epoch": 3.577733111946168,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003758017641890425,
      "loss": 5.274,
      "mean_token_accuracy": 0.20769727230072021,
      "num_tokens": 82951958.0,
      "step": 40940
    },
    {
      "entropy": 5.6721189498901365,
      "epoch": 3.5781700602988726,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003757739042775831,
      "loss": 5.3433,
      "mean_token_accuracy": 0.21333574652671813,
      "num_tokens": 82963014.0,
      "step": 40945
    },
    {
      "entropy": 5.804476022720337,
      "epoch": 3.5786070086515775,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003757460424332273,
      "loss": 5.3562,
      "mean_token_accuracy": 0.19801393002271653,
      "num_tokens": 82973081.0,
      "step": 40950
    },
    {
      "entropy": 5.839094829559326,
      "epoch": 3.579043957004282,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003757181786565097,
      "loss": 5.3592,
      "mean_token_accuracy": 0.19412908405065538,
      "num_tokens": 82982835.0,
      "step": 40955
    },
    {
      "entropy": 5.798557376861572,
      "epoch": 3.579480905356987,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003756903129479646,
      "loss": 5.3347,
      "mean_token_accuracy": 0.19634436070919037,
      "num_tokens": 82993275.0,
      "step": 40960
    },
    {
      "entropy": 5.834335613250732,
      "epoch": 3.5799178537096914,
      "grad_norm": 1.125,
      "learning_rate": 0.0003756624453081266,
      "loss": 5.42,
      "mean_token_accuracy": 0.19308097809553146,
      "num_tokens": 83003844.0,
      "step": 40965
    },
    {
      "entropy": 5.853808641433716,
      "epoch": 3.5803548020623963,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003756345757375302,
      "loss": 5.3893,
      "mean_token_accuracy": 0.19124084264039992,
      "num_tokens": 83014248.0,
      "step": 40970
    },
    {
      "entropy": 5.762586355209351,
      "epoch": 3.580791750415101,
      "grad_norm": 1.0,
      "learning_rate": 0.0003756067042367098,
      "loss": 5.2945,
      "mean_token_accuracy": 0.19865170419216155,
      "num_tokens": 83023967.0,
      "step": 40975
    },
    {
      "entropy": 5.836351156234741,
      "epoch": 3.5812286987678057,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003755788308062001,
      "loss": 5.4136,
      "mean_token_accuracy": 0.1896498903632164,
      "num_tokens": 83034167.0,
      "step": 40980
    },
    {
      "entropy": 5.749324464797974,
      "epoch": 3.58166564712051,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037555095544653574,
      "loss": 5.3419,
      "mean_token_accuracy": 0.19685045331716539,
      "num_tokens": 83043822.0,
      "step": 40985
    },
    {
      "entropy": 5.740227127075196,
      "epoch": 3.582102595473215,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037552307815825136,
      "loss": 5.369,
      "mean_token_accuracy": 0.20213617831468583,
      "num_tokens": 83054152.0,
      "step": 40990
    },
    {
      "entropy": 5.839321327209473,
      "epoch": 3.58253954382592,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00037549519894188174,
      "loss": 5.3246,
      "mean_token_accuracy": 0.197723850607872,
      "num_tokens": 83064553.0,
      "step": 40995
    },
    {
      "entropy": 5.7577540397644045,
      "epoch": 3.5829764921786245,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003754673177979614,
      "loss": 5.3543,
      "mean_token_accuracy": 0.2002020686864853,
      "num_tokens": 83073423.0,
      "step": 41000
    },
    {
      "entropy": 5.7863822937011715,
      "epoch": 3.583413440531329,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00037543943472702536,
      "loss": 5.3516,
      "mean_token_accuracy": 0.19783512353897095,
      "num_tokens": 83083774.0,
      "step": 41005
    },
    {
      "entropy": 5.733706140518189,
      "epoch": 3.583850388884034,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003754115497296084,
      "loss": 5.2518,
      "mean_token_accuracy": 0.2049364909529686,
      "num_tokens": 83092647.0,
      "step": 41010
    },
    {
      "entropy": 5.8353699207305905,
      "epoch": 3.5842873372367388,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037538366280624513,
      "loss": 5.5115,
      "mean_token_accuracy": 0.19016707688570023,
      "num_tokens": 83104070.0,
      "step": 41015
    },
    {
      "entropy": 5.817053127288818,
      "epoch": 3.5847242855894432,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003753557739574707,
      "loss": 5.319,
      "mean_token_accuracy": 0.20309409201145173,
      "num_tokens": 83113747.0,
      "step": 41020
    },
    {
      "entropy": 5.828705215454102,
      "epoch": 3.585161233942148,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037532788318382,
      "loss": 5.403,
      "mean_token_accuracy": 0.18746046721935272,
      "num_tokens": 83123678.0,
      "step": 41025
    },
    {
      "entropy": 5.870749998092651,
      "epoch": 3.5855981822948526,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003752999904858278,
      "loss": 5.3751,
      "mean_token_accuracy": 0.18988894373178483,
      "num_tokens": 83133232.0,
      "step": 41030
    },
    {
      "entropy": 5.871243286132812,
      "epoch": 3.5860351306475575,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037527209586402934,
      "loss": 5.3953,
      "mean_token_accuracy": 0.18845162391662598,
      "num_tokens": 83144351.0,
      "step": 41035
    },
    {
      "entropy": 5.80802731513977,
      "epoch": 3.586472079000262,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003752441993189595,
      "loss": 5.3811,
      "mean_token_accuracy": 0.19189390242099763,
      "num_tokens": 83155116.0,
      "step": 41040
    },
    {
      "entropy": 5.788115739822388,
      "epoch": 3.586909027352967,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003752163008511534,
      "loss": 5.4104,
      "mean_token_accuracy": 0.1897188350558281,
      "num_tokens": 83165227.0,
      "step": 41045
    },
    {
      "entropy": 5.86965856552124,
      "epoch": 3.587345975705672,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003751884004611461,
      "loss": 5.3994,
      "mean_token_accuracy": 0.19468041211366655,
      "num_tokens": 83175235.0,
      "step": 41050
    },
    {
      "entropy": 5.823133754730224,
      "epoch": 3.5877829240583763,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037516049814947277,
      "loss": 5.3463,
      "mean_token_accuracy": 0.19385128021240233,
      "num_tokens": 83185373.0,
      "step": 41055
    },
    {
      "entropy": 5.888617467880249,
      "epoch": 3.5882198724110808,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003751325939166687,
      "loss": 5.3977,
      "mean_token_accuracy": 0.18976622372865676,
      "num_tokens": 83195046.0,
      "step": 41060
    },
    {
      "entropy": 5.8028326511383055,
      "epoch": 3.5886568207637857,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037510468776326894,
      "loss": 5.3684,
      "mean_token_accuracy": 0.19926941245794297,
      "num_tokens": 83203817.0,
      "step": 41065
    },
    {
      "entropy": 5.867173051834106,
      "epoch": 3.5890937691164906,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003750767796898088,
      "loss": 5.4365,
      "mean_token_accuracy": 0.19192807972431183,
      "num_tokens": 83215096.0,
      "step": 41070
    },
    {
      "entropy": 5.81287841796875,
      "epoch": 3.589530717469195,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003750488696968236,
      "loss": 5.4437,
      "mean_token_accuracy": 0.19473422020673753,
      "num_tokens": 83225547.0,
      "step": 41075
    },
    {
      "entropy": 5.8396632194519045,
      "epoch": 3.5899676658219,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037502095778484866,
      "loss": 5.3265,
      "mean_token_accuracy": 0.1990486353635788,
      "num_tokens": 83235658.0,
      "step": 41080
    },
    {
      "entropy": 5.788891220092774,
      "epoch": 3.5904046141746044,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037499304395441943,
      "loss": 5.3372,
      "mean_token_accuracy": 0.192054583132267,
      "num_tokens": 83244819.0,
      "step": 41085
    },
    {
      "entropy": 5.707940864562988,
      "epoch": 3.5908415625273093,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003749651282060711,
      "loss": 5.2674,
      "mean_token_accuracy": 0.21149398237466813,
      "num_tokens": 83255729.0,
      "step": 41090
    },
    {
      "entropy": 5.816743087768555,
      "epoch": 3.591278510880014,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003749372105403392,
      "loss": 5.3815,
      "mean_token_accuracy": 0.19428260773420333,
      "num_tokens": 83265687.0,
      "step": 41095
    },
    {
      "entropy": 5.778873252868652,
      "epoch": 3.5917154592327187,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037490929095775934,
      "loss": 5.356,
      "mean_token_accuracy": 0.1955373004078865,
      "num_tokens": 83275558.0,
      "step": 41100
    },
    {
      "entropy": 5.796136093139649,
      "epoch": 3.5921524075854236,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037488136945886685,
      "loss": 5.3201,
      "mean_token_accuracy": 0.19885793030261995,
      "num_tokens": 83284744.0,
      "step": 41105
    },
    {
      "entropy": 5.778648614883423,
      "epoch": 3.592589355938128,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003748534460441975,
      "loss": 5.353,
      "mean_token_accuracy": 0.19549352079629898,
      "num_tokens": 83294372.0,
      "step": 41110
    },
    {
      "entropy": 5.827516078948975,
      "epoch": 3.5930263042908326,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037482552071428654,
      "loss": 5.3285,
      "mean_token_accuracy": 0.192726069688797,
      "num_tokens": 83305378.0,
      "step": 41115
    },
    {
      "entropy": 5.850354385375977,
      "epoch": 3.5934632526435375,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037479759346966985,
      "loss": 5.4221,
      "mean_token_accuracy": 0.18797239512205124,
      "num_tokens": 83315073.0,
      "step": 41120
    },
    {
      "entropy": 5.75220365524292,
      "epoch": 3.5939002009962424,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037476966431088305,
      "loss": 5.2577,
      "mean_token_accuracy": 0.20608129352331161,
      "num_tokens": 83324940.0,
      "step": 41125
    },
    {
      "entropy": 5.8255411148071286,
      "epoch": 3.594337149348947,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003747417332384618,
      "loss": 5.4023,
      "mean_token_accuracy": 0.19315814226865768,
      "num_tokens": 83335678.0,
      "step": 41130
    },
    {
      "entropy": 5.745960092544555,
      "epoch": 3.594774097701652,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037471380025294186,
      "loss": 5.2883,
      "mean_token_accuracy": 0.20721774250268937,
      "num_tokens": 83345682.0,
      "step": 41135
    },
    {
      "entropy": 5.748078727722168,
      "epoch": 3.5952110460543563,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000374685865354859,
      "loss": 5.3314,
      "mean_token_accuracy": 0.19895336478948594,
      "num_tokens": 83354934.0,
      "step": 41140
    },
    {
      "entropy": 5.781281423568726,
      "epoch": 3.595647994407061,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000374657928544749,
      "loss": 5.4038,
      "mean_token_accuracy": 0.19229533672332763,
      "num_tokens": 83365647.0,
      "step": 41145
    },
    {
      "entropy": 5.802182149887085,
      "epoch": 3.5960849427597656,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003746299898231477,
      "loss": 5.3621,
      "mean_token_accuracy": 0.19412537515163422,
      "num_tokens": 83376040.0,
      "step": 41150
    },
    {
      "entropy": 5.840118932723999,
      "epoch": 3.5965218911124706,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003746020491905909,
      "loss": 5.3313,
      "mean_token_accuracy": 0.1995715767145157,
      "num_tokens": 83386097.0,
      "step": 41155
    },
    {
      "entropy": 5.8965699672698975,
      "epoch": 3.5969588394651755,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037457410664761475,
      "loss": 5.4725,
      "mean_token_accuracy": 0.18152642846107483,
      "num_tokens": 83395555.0,
      "step": 41160
    },
    {
      "entropy": 5.84152193069458,
      "epoch": 3.59739578781788,
      "grad_norm": 1.8359375,
      "learning_rate": 0.000374546162194755,
      "loss": 5.3399,
      "mean_token_accuracy": 0.19444015622138977,
      "num_tokens": 83406195.0,
      "step": 41165
    },
    {
      "entropy": 5.798697757720947,
      "epoch": 3.5978327361705844,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037451821583254764,
      "loss": 5.3121,
      "mean_token_accuracy": 0.20354668349027633,
      "num_tokens": 83415254.0,
      "step": 41170
    },
    {
      "entropy": 5.766844749450684,
      "epoch": 3.5982696845232893,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037449026756152876,
      "loss": 5.3189,
      "mean_token_accuracy": 0.19942709654569626,
      "num_tokens": 83425112.0,
      "step": 41175
    },
    {
      "entropy": 5.794387435913086,
      "epoch": 3.5987066328759942,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037446231738223453,
      "loss": 5.4399,
      "mean_token_accuracy": 0.1824342966079712,
      "num_tokens": 83436185.0,
      "step": 41180
    },
    {
      "entropy": 5.892853116989135,
      "epoch": 3.5991435812286987,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003744343652952008,
      "loss": 5.3757,
      "mean_token_accuracy": 0.18987187892198562,
      "num_tokens": 83446540.0,
      "step": 41185
    },
    {
      "entropy": 5.822557830810547,
      "epoch": 3.5995805295814036,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037440641130096386,
      "loss": 5.307,
      "mean_token_accuracy": 0.20080591589212418,
      "num_tokens": 83457247.0,
      "step": 41190
    },
    {
      "entropy": 5.818590688705444,
      "epoch": 3.600017477934108,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003743784554000599,
      "loss": 5.371,
      "mean_token_accuracy": 0.1980339467525482,
      "num_tokens": 83468132.0,
      "step": 41195
    },
    {
      "entropy": 5.761891078948975,
      "epoch": 3.600454426286813,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003743504975930252,
      "loss": 5.3176,
      "mean_token_accuracy": 0.20129663348197938,
      "num_tokens": 83477824.0,
      "step": 41200
    },
    {
      "entropy": 5.820171260833741,
      "epoch": 3.6008913746395175,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003743225378803958,
      "loss": 5.3517,
      "mean_token_accuracy": 0.19204733669757842,
      "num_tokens": 83487352.0,
      "step": 41205
    },
    {
      "entropy": 5.775510501861572,
      "epoch": 3.6013283229922224,
      "grad_norm": 1.0,
      "learning_rate": 0.0003742945762627081,
      "loss": 5.3439,
      "mean_token_accuracy": 0.19850026816129684,
      "num_tokens": 83497957.0,
      "step": 41210
    },
    {
      "entropy": 5.787226819992066,
      "epoch": 3.6017652713449273,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003742666127404983,
      "loss": 5.2899,
      "mean_token_accuracy": 0.20154199600219727,
      "num_tokens": 83508213.0,
      "step": 41215
    },
    {
      "entropy": 5.794810390472412,
      "epoch": 3.6022022196976318,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003742386473143029,
      "loss": 5.3139,
      "mean_token_accuracy": 0.20602907687425615,
      "num_tokens": 83518694.0,
      "step": 41220
    },
    {
      "entropy": 5.772778940200806,
      "epoch": 3.6026391680503362,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037421067998465835,
      "loss": 5.3175,
      "mean_token_accuracy": 0.19861281663179398,
      "num_tokens": 83529077.0,
      "step": 41225
    },
    {
      "entropy": 5.746427869796753,
      "epoch": 3.603076116403041,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00037418271075210087,
      "loss": 5.3396,
      "mean_token_accuracy": 0.20041387975215913,
      "num_tokens": 83539502.0,
      "step": 41230
    },
    {
      "entropy": 5.78753662109375,
      "epoch": 3.603513064755746,
      "grad_norm": 1.109375,
      "learning_rate": 0.000374154739617167,
      "loss": 5.3099,
      "mean_token_accuracy": 0.20026807337999344,
      "num_tokens": 83549073.0,
      "step": 41235
    },
    {
      "entropy": 5.783473825454712,
      "epoch": 3.6039500131084505,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003741267665803934,
      "loss": 5.396,
      "mean_token_accuracy": 0.19571062326431274,
      "num_tokens": 83560404.0,
      "step": 41240
    },
    {
      "entropy": 5.766349029541016,
      "epoch": 3.6043869614611554,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003740987916423164,
      "loss": 5.2914,
      "mean_token_accuracy": 0.2015289232134819,
      "num_tokens": 83570421.0,
      "step": 41245
    },
    {
      "entropy": 5.823792314529419,
      "epoch": 3.60482390981386,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037407081480347263,
      "loss": 5.3816,
      "mean_token_accuracy": 0.19304774552583695,
      "num_tokens": 83581653.0,
      "step": 41250
    },
    {
      "entropy": 5.785693407058716,
      "epoch": 3.605260858166565,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003740428360643987,
      "loss": 5.3041,
      "mean_token_accuracy": 0.19534326195716858,
      "num_tokens": 83592190.0,
      "step": 41255
    },
    {
      "entropy": 5.814433813095093,
      "epoch": 3.6056978065192693,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003740148554256313,
      "loss": 5.3465,
      "mean_token_accuracy": 0.18950583189725875,
      "num_tokens": 83602495.0,
      "step": 41260
    },
    {
      "entropy": 5.815548658370972,
      "epoch": 3.606134754871974,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003739868728877071,
      "loss": 5.3393,
      "mean_token_accuracy": 0.20090124756097794,
      "num_tokens": 83612772.0,
      "step": 41265
    },
    {
      "entropy": 5.795007276535034,
      "epoch": 3.6065717032246787,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003739588884511628,
      "loss": 5.4096,
      "mean_token_accuracy": 0.19436945170164108,
      "num_tokens": 83623472.0,
      "step": 41270
    },
    {
      "entropy": 5.799902296066284,
      "epoch": 3.6070086515773836,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003739309021165352,
      "loss": 5.3289,
      "mean_token_accuracy": 0.1933260217308998,
      "num_tokens": 83633783.0,
      "step": 41275
    },
    {
      "entropy": 5.797708320617676,
      "epoch": 3.607445599930088,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037390291388436105,
      "loss": 5.3668,
      "mean_token_accuracy": 0.19536003321409226,
      "num_tokens": 83643594.0,
      "step": 41280
    },
    {
      "entropy": 5.773169231414795,
      "epoch": 3.607882548282793,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003738749237551772,
      "loss": 5.2368,
      "mean_token_accuracy": 0.20603047162294388,
      "num_tokens": 83653405.0,
      "step": 41285
    },
    {
      "entropy": 5.7371820449829105,
      "epoch": 3.608319496635498,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037384693172952047,
      "loss": 5.281,
      "mean_token_accuracy": 0.20062042474746705,
      "num_tokens": 83663233.0,
      "step": 41290
    },
    {
      "entropy": 5.800308179855347,
      "epoch": 3.6087564449882024,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003738189378079278,
      "loss": 5.3626,
      "mean_token_accuracy": 0.2003665030002594,
      "num_tokens": 83672899.0,
      "step": 41295
    },
    {
      "entropy": 5.835483455657959,
      "epoch": 3.609193393340907,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037379094199093603,
      "loss": 5.4225,
      "mean_token_accuracy": 0.19229428917169572,
      "num_tokens": 83683542.0,
      "step": 41300
    },
    {
      "entropy": 5.855385160446167,
      "epoch": 3.6096303416936117,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037376294427908237,
      "loss": 5.4286,
      "mean_token_accuracy": 0.1941140815615654,
      "num_tokens": 83693806.0,
      "step": 41305
    },
    {
      "entropy": 5.760533094406128,
      "epoch": 3.6100672900463167,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003737349446729036,
      "loss": 5.2893,
      "mean_token_accuracy": 0.19969107508659362,
      "num_tokens": 83704293.0,
      "step": 41310
    },
    {
      "entropy": 5.765359592437744,
      "epoch": 3.610504238399021,
      "grad_norm": 1.0,
      "learning_rate": 0.0003737069431729369,
      "loss": 5.2944,
      "mean_token_accuracy": 0.20256550312042237,
      "num_tokens": 83714321.0,
      "step": 41315
    },
    {
      "entropy": 5.712092685699463,
      "epoch": 3.610941186751726,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037367893977971933,
      "loss": 5.2643,
      "mean_token_accuracy": 0.204093836247921,
      "num_tokens": 83723815.0,
      "step": 41320
    },
    {
      "entropy": 5.791120815277099,
      "epoch": 3.6113781351044305,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037365093449378796,
      "loss": 5.3892,
      "mean_token_accuracy": 0.19286746382713318,
      "num_tokens": 83734036.0,
      "step": 41325
    },
    {
      "entropy": 5.886502313613891,
      "epoch": 3.6118150834571354,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037362292731567997,
      "loss": 5.3782,
      "mean_token_accuracy": 0.1873274102807045,
      "num_tokens": 83744343.0,
      "step": 41330
    },
    {
      "entropy": 5.894381809234619,
      "epoch": 3.61225203180984,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003735949182459325,
      "loss": 5.4038,
      "mean_token_accuracy": 0.19102613925933837,
      "num_tokens": 83754898.0,
      "step": 41335
    },
    {
      "entropy": 5.851542997360229,
      "epoch": 3.612688980162545,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003735669072850829,
      "loss": 5.3846,
      "mean_token_accuracy": 0.18903248012065887,
      "num_tokens": 83765964.0,
      "step": 41340
    },
    {
      "entropy": 5.812484931945801,
      "epoch": 3.6131259285152497,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037353889443366843,
      "loss": 5.3159,
      "mean_token_accuracy": 0.19368998259305953,
      "num_tokens": 83775920.0,
      "step": 41345
    },
    {
      "entropy": 5.797209310531616,
      "epoch": 3.613562876867954,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037351087969222626,
      "loss": 5.2646,
      "mean_token_accuracy": 0.20415127277374268,
      "num_tokens": 83784371.0,
      "step": 41350
    },
    {
      "entropy": 5.812798547744751,
      "epoch": 3.6139998252206587,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003734828630612938,
      "loss": 5.359,
      "mean_token_accuracy": 0.19315920770168304,
      "num_tokens": 83794618.0,
      "step": 41355
    },
    {
      "entropy": 5.79695496559143,
      "epoch": 3.6144367735733636,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003734548445414084,
      "loss": 5.3704,
      "mean_token_accuracy": 0.19639387130737304,
      "num_tokens": 83805070.0,
      "step": 41360
    },
    {
      "entropy": 5.801816463470459,
      "epoch": 3.6148737219260685,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037342682413310746,
      "loss": 5.4241,
      "mean_token_accuracy": 0.19009387493133545,
      "num_tokens": 83816257.0,
      "step": 41365
    },
    {
      "entropy": 5.790827417373658,
      "epoch": 3.615310670278773,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037339880183692855,
      "loss": 5.3632,
      "mean_token_accuracy": 0.18944750875234603,
      "num_tokens": 83826432.0,
      "step": 41370
    },
    {
      "entropy": 5.849508047103882,
      "epoch": 3.615747618631478,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037337077765340905,
      "loss": 5.3716,
      "mean_token_accuracy": 0.19306470602750778,
      "num_tokens": 83836746.0,
      "step": 41375
    },
    {
      "entropy": 5.766957092285156,
      "epoch": 3.6161845669841823,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037334275158308643,
      "loss": 5.3009,
      "mean_token_accuracy": 0.20128383487462997,
      "num_tokens": 83845783.0,
      "step": 41380
    },
    {
      "entropy": 5.797545337677002,
      "epoch": 3.6166215153368872,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003733147236264983,
      "loss": 5.3244,
      "mean_token_accuracy": 0.1946003556251526,
      "num_tokens": 83856981.0,
      "step": 41385
    },
    {
      "entropy": 5.8384932518005375,
      "epoch": 3.6170584636895917,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037328669378418236,
      "loss": 5.3995,
      "mean_token_accuracy": 0.19462761282920837,
      "num_tokens": 83867873.0,
      "step": 41390
    },
    {
      "entropy": 5.762905550003052,
      "epoch": 3.6174954120422966,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003732586620566761,
      "loss": 5.3261,
      "mean_token_accuracy": 0.19629526287317275,
      "num_tokens": 83877946.0,
      "step": 41395
    },
    {
      "entropy": 5.789417314529419,
      "epoch": 3.6179323603950015,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037323062844451713,
      "loss": 5.3362,
      "mean_token_accuracy": 0.19794358611106871,
      "num_tokens": 83888318.0,
      "step": 41400
    },
    {
      "entropy": 5.7423430442810055,
      "epoch": 3.618369308747706,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003732025929482433,
      "loss": 5.298,
      "mean_token_accuracy": 0.19858378767967225,
      "num_tokens": 83899092.0,
      "step": 41405
    },
    {
      "entropy": 5.760402965545654,
      "epoch": 3.6188062571004105,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003731745555683922,
      "loss": 5.3,
      "mean_token_accuracy": 0.2027610346674919,
      "num_tokens": 83908257.0,
      "step": 41410
    },
    {
      "entropy": 5.776785278320313,
      "epoch": 3.6192432054531154,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037314651630550186,
      "loss": 5.3791,
      "mean_token_accuracy": 0.19659262299537658,
      "num_tokens": 83918821.0,
      "step": 41415
    },
    {
      "entropy": 5.781262588500977,
      "epoch": 3.6196801538058203,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037311847516010975,
      "loss": 5.3054,
      "mean_token_accuracy": 0.20115043371915817,
      "num_tokens": 83928122.0,
      "step": 41420
    },
    {
      "entropy": 5.822286558151245,
      "epoch": 3.620117102158525,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003730904321327539,
      "loss": 5.4055,
      "mean_token_accuracy": 0.18871445804834366,
      "num_tokens": 83939904.0,
      "step": 41425
    },
    {
      "entropy": 5.840209913253784,
      "epoch": 3.6205540505112297,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003730623872239722,
      "loss": 5.3346,
      "mean_token_accuracy": 0.1974067896604538,
      "num_tokens": 83950631.0,
      "step": 41430
    },
    {
      "entropy": 5.840298271179199,
      "epoch": 3.620990998863934,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003730343404343025,
      "loss": 5.4219,
      "mean_token_accuracy": 0.1927458629012108,
      "num_tokens": 83960860.0,
      "step": 41435
    },
    {
      "entropy": 5.8428840160369875,
      "epoch": 3.621427947216639,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003730062917642828,
      "loss": 5.4273,
      "mean_token_accuracy": 0.18887946158647537,
      "num_tokens": 83971182.0,
      "step": 41440
    },
    {
      "entropy": 5.747141408920288,
      "epoch": 3.6218648955693435,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003729782412144511,
      "loss": 5.2744,
      "mean_token_accuracy": 0.20351493507623672,
      "num_tokens": 83981576.0,
      "step": 41445
    },
    {
      "entropy": 5.825237560272217,
      "epoch": 3.6223018439220485,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003729501887853453,
      "loss": 5.3328,
      "mean_token_accuracy": 0.1990129381418228,
      "num_tokens": 83991526.0,
      "step": 41450
    },
    {
      "entropy": 5.837203311920166,
      "epoch": 3.6227387922747534,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003729221344775036,
      "loss": 5.3823,
      "mean_token_accuracy": 0.19480076283216477,
      "num_tokens": 84001979.0,
      "step": 41455
    },
    {
      "entropy": 5.709698009490967,
      "epoch": 3.623175740627458,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003728940782914641,
      "loss": 5.2418,
      "mean_token_accuracy": 0.20314965695142745,
      "num_tokens": 84011977.0,
      "step": 41460
    },
    {
      "entropy": 5.801184463500976,
      "epoch": 3.6236126889801623,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037286602022776485,
      "loss": 5.3571,
      "mean_token_accuracy": 0.19441206157207488,
      "num_tokens": 84021826.0,
      "step": 41465
    },
    {
      "entropy": 5.773125457763672,
      "epoch": 3.6240496373328672,
      "grad_norm": 0.984375,
      "learning_rate": 0.00037283796028694403,
      "loss": 5.267,
      "mean_token_accuracy": 0.2026190772652626,
      "num_tokens": 84032180.0,
      "step": 41470
    },
    {
      "entropy": 5.842540502548218,
      "epoch": 3.624486585685572,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037280989846953986,
      "loss": 5.4332,
      "mean_token_accuracy": 0.18773707151412963,
      "num_tokens": 84043698.0,
      "step": 41475
    },
    {
      "entropy": 5.779453229904175,
      "epoch": 3.6249235340382766,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003727818347760906,
      "loss": 5.2834,
      "mean_token_accuracy": 0.20408952087163926,
      "num_tokens": 84053323.0,
      "step": 41480
    },
    {
      "entropy": 5.885993576049804,
      "epoch": 3.6253604823909815,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037275376920713454,
      "loss": 5.4865,
      "mean_token_accuracy": 0.18732829242944718,
      "num_tokens": 84064092.0,
      "step": 41485
    },
    {
      "entropy": 5.82270884513855,
      "epoch": 3.625797430743686,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037272570176320995,
      "loss": 5.3571,
      "mean_token_accuracy": 0.20206961184740066,
      "num_tokens": 84073661.0,
      "step": 41490
    },
    {
      "entropy": 5.737351942062378,
      "epoch": 3.626234379096391,
      "grad_norm": 1.0,
      "learning_rate": 0.0003726976324448552,
      "loss": 5.369,
      "mean_token_accuracy": 0.19839295893907546,
      "num_tokens": 84083199.0,
      "step": 41495
    },
    {
      "entropy": 5.84652304649353,
      "epoch": 3.6266713274490954,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003726695612526086,
      "loss": 5.3714,
      "mean_token_accuracy": 0.19686997979879378,
      "num_tokens": 84093600.0,
      "step": 41500
    },
    {
      "entropy": 5.857052993774414,
      "epoch": 3.6271082758018003,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003726414881870087,
      "loss": 5.3933,
      "mean_token_accuracy": 0.19224451780319213,
      "num_tokens": 84103147.0,
      "step": 41505
    },
    {
      "entropy": 5.70027642250061,
      "epoch": 3.627545224154505,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037261341324859386,
      "loss": 5.2872,
      "mean_token_accuracy": 0.20519742667675017,
      "num_tokens": 84113719.0,
      "step": 41510
    },
    {
      "entropy": 5.801758718490601,
      "epoch": 3.6279821725072097,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00037258533643790264,
      "loss": 5.302,
      "mean_token_accuracy": 0.1986786752939224,
      "num_tokens": 84123563.0,
      "step": 41515
    },
    {
      "entropy": 5.81337685585022,
      "epoch": 3.628419120859914,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00037255725775547357,
      "loss": 5.4,
      "mean_token_accuracy": 0.19252602905035018,
      "num_tokens": 84134753.0,
      "step": 41520
    },
    {
      "entropy": 5.852410221099854,
      "epoch": 3.628856069212619,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037252917720184507,
      "loss": 5.3866,
      "mean_token_accuracy": 0.1848103731870651,
      "num_tokens": 84146549.0,
      "step": 41525
    },
    {
      "entropy": 5.8929705142974855,
      "epoch": 3.629293017565324,
      "grad_norm": 1.125,
      "learning_rate": 0.0003725010947775559,
      "loss": 5.4337,
      "mean_token_accuracy": 0.1938855990767479,
      "num_tokens": 84157069.0,
      "step": 41530
    },
    {
      "entropy": 5.77478551864624,
      "epoch": 3.6297299659180284,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037247301048314464,
      "loss": 5.3831,
      "mean_token_accuracy": 0.19540640562772751,
      "num_tokens": 84167516.0,
      "step": 41535
    },
    {
      "entropy": 5.796593141555786,
      "epoch": 3.6301669142707333,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037244492431914996,
      "loss": 5.3597,
      "mean_token_accuracy": 0.1973887249827385,
      "num_tokens": 84178539.0,
      "step": 41540
    },
    {
      "entropy": 5.901836347579956,
      "epoch": 3.630603862623438,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00037241683628611055,
      "loss": 5.3344,
      "mean_token_accuracy": 0.20174923092126845,
      "num_tokens": 84188605.0,
      "step": 41545
    },
    {
      "entropy": 5.780180883407593,
      "epoch": 3.6310408109761427,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037238874638456524,
      "loss": 5.2983,
      "mean_token_accuracy": 0.203644959628582,
      "num_tokens": 84197478.0,
      "step": 41550
    },
    {
      "entropy": 5.741819429397583,
      "epoch": 3.631477759328847,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003723606546150527,
      "loss": 5.3885,
      "mean_token_accuracy": 0.1909459948539734,
      "num_tokens": 84207455.0,
      "step": 41555
    },
    {
      "entropy": 5.769747972488403,
      "epoch": 3.631914707681552,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003723325609781118,
      "loss": 5.2439,
      "mean_token_accuracy": 0.209821455180645,
      "num_tokens": 84217336.0,
      "step": 41560
    },
    {
      "entropy": 5.801977729797363,
      "epoch": 3.632351656034257,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037230446547428136,
      "loss": 5.3193,
      "mean_token_accuracy": 0.19609753042459488,
      "num_tokens": 84227474.0,
      "step": 41565
    },
    {
      "entropy": 5.776434516906738,
      "epoch": 3.6327886043869615,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003722763681041003,
      "loss": 5.3731,
      "mean_token_accuracy": 0.19377435594797135,
      "num_tokens": 84237382.0,
      "step": 41570
    },
    {
      "entropy": 5.823347377777099,
      "epoch": 3.633225552739666,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003722482688681075,
      "loss": 5.3864,
      "mean_token_accuracy": 0.18710898011922836,
      "num_tokens": 84247816.0,
      "step": 41575
    },
    {
      "entropy": 5.843149471282959,
      "epoch": 3.633662501092371,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000372220167766842,
      "loss": 5.3262,
      "mean_token_accuracy": 0.19602570682764053,
      "num_tokens": 84257056.0,
      "step": 41580
    },
    {
      "entropy": 5.894835615158081,
      "epoch": 3.634099449445076,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037219206480084273,
      "loss": 5.4661,
      "mean_token_accuracy": 0.1917251631617546,
      "num_tokens": 84267499.0,
      "step": 41585
    },
    {
      "entropy": 5.795135545730591,
      "epoch": 3.6345363977977803,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003721639599706487,
      "loss": 5.3004,
      "mean_token_accuracy": 0.20007604211568833,
      "num_tokens": 84277548.0,
      "step": 41590
    },
    {
      "entropy": 5.749506092071533,
      "epoch": 3.634973346150485,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037213585327679904,
      "loss": 5.2939,
      "mean_token_accuracy": 0.19687765687704087,
      "num_tokens": 84287461.0,
      "step": 41595
    },
    {
      "entropy": 5.812924242019653,
      "epoch": 3.6354102945031896,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037210774471983283,
      "loss": 5.313,
      "mean_token_accuracy": 0.19585035294294356,
      "num_tokens": 84297878.0,
      "step": 41600
    },
    {
      "entropy": 5.757561349868775,
      "epoch": 3.6358472428558946,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037207963430028915,
      "loss": 5.2703,
      "mean_token_accuracy": 0.2049698829650879,
      "num_tokens": 84307850.0,
      "step": 41605
    },
    {
      "entropy": 5.812178897857666,
      "epoch": 3.636284191208599,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003720515220187072,
      "loss": 5.3886,
      "mean_token_accuracy": 0.1914113536477089,
      "num_tokens": 84318111.0,
      "step": 41610
    },
    {
      "entropy": 5.713361692428589,
      "epoch": 3.636721139561304,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003720234078756262,
      "loss": 5.2103,
      "mean_token_accuracy": 0.20568044483661652,
      "num_tokens": 84327118.0,
      "step": 41615
    },
    {
      "entropy": 5.798464393615722,
      "epoch": 3.6371580879140084,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003719952918715854,
      "loss": 5.3857,
      "mean_token_accuracy": 0.19632071554660796,
      "num_tokens": 84338428.0,
      "step": 41620
    },
    {
      "entropy": 5.868563604354859,
      "epoch": 3.6375950362667133,
      "grad_norm": 0.984375,
      "learning_rate": 0.00037196717400712415,
      "loss": 5.4466,
      "mean_token_accuracy": 0.18657319545745848,
      "num_tokens": 84349117.0,
      "step": 41625
    },
    {
      "entropy": 5.855116605758667,
      "epoch": 3.638031984619418,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003719390542827816,
      "loss": 5.3962,
      "mean_token_accuracy": 0.19391593486070632,
      "num_tokens": 84358807.0,
      "step": 41630
    },
    {
      "entropy": 5.770904541015625,
      "epoch": 3.6384689329721227,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003719109326990973,
      "loss": 5.3559,
      "mean_token_accuracy": 0.19891505539417267,
      "num_tokens": 84368970.0,
      "step": 41635
    },
    {
      "entropy": 5.7702521800994875,
      "epoch": 3.6389058813248276,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037188280925661043,
      "loss": 5.2779,
      "mean_token_accuracy": 0.20978891253471374,
      "num_tokens": 84378475.0,
      "step": 41640
    },
    {
      "entropy": 5.784608793258667,
      "epoch": 3.639342829677532,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003718546839558606,
      "loss": 5.2802,
      "mean_token_accuracy": 0.20948150008916855,
      "num_tokens": 84387832.0,
      "step": 41645
    },
    {
      "entropy": 5.779055452346801,
      "epoch": 3.6397797780302366,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037182655679738713,
      "loss": 5.2881,
      "mean_token_accuracy": 0.20291053652763366,
      "num_tokens": 84398316.0,
      "step": 41650
    },
    {
      "entropy": 5.735073614120483,
      "epoch": 3.6402167263829415,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003717984277817296,
      "loss": 5.2545,
      "mean_token_accuracy": 0.19582271575927734,
      "num_tokens": 84408624.0,
      "step": 41655
    },
    {
      "entropy": 5.7958158493042,
      "epoch": 3.6406536747356464,
      "grad_norm": 0.96875,
      "learning_rate": 0.00037177029690942744,
      "loss": 5.2645,
      "mean_token_accuracy": 0.20450455099344253,
      "num_tokens": 84417422.0,
      "step": 41660
    },
    {
      "entropy": 5.790545320510864,
      "epoch": 3.641090623088351,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003717421641810202,
      "loss": 5.387,
      "mean_token_accuracy": 0.18857968002557754,
      "num_tokens": 84427857.0,
      "step": 41665
    },
    {
      "entropy": 5.892370080947876,
      "epoch": 3.6415275714410558,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037171402959704767,
      "loss": 5.3799,
      "mean_token_accuracy": 0.18536862432956697,
      "num_tokens": 84438468.0,
      "step": 41670
    },
    {
      "entropy": 5.794828033447265,
      "epoch": 3.6419645197937602,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037168589315804944,
      "loss": 5.3575,
      "mean_token_accuracy": 0.19685754477977752,
      "num_tokens": 84449371.0,
      "step": 41675
    },
    {
      "entropy": 5.7731770992279055,
      "epoch": 3.642401468146465,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003716577548645651,
      "loss": 5.3181,
      "mean_token_accuracy": 0.19486112147569656,
      "num_tokens": 84458569.0,
      "step": 41680
    },
    {
      "entropy": 5.795122671127319,
      "epoch": 3.6428384164991696,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003716296147171343,
      "loss": 5.4142,
      "mean_token_accuracy": 0.19798565655946732,
      "num_tokens": 84469244.0,
      "step": 41685
    },
    {
      "entropy": 5.73502631187439,
      "epoch": 3.6432753648518745,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003716014727162969,
      "loss": 5.3008,
      "mean_token_accuracy": 0.2147839531302452,
      "num_tokens": 84478902.0,
      "step": 41690
    },
    {
      "entropy": 5.714601469039917,
      "epoch": 3.6437123132045794,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037157332886259253,
      "loss": 5.2919,
      "mean_token_accuracy": 0.20083983540534972,
      "num_tokens": 84489166.0,
      "step": 41695
    },
    {
      "entropy": 5.789045953750611,
      "epoch": 3.644149261557284,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003715451831565613,
      "loss": 5.2869,
      "mean_token_accuracy": 0.19994858354330064,
      "num_tokens": 84499093.0,
      "step": 41700
    },
    {
      "entropy": 5.807829427719116,
      "epoch": 3.6445862099099884,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037151703559874273,
      "loss": 5.3025,
      "mean_token_accuracy": 0.19926489293575286,
      "num_tokens": 84508898.0,
      "step": 41705
    },
    {
      "entropy": 5.832468461990357,
      "epoch": 3.6450231582626933,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037148888618967677,
      "loss": 5.4058,
      "mean_token_accuracy": 0.19182953834533692,
      "num_tokens": 84518277.0,
      "step": 41710
    },
    {
      "entropy": 5.826121854782104,
      "epoch": 3.645460106615398,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003714607349299035,
      "loss": 5.3642,
      "mean_token_accuracy": 0.19068333506584167,
      "num_tokens": 84527882.0,
      "step": 41715
    },
    {
      "entropy": 5.804207801818848,
      "epoch": 3.6458970549681027,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037143258181996286,
      "loss": 5.3273,
      "mean_token_accuracy": 0.20188977122306823,
      "num_tokens": 84538065.0,
      "step": 41720
    },
    {
      "entropy": 5.806598234176636,
      "epoch": 3.6463340033208076,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003714044268603947,
      "loss": 5.336,
      "mean_token_accuracy": 0.1988108366727829,
      "num_tokens": 84547923.0,
      "step": 41725
    },
    {
      "entropy": 5.7683751583099365,
      "epoch": 3.646770951673512,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003713762700517392,
      "loss": 5.3451,
      "mean_token_accuracy": 0.19173361957073212,
      "num_tokens": 84558725.0,
      "step": 41730
    },
    {
      "entropy": 5.8183153629302975,
      "epoch": 3.647207900026217,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037134811139453625,
      "loss": 5.3397,
      "mean_token_accuracy": 0.20042741745710374,
      "num_tokens": 84568945.0,
      "step": 41735
    },
    {
      "entropy": 5.833605527877808,
      "epoch": 3.6476448483789214,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000371319950889326,
      "loss": 5.4122,
      "mean_token_accuracy": 0.19202979803085327,
      "num_tokens": 84579230.0,
      "step": 41740
    },
    {
      "entropy": 5.84570198059082,
      "epoch": 3.6480817967316264,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003712917885366487,
      "loss": 5.3447,
      "mean_token_accuracy": 0.19869971722364427,
      "num_tokens": 84588966.0,
      "step": 41745
    },
    {
      "entropy": 5.731508779525757,
      "epoch": 3.6485187450843313,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003712636243370444,
      "loss": 5.2823,
      "mean_token_accuracy": 0.19990544617176056,
      "num_tokens": 84598631.0,
      "step": 41750
    },
    {
      "entropy": 5.789768123626709,
      "epoch": 3.6489556934370357,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037123545829105336,
      "loss": 5.3383,
      "mean_token_accuracy": 0.19249698221683503,
      "num_tokens": 84609849.0,
      "step": 41755
    },
    {
      "entropy": 5.797710180282593,
      "epoch": 3.64939264178974,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003712072903992159,
      "loss": 5.3502,
      "mean_token_accuracy": 0.19587461203336715,
      "num_tokens": 84619959.0,
      "step": 41760
    },
    {
      "entropy": 5.845406818389892,
      "epoch": 3.649829590142445,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003711791206620721,
      "loss": 5.2926,
      "mean_token_accuracy": 0.20063208937644958,
      "num_tokens": 84630224.0,
      "step": 41765
    },
    {
      "entropy": 5.791042041778565,
      "epoch": 3.65026653849515,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003711509490801624,
      "loss": 5.3661,
      "mean_token_accuracy": 0.19632393568754197,
      "num_tokens": 84641595.0,
      "step": 41770
    },
    {
      "entropy": 5.721117925643921,
      "epoch": 3.6507034868478545,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000371122775654027,
      "loss": 5.2938,
      "mean_token_accuracy": 0.1999514415860176,
      "num_tokens": 84652406.0,
      "step": 41775
    },
    {
      "entropy": 5.764357233047486,
      "epoch": 3.6511404352005594,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003710946003842065,
      "loss": 5.2404,
      "mean_token_accuracy": 0.21350652426481248,
      "num_tokens": 84662319.0,
      "step": 41780
    },
    {
      "entropy": 5.817035102844239,
      "epoch": 3.651577383553264,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003710664232712412,
      "loss": 5.3468,
      "mean_token_accuracy": 0.19267132878303528,
      "num_tokens": 84673086.0,
      "step": 41785
    },
    {
      "entropy": 5.767650747299195,
      "epoch": 3.652014331905969,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003710382443156716,
      "loss": 5.4105,
      "mean_token_accuracy": 0.19405241161584855,
      "num_tokens": 84684536.0,
      "step": 41790
    },
    {
      "entropy": 5.788109493255615,
      "epoch": 3.6524512802586733,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003710100635180381,
      "loss": 5.3349,
      "mean_token_accuracy": 0.19879642724990845,
      "num_tokens": 84694975.0,
      "step": 41795
    },
    {
      "entropy": 5.736692667007446,
      "epoch": 3.652888228611378,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003709818808788813,
      "loss": 5.3132,
      "mean_token_accuracy": 0.19707499742507933,
      "num_tokens": 84704290.0,
      "step": 41800
    },
    {
      "entropy": 5.865963172912598,
      "epoch": 3.653325176964083,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003709536963987417,
      "loss": 5.332,
      "mean_token_accuracy": 0.19428955167531967,
      "num_tokens": 84714280.0,
      "step": 41805
    },
    {
      "entropy": 5.869581985473633,
      "epoch": 3.6537621253167876,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037092551007815997,
      "loss": 5.3854,
      "mean_token_accuracy": 0.19985494911670684,
      "num_tokens": 84725127.0,
      "step": 41810
    },
    {
      "entropy": 5.810454940795898,
      "epoch": 3.654199073669492,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003708973219176767,
      "loss": 5.3626,
      "mean_token_accuracy": 0.19266560971736907,
      "num_tokens": 84736385.0,
      "step": 41815
    },
    {
      "entropy": 5.772128963470459,
      "epoch": 3.654636022022197,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00037086913191783247,
      "loss": 5.2912,
      "mean_token_accuracy": 0.2120218649506569,
      "num_tokens": 84746489.0,
      "step": 41820
    },
    {
      "entropy": 5.788606214523315,
      "epoch": 3.655072970374902,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0003708409400791681,
      "loss": 5.343,
      "mean_token_accuracy": 0.19396435022354125,
      "num_tokens": 84756743.0,
      "step": 41825
    },
    {
      "entropy": 5.785589981079101,
      "epoch": 3.6555099187276063,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037081274640222425,
      "loss": 5.3361,
      "mean_token_accuracy": 0.19763322174549103,
      "num_tokens": 84767647.0,
      "step": 41830
    },
    {
      "entropy": 5.866511058807373,
      "epoch": 3.6559468670803112,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003707845508875417,
      "loss": 5.3892,
      "mean_token_accuracy": 0.18773798644542694,
      "num_tokens": 84778122.0,
      "step": 41835
    },
    {
      "entropy": 5.8472949981689455,
      "epoch": 3.6563838154330157,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003707563535356614,
      "loss": 5.381,
      "mean_token_accuracy": 0.20031924098730086,
      "num_tokens": 84787300.0,
      "step": 41840
    },
    {
      "entropy": 5.777621698379517,
      "epoch": 3.6568207637857206,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003707281543471239,
      "loss": 5.3534,
      "mean_token_accuracy": 0.19339047223329545,
      "num_tokens": 84798271.0,
      "step": 41845
    },
    {
      "entropy": 5.861323022842408,
      "epoch": 3.657257712138425,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003706999533224704,
      "loss": 5.4286,
      "mean_token_accuracy": 0.1897789254784584,
      "num_tokens": 84810377.0,
      "step": 41850
    },
    {
      "entropy": 5.838926458358765,
      "epoch": 3.65769466049113,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003706717504622415,
      "loss": 5.3192,
      "mean_token_accuracy": 0.1975908949971199,
      "num_tokens": 84821319.0,
      "step": 41855
    },
    {
      "entropy": 5.845651626586914,
      "epoch": 3.658131608843835,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003706435457669784,
      "loss": 5.3614,
      "mean_token_accuracy": 0.1927065759897232,
      "num_tokens": 84831558.0,
      "step": 41860
    },
    {
      "entropy": 5.808934688568115,
      "epoch": 3.6585685571965394,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037061533923722195,
      "loss": 5.3212,
      "mean_token_accuracy": 0.1968322515487671,
      "num_tokens": 84840983.0,
      "step": 41865
    },
    {
      "entropy": 5.699240446090698,
      "epoch": 3.659005505549244,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037058713087351317,
      "loss": 5.257,
      "mean_token_accuracy": 0.2088940143585205,
      "num_tokens": 84851304.0,
      "step": 41870
    },
    {
      "entropy": 5.706376218795777,
      "epoch": 3.6594424539019488,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037055892067639314,
      "loss": 5.3388,
      "mean_token_accuracy": 0.1972159191966057,
      "num_tokens": 84862849.0,
      "step": 41875
    },
    {
      "entropy": 5.812178277969361,
      "epoch": 3.6598794022546537,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037053070864640294,
      "loss": 5.2898,
      "mean_token_accuracy": 0.20249702781438828,
      "num_tokens": 84873591.0,
      "step": 41880
    },
    {
      "entropy": 5.817550706863403,
      "epoch": 3.660316350607358,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037050249478408364,
      "loss": 5.2927,
      "mean_token_accuracy": 0.20389047861099244,
      "num_tokens": 84883701.0,
      "step": 41885
    },
    {
      "entropy": 5.7669981002807615,
      "epoch": 3.660753298960063,
      "grad_norm": 1.0,
      "learning_rate": 0.0003704742790899765,
      "loss": 5.328,
      "mean_token_accuracy": 0.19241720587015151,
      "num_tokens": 84893842.0,
      "step": 41890
    },
    {
      "entropy": 5.841239166259766,
      "epoch": 3.6611902473127675,
      "grad_norm": 1.0,
      "learning_rate": 0.0003704460615646227,
      "loss": 5.3983,
      "mean_token_accuracy": 0.18803357779979707,
      "num_tokens": 84903797.0,
      "step": 41895
    },
    {
      "entropy": 5.837640333175659,
      "epoch": 3.6616271956654725,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037041784220856336,
      "loss": 5.4332,
      "mean_token_accuracy": 0.19758789241313934,
      "num_tokens": 84915459.0,
      "step": 41900
    },
    {
      "entropy": 5.793707370758057,
      "epoch": 3.662064144018177,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003703896210223399,
      "loss": 5.37,
      "mean_token_accuracy": 0.19661697894334793,
      "num_tokens": 84926067.0,
      "step": 41905
    },
    {
      "entropy": 5.874314641952514,
      "epoch": 3.662501092370882,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037036139800649336,
      "loss": 5.3469,
      "mean_token_accuracy": 0.1989818125963211,
      "num_tokens": 84936612.0,
      "step": 41910
    },
    {
      "entropy": 5.73343071937561,
      "epoch": 3.6629380407235863,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003703331731615654,
      "loss": 5.2609,
      "mean_token_accuracy": 0.20246677100658417,
      "num_tokens": 84946620.0,
      "step": 41915
    },
    {
      "entropy": 5.797532653808593,
      "epoch": 3.663374989076291,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037030494648809716,
      "loss": 5.3116,
      "mean_token_accuracy": 0.20083496123552322,
      "num_tokens": 84955890.0,
      "step": 41920
    },
    {
      "entropy": 5.649131393432617,
      "epoch": 3.6638119374289957,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003702767179866301,
      "loss": 5.2784,
      "mean_token_accuracy": 0.1961122140288353,
      "num_tokens": 84964694.0,
      "step": 41925
    },
    {
      "entropy": 5.760038661956787,
      "epoch": 3.6642488857817006,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003702484876577057,
      "loss": 5.313,
      "mean_token_accuracy": 0.20307784229516984,
      "num_tokens": 84974203.0,
      "step": 41930
    },
    {
      "entropy": 5.785346794128418,
      "epoch": 3.6646858341344055,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037022025550186537,
      "loss": 5.2576,
      "mean_token_accuracy": 0.2002711519598961,
      "num_tokens": 84984326.0,
      "step": 41935
    },
    {
      "entropy": 5.787569808959961,
      "epoch": 3.66512278248711,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003701920215196506,
      "loss": 5.2877,
      "mean_token_accuracy": 0.19810759425163268,
      "num_tokens": 84993284.0,
      "step": 41940
    },
    {
      "entropy": 5.765491914749146,
      "epoch": 3.665559730839815,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003701637857116031,
      "loss": 5.3111,
      "mean_token_accuracy": 0.20033330023288726,
      "num_tokens": 85003239.0,
      "step": 41945
    },
    {
      "entropy": 5.793357992172242,
      "epoch": 3.6659966791925194,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003701355480782642,
      "loss": 5.3286,
      "mean_token_accuracy": 0.19958533793687822,
      "num_tokens": 85015493.0,
      "step": 41950
    },
    {
      "entropy": 5.763089275360107,
      "epoch": 3.6664336275452243,
      "grad_norm": 0.96875,
      "learning_rate": 0.00037010730862017575,
      "loss": 5.3007,
      "mean_token_accuracy": 0.1988338425755501,
      "num_tokens": 85025677.0,
      "step": 41955
    },
    {
      "entropy": 5.766998672485352,
      "epoch": 3.6668705758979288,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037007906733787913,
      "loss": 5.3195,
      "mean_token_accuracy": 0.2009044259786606,
      "num_tokens": 85034908.0,
      "step": 41960
    },
    {
      "entropy": 5.8055376529693605,
      "epoch": 3.6673075242506337,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003700508242319163,
      "loss": 5.3679,
      "mean_token_accuracy": 0.19113693833351136,
      "num_tokens": 85045416.0,
      "step": 41965
    },
    {
      "entropy": 5.878468990325928,
      "epoch": 3.667744472603338,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037002257930282887,
      "loss": 5.4371,
      "mean_token_accuracy": 0.19002807289361953,
      "num_tokens": 85056259.0,
      "step": 41970
    },
    {
      "entropy": 5.817567777633667,
      "epoch": 3.668181420956043,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003699943325511585,
      "loss": 5.3201,
      "mean_token_accuracy": 0.2015848070383072,
      "num_tokens": 85065579.0,
      "step": 41975
    },
    {
      "entropy": 5.748672723770142,
      "epoch": 3.6686183693087475,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003699660839774471,
      "loss": 5.3366,
      "mean_token_accuracy": 0.19641423374414443,
      "num_tokens": 85075084.0,
      "step": 41980
    },
    {
      "entropy": 5.754591178894043,
      "epoch": 3.6690553176614524,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003699378335822364,
      "loss": 5.326,
      "mean_token_accuracy": 0.2050619378685951,
      "num_tokens": 85084536.0,
      "step": 41985
    },
    {
      "entropy": 5.821064615249634,
      "epoch": 3.6694922660141573,
      "grad_norm": 1.125,
      "learning_rate": 0.0003699095813660683,
      "loss": 5.3226,
      "mean_token_accuracy": 0.203002367913723,
      "num_tokens": 85094076.0,
      "step": 41990
    },
    {
      "entropy": 5.771641635894776,
      "epoch": 3.669929214366862,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003698813273294848,
      "loss": 5.2856,
      "mean_token_accuracy": 0.19610814452171327,
      "num_tokens": 85103784.0,
      "step": 41995
    },
    {
      "entropy": 5.755959987640381,
      "epoch": 3.6703661627195663,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036985307147302765,
      "loss": 5.3447,
      "mean_token_accuracy": 0.19964399933815002,
      "num_tokens": 85114066.0,
      "step": 42000
    },
    {
      "epoch": 3.6703661627195663,
      "eval_entropy": 5.525017110168329,
      "eval_loss": 5.404279708862305,
      "eval_mean_token_accuracy": 0.20188119776123462,
      "eval_num_tokens": 85114066.0,
      "eval_runtime": 28.3606,
      "eval_samples_per_second": 1305.121,
      "eval_steps_per_second": 163.149,
      "step": 42000
    },
    {
      "entropy": 5.781780481338501,
      "epoch": 3.670803111072271,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000369824813797239,
      "loss": 5.3711,
      "mean_token_accuracy": 0.19192519932985305,
      "num_tokens": 85123881.0,
      "step": 42005
    },
    {
      "entropy": 5.844514274597168,
      "epoch": 3.671240059424976,
      "grad_norm": 1.125,
      "learning_rate": 0.0003697965543026606,
      "loss": 5.3744,
      "mean_token_accuracy": 0.19017427563667297,
      "num_tokens": 85135159.0,
      "step": 42010
    },
    {
      "entropy": 5.870775270462036,
      "epoch": 3.6716770077776806,
      "grad_norm": 1.0,
      "learning_rate": 0.0003697682929898347,
      "loss": 5.3918,
      "mean_token_accuracy": 0.1993008777499199,
      "num_tokens": 85144718.0,
      "step": 42015
    },
    {
      "entropy": 5.680697345733643,
      "epoch": 3.6721139561303855,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036974002985930315,
      "loss": 5.2779,
      "mean_token_accuracy": 0.20762051194906234,
      "num_tokens": 85154094.0,
      "step": 42020
    },
    {
      "entropy": 5.769920730590821,
      "epoch": 3.67255090448309,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003697117649116083,
      "loss": 5.2914,
      "mean_token_accuracy": 0.19667955338954926,
      "num_tokens": 85163636.0,
      "step": 42025
    },
    {
      "entropy": 5.80889048576355,
      "epoch": 3.672987852835795,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003696834981472921,
      "loss": 5.3153,
      "mean_token_accuracy": 0.20387952476739885,
      "num_tokens": 85172977.0,
      "step": 42030
    },
    {
      "entropy": 5.734192323684693,
      "epoch": 3.6734248011884993,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036965522956689683,
      "loss": 5.3368,
      "mean_token_accuracy": 0.20171502679586412,
      "num_tokens": 85182956.0,
      "step": 42035
    },
    {
      "entropy": 5.814190196990967,
      "epoch": 3.6738617495412043,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003696269591709647,
      "loss": 5.409,
      "mean_token_accuracy": 0.19260480105876923,
      "num_tokens": 85193317.0,
      "step": 42040
    },
    {
      "entropy": 5.852217674255371,
      "epoch": 3.674298697893909,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036959868696003785,
      "loss": 5.3777,
      "mean_token_accuracy": 0.19375605136156082,
      "num_tokens": 85203724.0,
      "step": 42045
    },
    {
      "entropy": 5.806982469558716,
      "epoch": 3.6747356462466136,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036957041293465864,
      "loss": 5.3369,
      "mean_token_accuracy": 0.1943640723824501,
      "num_tokens": 85213409.0,
      "step": 42050
    },
    {
      "entropy": 5.846463346481324,
      "epoch": 3.675172594599318,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003695421370953694,
      "loss": 5.3833,
      "mean_token_accuracy": 0.19147460907697678,
      "num_tokens": 85223975.0,
      "step": 42055
    },
    {
      "entropy": 5.777075099945068,
      "epoch": 3.675609542952023,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00036951385944271233,
      "loss": 5.3406,
      "mean_token_accuracy": 0.19490243643522262,
      "num_tokens": 85234522.0,
      "step": 42060
    },
    {
      "entropy": 5.753358316421509,
      "epoch": 3.676046491304728,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036948557997722994,
      "loss": 5.335,
      "mean_token_accuracy": 0.19429556727409364,
      "num_tokens": 85243449.0,
      "step": 42065
    },
    {
      "entropy": 5.7812141418457035,
      "epoch": 3.6764834396574324,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036945729869946464,
      "loss": 5.3091,
      "mean_token_accuracy": 0.19638511687517166,
      "num_tokens": 85254328.0,
      "step": 42070
    },
    {
      "entropy": 5.815943861007691,
      "epoch": 3.6769203880101373,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00036942901560995886,
      "loss": 5.3185,
      "mean_token_accuracy": 0.19829144030809404,
      "num_tokens": 85264827.0,
      "step": 42075
    },
    {
      "entropy": 5.767700576782227,
      "epoch": 3.677357336362842,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003694007307092551,
      "loss": 5.2705,
      "mean_token_accuracy": 0.20109273344278336,
      "num_tokens": 85274564.0,
      "step": 42080
    },
    {
      "entropy": 5.748057842254639,
      "epoch": 3.6777942847155467,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003693724439978958,
      "loss": 5.2537,
      "mean_token_accuracy": 0.20757106840610504,
      "num_tokens": 85284086.0,
      "step": 42085
    },
    {
      "entropy": 5.853471565246582,
      "epoch": 3.678231233068251,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003693441554764236,
      "loss": 5.4134,
      "mean_token_accuracy": 0.1944814145565033,
      "num_tokens": 85294145.0,
      "step": 42090
    },
    {
      "entropy": 5.660535192489624,
      "epoch": 3.678668181420956,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003693158651453811,
      "loss": 5.2298,
      "mean_token_accuracy": 0.2162814736366272,
      "num_tokens": 85304645.0,
      "step": 42095
    },
    {
      "entropy": 5.793001699447632,
      "epoch": 3.679105129773661,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036928757300531083,
      "loss": 5.2513,
      "mean_token_accuracy": 0.20078397542238235,
      "num_tokens": 85313844.0,
      "step": 42100
    },
    {
      "entropy": 5.741316652297973,
      "epoch": 3.6795420781263655,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036925927905675555,
      "loss": 5.3575,
      "mean_token_accuracy": 0.20589490085840226,
      "num_tokens": 85322660.0,
      "step": 42105
    },
    {
      "entropy": 5.846071481704712,
      "epoch": 3.67997902647907,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003692309833002579,
      "loss": 5.3832,
      "mean_token_accuracy": 0.19385352730751038,
      "num_tokens": 85334053.0,
      "step": 42110
    },
    {
      "entropy": 5.839386653900147,
      "epoch": 3.680415974831775,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00036920268573636055,
      "loss": 5.2485,
      "mean_token_accuracy": 0.20648382157087325,
      "num_tokens": 85343890.0,
      "step": 42115
    },
    {
      "entropy": 5.716190528869629,
      "epoch": 3.6808529231844798,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036917438636560645,
      "loss": 5.2765,
      "mean_token_accuracy": 0.2048608258366585,
      "num_tokens": 85353325.0,
      "step": 42120
    },
    {
      "entropy": 5.795178651809692,
      "epoch": 3.6812898715371842,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003691460851885382,
      "loss": 5.4279,
      "mean_token_accuracy": 0.19062014818191528,
      "num_tokens": 85363072.0,
      "step": 42125
    },
    {
      "entropy": 5.756953859329224,
      "epoch": 3.681726819889889,
      "grad_norm": 1.0,
      "learning_rate": 0.0003691177822056987,
      "loss": 5.2401,
      "mean_token_accuracy": 0.1989774748682976,
      "num_tokens": 85372709.0,
      "step": 42130
    },
    {
      "entropy": 5.871082019805908,
      "epoch": 3.6821637682425936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036908947741763084,
      "loss": 5.398,
      "mean_token_accuracy": 0.20362657755613328,
      "num_tokens": 85383143.0,
      "step": 42135
    },
    {
      "entropy": 5.7692962169647215,
      "epoch": 3.6826007165952985,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003690611708248774,
      "loss": 5.3136,
      "mean_token_accuracy": 0.20530474185943604,
      "num_tokens": 85393245.0,
      "step": 42140
    },
    {
      "entropy": 5.756060171127319,
      "epoch": 3.683037664948003,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036903286242798155,
      "loss": 5.2789,
      "mean_token_accuracy": 0.1979948952794075,
      "num_tokens": 85404031.0,
      "step": 42145
    },
    {
      "entropy": 5.797820138931274,
      "epoch": 3.683474613300708,
      "grad_norm": 0.90625,
      "learning_rate": 0.0003690045522274861,
      "loss": 5.3784,
      "mean_token_accuracy": 0.19439157098531723,
      "num_tokens": 85414561.0,
      "step": 42150
    },
    {
      "entropy": 5.844581508636475,
      "epoch": 3.683911561653413,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036897624022393403,
      "loss": 5.3948,
      "mean_token_accuracy": 0.1906420558691025,
      "num_tokens": 85424755.0,
      "step": 42155
    },
    {
      "entropy": 5.775044250488281,
      "epoch": 3.6843485100061173,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036894792641786853,
      "loss": 5.3514,
      "mean_token_accuracy": 0.19816180765628816,
      "num_tokens": 85434086.0,
      "step": 42160
    },
    {
      "entropy": 5.759356832504272,
      "epoch": 3.6847854583588218,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036891961080983245,
      "loss": 5.3237,
      "mean_token_accuracy": 0.19435893446207048,
      "num_tokens": 85443540.0,
      "step": 42165
    },
    {
      "entropy": 5.828338718414306,
      "epoch": 3.6852224067115267,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003688912934003691,
      "loss": 5.3298,
      "mean_token_accuracy": 0.20633936375379563,
      "num_tokens": 85452813.0,
      "step": 42170
    },
    {
      "entropy": 5.835817813873291,
      "epoch": 3.6856593550642316,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036886297419002153,
      "loss": 5.3641,
      "mean_token_accuracy": 0.19277258217334747,
      "num_tokens": 85461838.0,
      "step": 42175
    },
    {
      "entropy": 5.824565505981445,
      "epoch": 3.686096303416936,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003688346531793329,
      "loss": 5.4087,
      "mean_token_accuracy": 0.18659051060676574,
      "num_tokens": 85472272.0,
      "step": 42180
    },
    {
      "entropy": 5.830305051803589,
      "epoch": 3.686533251769641,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003688063303688465,
      "loss": 5.3222,
      "mean_token_accuracy": 0.20039959698915483,
      "num_tokens": 85482663.0,
      "step": 42185
    },
    {
      "entropy": 5.81942458152771,
      "epoch": 3.6869702001223454,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003687780057591055,
      "loss": 5.3467,
      "mean_token_accuracy": 0.19081711769104004,
      "num_tokens": 85492697.0,
      "step": 42190
    },
    {
      "entropy": 5.770183181762695,
      "epoch": 3.6874071484750504,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003687496793506532,
      "loss": 5.2867,
      "mean_token_accuracy": 0.20438645631074906,
      "num_tokens": 85501785.0,
      "step": 42195
    },
    {
      "entropy": 5.785554552078247,
      "epoch": 3.687844096827755,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036872135114403293,
      "loss": 5.303,
      "mean_token_accuracy": 0.20062290579080583,
      "num_tokens": 85512493.0,
      "step": 42200
    },
    {
      "entropy": 5.824826860427857,
      "epoch": 3.6882810451804597,
      "grad_norm": 1.125,
      "learning_rate": 0.000368693021139788,
      "loss": 5.3412,
      "mean_token_accuracy": 0.19871870428323746,
      "num_tokens": 85521118.0,
      "step": 42205
    },
    {
      "entropy": 5.84834041595459,
      "epoch": 3.6887179935331647,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003686646893384619,
      "loss": 5.3988,
      "mean_token_accuracy": 0.19240282475948334,
      "num_tokens": 85531048.0,
      "step": 42210
    },
    {
      "entropy": 5.837186050415039,
      "epoch": 3.689154941885869,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000368636355740598,
      "loss": 5.2986,
      "mean_token_accuracy": 0.19629057943820954,
      "num_tokens": 85541526.0,
      "step": 42215
    },
    {
      "entropy": 5.762677192687988,
      "epoch": 3.6895918902385736,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003686080203467397,
      "loss": 5.302,
      "mean_token_accuracy": 0.20134770274162292,
      "num_tokens": 85551588.0,
      "step": 42220
    },
    {
      "entropy": 5.74679479598999,
      "epoch": 3.6900288385912785,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003685796831574305,
      "loss": 5.3069,
      "mean_token_accuracy": 0.2032110646367073,
      "num_tokens": 85560775.0,
      "step": 42225
    },
    {
      "entropy": 5.7416235446929935,
      "epoch": 3.6904657869439834,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003685513441732139,
      "loss": 5.336,
      "mean_token_accuracy": 0.19367437809705734,
      "num_tokens": 85571495.0,
      "step": 42230
    },
    {
      "entropy": 5.835062646865845,
      "epoch": 3.690902735296688,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003685230033946336,
      "loss": 5.3389,
      "mean_token_accuracy": 0.1948864534497261,
      "num_tokens": 85583782.0,
      "step": 42235
    },
    {
      "entropy": 5.821642637252808,
      "epoch": 3.691339683649393,
      "grad_norm": 1.0625,
      "learning_rate": 0.000368494660822233,
      "loss": 5.4139,
      "mean_token_accuracy": 0.1939649313688278,
      "num_tokens": 85594342.0,
      "step": 42240
    },
    {
      "entropy": 5.819929790496826,
      "epoch": 3.6917766320020973,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003684663164565558,
      "loss": 5.3171,
      "mean_token_accuracy": 0.19997702538967133,
      "num_tokens": 85604829.0,
      "step": 42245
    },
    {
      "entropy": 5.784072399139404,
      "epoch": 3.692213580354802,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036843797029814575,
      "loss": 5.2846,
      "mean_token_accuracy": 0.20353931039571763,
      "num_tokens": 85614530.0,
      "step": 42250
    },
    {
      "entropy": 5.802684020996094,
      "epoch": 3.6926505287075067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003684096223475464,
      "loss": 5.3276,
      "mean_token_accuracy": 0.19750421792268752,
      "num_tokens": 85624110.0,
      "step": 42255
    },
    {
      "entropy": 5.82811336517334,
      "epoch": 3.6930874770602116,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003683812726053017,
      "loss": 5.3589,
      "mean_token_accuracy": 0.19989126026630402,
      "num_tokens": 85634345.0,
      "step": 42260
    },
    {
      "entropy": 5.781299591064453,
      "epoch": 3.693524425412916,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003683529210719551,
      "loss": 5.2941,
      "mean_token_accuracy": 0.20405644476413726,
      "num_tokens": 85644813.0,
      "step": 42265
    },
    {
      "entropy": 5.77764048576355,
      "epoch": 3.693961373765621,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003683245677480506,
      "loss": 5.3837,
      "mean_token_accuracy": 0.19028246402740479,
      "num_tokens": 85655848.0,
      "step": 42270
    },
    {
      "entropy": 5.811873149871826,
      "epoch": 3.6943983221183254,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000368296212634132,
      "loss": 5.2783,
      "mean_token_accuracy": 0.19540243595838547,
      "num_tokens": 85664672.0,
      "step": 42275
    },
    {
      "entropy": 5.831072664260864,
      "epoch": 3.6948352704710303,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003682678557307432,
      "loss": 5.3772,
      "mean_token_accuracy": 0.18965646177530288,
      "num_tokens": 85674179.0,
      "step": 42280
    },
    {
      "entropy": 5.80156455039978,
      "epoch": 3.6952722188237352,
      "grad_norm": 1.0625,
      "learning_rate": 0.000368239497038428,
      "loss": 5.3748,
      "mean_token_accuracy": 0.1936066895723343,
      "num_tokens": 85684244.0,
      "step": 42285
    },
    {
      "entropy": 5.851490592956543,
      "epoch": 3.6957091671764397,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003682111365577304,
      "loss": 5.4025,
      "mean_token_accuracy": 0.19230435639619828,
      "num_tokens": 85694998.0,
      "step": 42290
    },
    {
      "entropy": 5.814294004440308,
      "epoch": 3.696146115529144,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00036818277428919433,
      "loss": 5.3522,
      "mean_token_accuracy": 0.19198853820562362,
      "num_tokens": 85705083.0,
      "step": 42295
    },
    {
      "entropy": 5.76931414604187,
      "epoch": 3.696583063881849,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003681544102333638,
      "loss": 5.342,
      "mean_token_accuracy": 0.19956194907426833,
      "num_tokens": 85714442.0,
      "step": 42300
    },
    {
      "entropy": 5.806853151321411,
      "epoch": 3.697020012234554,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036812604439078297,
      "loss": 5.2623,
      "mean_token_accuracy": 0.2085899755358696,
      "num_tokens": 85724124.0,
      "step": 42305
    },
    {
      "entropy": 5.819533443450927,
      "epoch": 3.6974569605872585,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036809767676199585,
      "loss": 5.3314,
      "mean_token_accuracy": 0.19715208262205125,
      "num_tokens": 85734271.0,
      "step": 42310
    },
    {
      "entropy": 5.799130058288574,
      "epoch": 3.6978939089399634,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036806930734754634,
      "loss": 5.3577,
      "mean_token_accuracy": 0.19477418065071106,
      "num_tokens": 85744008.0,
      "step": 42315
    },
    {
      "entropy": 5.8127869129180905,
      "epoch": 3.698330857292668,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036804093614797883,
      "loss": 5.3477,
      "mean_token_accuracy": 0.19751254618167877,
      "num_tokens": 85755377.0,
      "step": 42320
    },
    {
      "entropy": 5.813068771362305,
      "epoch": 3.6987678056453728,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003680125631638375,
      "loss": 5.1854,
      "mean_token_accuracy": 0.2099561169743538,
      "num_tokens": 85764640.0,
      "step": 42325
    },
    {
      "entropy": 5.734732294082642,
      "epoch": 3.6992047539980772,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036798418839566633,
      "loss": 5.2844,
      "mean_token_accuracy": 0.20378794968128205,
      "num_tokens": 85774282.0,
      "step": 42330
    },
    {
      "entropy": 5.726059436798096,
      "epoch": 3.699641702350782,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036795581184400977,
      "loss": 5.2904,
      "mean_token_accuracy": 0.19992888867855071,
      "num_tokens": 85783936.0,
      "step": 42335
    },
    {
      "entropy": 5.853030633926392,
      "epoch": 3.700078650703487,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000367927433509412,
      "loss": 5.3849,
      "mean_token_accuracy": 0.1918627217411995,
      "num_tokens": 85794427.0,
      "step": 42340
    },
    {
      "entropy": 5.82481107711792,
      "epoch": 3.7005155990561915,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036789905339241737,
      "loss": 5.4011,
      "mean_token_accuracy": 0.1901980385184288,
      "num_tokens": 85804367.0,
      "step": 42345
    },
    {
      "entropy": 5.748404788970947,
      "epoch": 3.700952547408896,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003678706714935702,
      "loss": 5.228,
      "mean_token_accuracy": 0.20549803972244263,
      "num_tokens": 85814393.0,
      "step": 42350
    },
    {
      "entropy": 5.777960968017578,
      "epoch": 3.701389495761601,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003678422878134149,
      "loss": 5.3001,
      "mean_token_accuracy": 0.2008540377020836,
      "num_tokens": 85824979.0,
      "step": 42355
    },
    {
      "entropy": 5.820379590988159,
      "epoch": 3.701826444114306,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036781390235249593,
      "loss": 5.3777,
      "mean_token_accuracy": 0.1946059376001358,
      "num_tokens": 85835181.0,
      "step": 42360
    },
    {
      "entropy": 5.7975935459136965,
      "epoch": 3.7022633924670103,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003677855151113576,
      "loss": 5.3326,
      "mean_token_accuracy": 0.1990017592906952,
      "num_tokens": 85844720.0,
      "step": 42365
    },
    {
      "entropy": 5.855209684371948,
      "epoch": 3.702700340819715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003677571260905445,
      "loss": 5.3832,
      "mean_token_accuracy": 0.1933203235268593,
      "num_tokens": 85854085.0,
      "step": 42370
    },
    {
      "entropy": 5.812873697280883,
      "epoch": 3.7031372891724197,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003677287352906012,
      "loss": 5.3506,
      "mean_token_accuracy": 0.19438573569059373,
      "num_tokens": 85864675.0,
      "step": 42375
    },
    {
      "entropy": 5.813150787353516,
      "epoch": 3.7035742375251246,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036770034271207204,
      "loss": 5.2929,
      "mean_token_accuracy": 0.2037007376551628,
      "num_tokens": 85875122.0,
      "step": 42380
    },
    {
      "entropy": 5.786781263351441,
      "epoch": 3.704011185877829,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00036767194835550185,
      "loss": 5.2512,
      "mean_token_accuracy": 0.20480717867612838,
      "num_tokens": 85884485.0,
      "step": 42385
    },
    {
      "entropy": 5.725072765350342,
      "epoch": 3.704448134230534,
      "grad_norm": 1.015625,
      "learning_rate": 0.000367643552221435,
      "loss": 5.3348,
      "mean_token_accuracy": 0.19796541035175325,
      "num_tokens": 85895033.0,
      "step": 42390
    },
    {
      "entropy": 5.780206537246704,
      "epoch": 3.704885082583239,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036761515431041634,
      "loss": 5.288,
      "mean_token_accuracy": 0.19897780418395997,
      "num_tokens": 85904882.0,
      "step": 42395
    },
    {
      "entropy": 5.821351814270019,
      "epoch": 3.7053220309359434,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036758675462299044,
      "loss": 5.3159,
      "mean_token_accuracy": 0.1986992210149765,
      "num_tokens": 85914173.0,
      "step": 42400
    },
    {
      "entropy": 5.749496126174927,
      "epoch": 3.705758979288648,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003675583531597022,
      "loss": 5.357,
      "mean_token_accuracy": 0.19218400865793228,
      "num_tokens": 85923682.0,
      "step": 42405
    },
    {
      "entropy": 5.805094146728516,
      "epoch": 3.7061959276413527,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036752994992109617,
      "loss": 5.3126,
      "mean_token_accuracy": 0.19734741002321243,
      "num_tokens": 85934061.0,
      "step": 42410
    },
    {
      "entropy": 5.8146717071533205,
      "epoch": 3.7066328759940577,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003675015449077172,
      "loss": 5.3331,
      "mean_token_accuracy": 0.19489277452230452,
      "num_tokens": 85943568.0,
      "step": 42415
    },
    {
      "entropy": 5.800482940673828,
      "epoch": 3.707069824346762,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036747313812011007,
      "loss": 5.3107,
      "mean_token_accuracy": 0.2034081533551216,
      "num_tokens": 85953998.0,
      "step": 42420
    },
    {
      "entropy": 5.758801603317261,
      "epoch": 3.707506772699467,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003674447295588198,
      "loss": 5.2951,
      "mean_token_accuracy": 0.20061323791742325,
      "num_tokens": 85963520.0,
      "step": 42425
    },
    {
      "entropy": 5.751541709899902,
      "epoch": 3.7079437210521715,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003674163192243912,
      "loss": 5.3034,
      "mean_token_accuracy": 0.1985554113984108,
      "num_tokens": 85974142.0,
      "step": 42430
    },
    {
      "entropy": 5.791504669189453,
      "epoch": 3.7083806694048764,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036738790711736907,
      "loss": 5.3466,
      "mean_token_accuracy": 0.19681477248668672,
      "num_tokens": 85984724.0,
      "step": 42435
    },
    {
      "entropy": 5.755845403671264,
      "epoch": 3.708817617757581,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003673594932382985,
      "loss": 5.2326,
      "mean_token_accuracy": 0.20829349905252456,
      "num_tokens": 85995475.0,
      "step": 42440
    },
    {
      "entropy": 5.734438467025757,
      "epoch": 3.709254566110286,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003673310775877244,
      "loss": 5.314,
      "mean_token_accuracy": 0.2063568577170372,
      "num_tokens": 86005672.0,
      "step": 42445
    },
    {
      "entropy": 5.802830839157105,
      "epoch": 3.7096915144629907,
      "grad_norm": 1.109375,
      "learning_rate": 0.000367302660166192,
      "loss": 5.3878,
      "mean_token_accuracy": 0.19345163851976394,
      "num_tokens": 86016918.0,
      "step": 42450
    },
    {
      "entropy": 5.831452703475952,
      "epoch": 3.710128462815695,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00036727424097424615,
      "loss": 5.3645,
      "mean_token_accuracy": 0.19245204776525499,
      "num_tokens": 86029345.0,
      "step": 42455
    },
    {
      "entropy": 5.8306052684783936,
      "epoch": 3.7105654111683997,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036724582001243196,
      "loss": 5.3113,
      "mean_token_accuracy": 0.20018555521965026,
      "num_tokens": 86039368.0,
      "step": 42460
    },
    {
      "entropy": 5.822173309326172,
      "epoch": 3.7110023595211046,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003672173972812946,
      "loss": 5.3603,
      "mean_token_accuracy": 0.1904800847172737,
      "num_tokens": 86048220.0,
      "step": 42465
    },
    {
      "entropy": 5.856272649765015,
      "epoch": 3.7114393078738095,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036718897278137933,
      "loss": 5.3811,
      "mean_token_accuracy": 0.18874009549617768,
      "num_tokens": 86058316.0,
      "step": 42470
    },
    {
      "entropy": 5.771484470367431,
      "epoch": 3.711876256226514,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003671605465132312,
      "loss": 5.2887,
      "mean_token_accuracy": 0.20048342794179916,
      "num_tokens": 86068995.0,
      "step": 42475
    },
    {
      "entropy": 5.795468950271607,
      "epoch": 3.712313204579219,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036713211847739553,
      "loss": 5.2905,
      "mean_token_accuracy": 0.2072661265730858,
      "num_tokens": 86080604.0,
      "step": 42480
    },
    {
      "entropy": 5.759581851959228,
      "epoch": 3.7127501529319233,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036710368867441743,
      "loss": 5.3522,
      "mean_token_accuracy": 0.20045081824064254,
      "num_tokens": 86092636.0,
      "step": 42485
    },
    {
      "entropy": 5.777671670913696,
      "epoch": 3.7131871012846283,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003670752571048425,
      "loss": 5.3296,
      "mean_token_accuracy": 0.194071988761425,
      "num_tokens": 86101454.0,
      "step": 42490
    },
    {
      "entropy": 5.791636180877686,
      "epoch": 3.7136240496373327,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036704682376921576,
      "loss": 5.3522,
      "mean_token_accuracy": 0.20143762081861497,
      "num_tokens": 86111460.0,
      "step": 42495
    },
    {
      "entropy": 5.839567756652832,
      "epoch": 3.7140609979900376,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003670183886680828,
      "loss": 5.3425,
      "mean_token_accuracy": 0.1993425816297531,
      "num_tokens": 86121347.0,
      "step": 42500
    },
    {
      "entropy": 5.8239343643188475,
      "epoch": 3.7144979463427426,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036698995180198886,
      "loss": 5.4335,
      "mean_token_accuracy": 0.19088583290576935,
      "num_tokens": 86132084.0,
      "step": 42505
    },
    {
      "entropy": 5.692169189453125,
      "epoch": 3.714934894695447,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036696151317147944,
      "loss": 5.2494,
      "mean_token_accuracy": 0.21124820411205292,
      "num_tokens": 86141300.0,
      "step": 42510
    },
    {
      "entropy": 5.761719560623169,
      "epoch": 3.7153718430481515,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036693307277710005,
      "loss": 5.3165,
      "mean_token_accuracy": 0.19989665299654008,
      "num_tokens": 86151640.0,
      "step": 42515
    },
    {
      "entropy": 5.738762426376343,
      "epoch": 3.7158087914008564,
      "grad_norm": 0.96875,
      "learning_rate": 0.00036690463061939615,
      "loss": 5.1992,
      "mean_token_accuracy": 0.21097466051578523,
      "num_tokens": 86162442.0,
      "step": 42520
    },
    {
      "entropy": 5.750937747955322,
      "epoch": 3.7162457397535613,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003668761866989132,
      "loss": 5.3258,
      "mean_token_accuracy": 0.19528876394033431,
      "num_tokens": 86172824.0,
      "step": 42525
    },
    {
      "entropy": 5.7376752376556395,
      "epoch": 3.716682688106266,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003668477410161969,
      "loss": 5.2587,
      "mean_token_accuracy": 0.2070658713579178,
      "num_tokens": 86182829.0,
      "step": 42530
    },
    {
      "entropy": 5.778853130340576,
      "epoch": 3.7171196364589707,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036681929357179285,
      "loss": 5.3805,
      "mean_token_accuracy": 0.19320559054613112,
      "num_tokens": 86193656.0,
      "step": 42535
    },
    {
      "entropy": 5.734366369247437,
      "epoch": 3.717556584811675,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003667908443662466,
      "loss": 5.2844,
      "mean_token_accuracy": 0.2076038658618927,
      "num_tokens": 86203025.0,
      "step": 42540
    },
    {
      "entropy": 5.697314834594726,
      "epoch": 3.71799353316438,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036676239340010386,
      "loss": 5.2821,
      "mean_token_accuracy": 0.20268132984638215,
      "num_tokens": 86213029.0,
      "step": 42545
    },
    {
      "entropy": 5.882200717926025,
      "epoch": 3.7184304815170846,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003667339406739103,
      "loss": 5.3566,
      "mean_token_accuracy": 0.1833513006567955,
      "num_tokens": 86222725.0,
      "step": 42550
    },
    {
      "entropy": 5.800853252410889,
      "epoch": 3.7188674298697895,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003667054861882117,
      "loss": 5.4045,
      "mean_token_accuracy": 0.18629931062459945,
      "num_tokens": 86232514.0,
      "step": 42555
    },
    {
      "entropy": 5.846207237243652,
      "epoch": 3.7193043782224944,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036667702994355386,
      "loss": 5.4508,
      "mean_token_accuracy": 0.18847987353801726,
      "num_tokens": 86242840.0,
      "step": 42560
    },
    {
      "entropy": 5.896905469894409,
      "epoch": 3.719741326575199,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003666485719404825,
      "loss": 5.4472,
      "mean_token_accuracy": 0.18420123904943467,
      "num_tokens": 86253221.0,
      "step": 42565
    },
    {
      "entropy": 5.839185523986816,
      "epoch": 3.7201782749279033,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036662011217954355,
      "loss": 5.3493,
      "mean_token_accuracy": 0.1904863655567169,
      "num_tokens": 86263154.0,
      "step": 42570
    },
    {
      "entropy": 5.808301973342895,
      "epoch": 3.7206152232806082,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003665916506612828,
      "loss": 5.3951,
      "mean_token_accuracy": 0.18705807477235795,
      "num_tokens": 86273041.0,
      "step": 42575
    },
    {
      "entropy": 5.79590106010437,
      "epoch": 3.721052171633313,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036656318738624626,
      "loss": 5.3247,
      "mean_token_accuracy": 0.20185233056545257,
      "num_tokens": 86284669.0,
      "step": 42580
    },
    {
      "entropy": 5.703448104858398,
      "epoch": 3.7214891199860176,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003665347223549797,
      "loss": 5.2044,
      "mean_token_accuracy": 0.20574459284543992,
      "num_tokens": 86294186.0,
      "step": 42585
    },
    {
      "entropy": 5.717581796646118,
      "epoch": 3.7219260683387225,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003665062555680293,
      "loss": 5.2748,
      "mean_token_accuracy": 0.19382278025150299,
      "num_tokens": 86304055.0,
      "step": 42590
    },
    {
      "entropy": 5.840995502471924,
      "epoch": 3.722363016691427,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036647778702594087,
      "loss": 5.3541,
      "mean_token_accuracy": 0.19456524699926375,
      "num_tokens": 86314258.0,
      "step": 42595
    },
    {
      "entropy": 5.751426076889038,
      "epoch": 3.722799965044132,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036644931672926063,
      "loss": 5.2912,
      "mean_token_accuracy": 0.20031103789806365,
      "num_tokens": 86324829.0,
      "step": 42600
    },
    {
      "entropy": 5.775260972976684,
      "epoch": 3.7232369133968364,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003664208446785346,
      "loss": 5.3446,
      "mean_token_accuracy": 0.19698090851306915,
      "num_tokens": 86334901.0,
      "step": 42605
    },
    {
      "entropy": 5.8000650882720945,
      "epoch": 3.7236738617495413,
      "grad_norm": 1.125,
      "learning_rate": 0.00036639237087430875,
      "loss": 5.3415,
      "mean_token_accuracy": 0.19472158402204515,
      "num_tokens": 86345798.0,
      "step": 42610
    },
    {
      "entropy": 5.8888627052307125,
      "epoch": 3.7241108101022458,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003663638953171295,
      "loss": 5.4313,
      "mean_token_accuracy": 0.18676805794239043,
      "num_tokens": 86355676.0,
      "step": 42615
    },
    {
      "entropy": 5.788529729843139,
      "epoch": 3.7245477584549507,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036633541800754277,
      "loss": 5.3409,
      "mean_token_accuracy": 0.19825909435749053,
      "num_tokens": 86366590.0,
      "step": 42620
    },
    {
      "entropy": 5.760331773757935,
      "epoch": 3.724984706807655,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036630693894609487,
      "loss": 5.3724,
      "mean_token_accuracy": 0.19280479848384857,
      "num_tokens": 86376495.0,
      "step": 42625
    },
    {
      "entropy": 5.736511182785034,
      "epoch": 3.72542165516036,
      "grad_norm": 0.953125,
      "learning_rate": 0.00036627845813333204,
      "loss": 5.3344,
      "mean_token_accuracy": 0.19371307790279388,
      "num_tokens": 86386990.0,
      "step": 42630
    },
    {
      "entropy": 5.827192306518555,
      "epoch": 3.725858603513065,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003662499755698006,
      "loss": 5.3139,
      "mean_token_accuracy": 0.19552787989377976,
      "num_tokens": 86396989.0,
      "step": 42635
    },
    {
      "entropy": 5.783801031112671,
      "epoch": 3.7262955518657694,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036622149125604676,
      "loss": 5.3126,
      "mean_token_accuracy": 0.20075297951698304,
      "num_tokens": 86407870.0,
      "step": 42640
    },
    {
      "entropy": 5.7362171649932865,
      "epoch": 3.726732500218474,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003661930051926169,
      "loss": 5.2773,
      "mean_token_accuracy": 0.2074267163872719,
      "num_tokens": 86417831.0,
      "step": 42645
    },
    {
      "entropy": 5.7945075035095215,
      "epoch": 3.727169448571179,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003661645173800574,
      "loss": 5.4217,
      "mean_token_accuracy": 0.19808091819286347,
      "num_tokens": 86428186.0,
      "step": 42650
    },
    {
      "entropy": 5.757077217102051,
      "epoch": 3.7276063969238837,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036613602781891474,
      "loss": 5.3725,
      "mean_token_accuracy": 0.19747121781110763,
      "num_tokens": 86438592.0,
      "step": 42655
    },
    {
      "entropy": 5.754097366333008,
      "epoch": 3.728043345276588,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036610753650973536,
      "loss": 5.2459,
      "mean_token_accuracy": 0.20333611965179443,
      "num_tokens": 86448488.0,
      "step": 42660
    },
    {
      "entropy": 5.779180383682251,
      "epoch": 3.728480293629293,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003660790434530656,
      "loss": 5.3068,
      "mean_token_accuracy": 0.20631697326898574,
      "num_tokens": 86459524.0,
      "step": 42665
    },
    {
      "entropy": 5.717403984069824,
      "epoch": 3.7289172419819976,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036605054864945213,
      "loss": 5.1788,
      "mean_token_accuracy": 0.21165538132190703,
      "num_tokens": 86470264.0,
      "step": 42670
    },
    {
      "entropy": 5.752198648452759,
      "epoch": 3.7293541903347025,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003660220520994413,
      "loss": 5.3388,
      "mean_token_accuracy": 0.19545484483242034,
      "num_tokens": 86481155.0,
      "step": 42675
    },
    {
      "entropy": 5.831137704849243,
      "epoch": 3.729791138687407,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036599355380358004,
      "loss": 5.3618,
      "mean_token_accuracy": 0.1894158124923706,
      "num_tokens": 86491666.0,
      "step": 42680
    },
    {
      "entropy": 5.822264862060547,
      "epoch": 3.730228087040112,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036596505376241455,
      "loss": 5.4002,
      "mean_token_accuracy": 0.19341200441122056,
      "num_tokens": 86501953.0,
      "step": 42685
    },
    {
      "entropy": 5.712327098846435,
      "epoch": 3.730665035392817,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003659365519764917,
      "loss": 5.2541,
      "mean_token_accuracy": 0.20507272481918334,
      "num_tokens": 86512658.0,
      "step": 42690
    },
    {
      "entropy": 5.773210144042968,
      "epoch": 3.7311019837455213,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036590804844635817,
      "loss": 5.4324,
      "mean_token_accuracy": 0.19147348999977112,
      "num_tokens": 86522856.0,
      "step": 42695
    },
    {
      "entropy": 5.886049079895019,
      "epoch": 3.7315389320982257,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00036587954317256064,
      "loss": 5.3751,
      "mean_token_accuracy": 0.19214088916778566,
      "num_tokens": 86533650.0,
      "step": 42700
    },
    {
      "entropy": 5.888300228118896,
      "epoch": 3.7319758804509306,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003658510361556458,
      "loss": 5.3791,
      "mean_token_accuracy": 0.19621682167053223,
      "num_tokens": 86543801.0,
      "step": 42705
    },
    {
      "entropy": 5.780824899673462,
      "epoch": 3.7324128288036356,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003658225273961605,
      "loss": 5.3263,
      "mean_token_accuracy": 0.19352971762418747,
      "num_tokens": 86553637.0,
      "step": 42710
    },
    {
      "entropy": 5.803317022323609,
      "epoch": 3.73284977715634,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003657940168946516,
      "loss": 5.3992,
      "mean_token_accuracy": 0.19434398859739305,
      "num_tokens": 86563531.0,
      "step": 42715
    },
    {
      "entropy": 5.704267692565918,
      "epoch": 3.733286725509045,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003657655046516658,
      "loss": 5.276,
      "mean_token_accuracy": 0.2109770104289055,
      "num_tokens": 86573675.0,
      "step": 42720
    },
    {
      "entropy": 5.717259359359741,
      "epoch": 3.7337236738617494,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036573699066775006,
      "loss": 5.2896,
      "mean_token_accuracy": 0.19775913655757904,
      "num_tokens": 86583817.0,
      "step": 42725
    },
    {
      "entropy": 5.753678464889527,
      "epoch": 3.7341606222144543,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036570847494345134,
      "loss": 5.3344,
      "mean_token_accuracy": 0.19563426822423935,
      "num_tokens": 86593170.0,
      "step": 42730
    },
    {
      "entropy": 5.793945741653443,
      "epoch": 3.734597570567159,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003656799574793166,
      "loss": 5.3581,
      "mean_token_accuracy": 0.1945022836327553,
      "num_tokens": 86603224.0,
      "step": 42735
    },
    {
      "entropy": 5.769664192199707,
      "epoch": 3.7350345189198637,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00036565143827589267,
      "loss": 5.2762,
      "mean_token_accuracy": 0.20380765348672866,
      "num_tokens": 86613442.0,
      "step": 42740
    },
    {
      "entropy": 5.769879674911499,
      "epoch": 3.7354714672725686,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003656229173337266,
      "loss": 5.2385,
      "mean_token_accuracy": 0.20301606506109238,
      "num_tokens": 86623459.0,
      "step": 42745
    },
    {
      "entropy": 5.670239877700806,
      "epoch": 3.735908415625273,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003655943946533656,
      "loss": 5.2234,
      "mean_token_accuracy": 0.2146788001060486,
      "num_tokens": 86633407.0,
      "step": 42750
    },
    {
      "entropy": 5.738392925262451,
      "epoch": 3.7363453639779776,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003655658702353566,
      "loss": 5.3605,
      "mean_token_accuracy": 0.19323530048131943,
      "num_tokens": 86643902.0,
      "step": 42755
    },
    {
      "entropy": 5.777231597900391,
      "epoch": 3.7367823123306825,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003655373440802467,
      "loss": 5.3128,
      "mean_token_accuracy": 0.20155907571315765,
      "num_tokens": 86653871.0,
      "step": 42760
    },
    {
      "entropy": 5.795123338699341,
      "epoch": 3.7372192606833874,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003655088161885831,
      "loss": 5.2624,
      "mean_token_accuracy": 0.20822289437055588,
      "num_tokens": 86663890.0,
      "step": 42765
    },
    {
      "entropy": 5.702305173873901,
      "epoch": 3.737656209036092,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003654802865609131,
      "loss": 5.2905,
      "mean_token_accuracy": 0.20838228315114976,
      "num_tokens": 86672954.0,
      "step": 42770
    },
    {
      "entropy": 5.738843822479248,
      "epoch": 3.7380931573887968,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003654517551977837,
      "loss": 5.3776,
      "mean_token_accuracy": 0.1942624941468239,
      "num_tokens": 86682432.0,
      "step": 42775
    },
    {
      "entropy": 5.838733291625976,
      "epoch": 3.7385301057415012,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00036542322209974227,
      "loss": 5.2988,
      "mean_token_accuracy": 0.19756577610969545,
      "num_tokens": 86691832.0,
      "step": 42780
    },
    {
      "entropy": 5.8452709197998045,
      "epoch": 3.738967054094206,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000365394687267336,
      "loss": 5.4071,
      "mean_token_accuracy": 0.19488171488046646,
      "num_tokens": 86703284.0,
      "step": 42785
    },
    {
      "entropy": 5.836964797973633,
      "epoch": 3.7394040024469106,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036536615070111234,
      "loss": 5.392,
      "mean_token_accuracy": 0.19656782150268554,
      "num_tokens": 86713619.0,
      "step": 42790
    },
    {
      "entropy": 5.80809645652771,
      "epoch": 3.7398409507996155,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003653376124016185,
      "loss": 5.3069,
      "mean_token_accuracy": 0.19372112303972244,
      "num_tokens": 86723802.0,
      "step": 42795
    },
    {
      "entropy": 5.776024532318115,
      "epoch": 3.7402778991523205,
      "grad_norm": 1.125,
      "learning_rate": 0.000365309072369402,
      "loss": 5.325,
      "mean_token_accuracy": 0.19943070858716966,
      "num_tokens": 86733855.0,
      "step": 42800
    },
    {
      "entropy": 5.746087217330933,
      "epoch": 3.740714847505025,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036528053060501,
      "loss": 5.3054,
      "mean_token_accuracy": 0.203433758020401,
      "num_tokens": 86742969.0,
      "step": 42805
    },
    {
      "entropy": 5.779957008361817,
      "epoch": 3.7411517958577294,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003652519871089903,
      "loss": 5.3407,
      "mean_token_accuracy": 0.19762025475502015,
      "num_tokens": 86753749.0,
      "step": 42810
    },
    {
      "entropy": 5.8242357730865475,
      "epoch": 3.7415887442104343,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003652234418818901,
      "loss": 5.3363,
      "mean_token_accuracy": 0.1982959493994713,
      "num_tokens": 86764287.0,
      "step": 42815
    },
    {
      "entropy": 5.793248844146729,
      "epoch": 3.742025692563139,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036519489492425704,
      "loss": 5.3868,
      "mean_token_accuracy": 0.1969745099544525,
      "num_tokens": 86774711.0,
      "step": 42820
    },
    {
      "entropy": 5.663901853561401,
      "epoch": 3.7424626409158437,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003651663462366387,
      "loss": 5.2709,
      "mean_token_accuracy": 0.20856418758630751,
      "num_tokens": 86784551.0,
      "step": 42825
    },
    {
      "entropy": 5.70191011428833,
      "epoch": 3.7428995892685486,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003651377958195825,
      "loss": 5.1893,
      "mean_token_accuracy": 0.21186779141426088,
      "num_tokens": 86794258.0,
      "step": 42830
    },
    {
      "entropy": 5.763357496261596,
      "epoch": 3.743336537621253,
      "grad_norm": 1.125,
      "learning_rate": 0.0003651092436736362,
      "loss": 5.29,
      "mean_token_accuracy": 0.19910774976015091,
      "num_tokens": 86804332.0,
      "step": 42835
    },
    {
      "entropy": 5.781505107879639,
      "epoch": 3.743773485973958,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003650806897993473,
      "loss": 5.2535,
      "mean_token_accuracy": 0.20718594640493393,
      "num_tokens": 86814593.0,
      "step": 42840
    },
    {
      "entropy": 5.78445200920105,
      "epoch": 3.7442104343266625,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036505213419726364,
      "loss": 5.39,
      "mean_token_accuracy": 0.19382372498512268,
      "num_tokens": 86824489.0,
      "step": 42845
    },
    {
      "entropy": 5.729154300689697,
      "epoch": 3.7446473826793674,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036502357686793284,
      "loss": 5.3299,
      "mean_token_accuracy": 0.20525248050689698,
      "num_tokens": 86835052.0,
      "step": 42850
    },
    {
      "entropy": 5.774785709381104,
      "epoch": 3.7450843310320723,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003649950178119027,
      "loss": 5.2859,
      "mean_token_accuracy": 0.2069399043917656,
      "num_tokens": 86845183.0,
      "step": 42855
    },
    {
      "entropy": 5.793636989593506,
      "epoch": 3.7455212793847767,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003649664570297209,
      "loss": 5.2742,
      "mean_token_accuracy": 0.20625607967376708,
      "num_tokens": 86854752.0,
      "step": 42860
    },
    {
      "entropy": 5.829070234298706,
      "epoch": 3.745958227737481,
      "grad_norm": 1.0,
      "learning_rate": 0.00036493789452193533,
      "loss": 5.3208,
      "mean_token_accuracy": 0.1990916073322296,
      "num_tokens": 86865876.0,
      "step": 42865
    },
    {
      "entropy": 5.762755823135376,
      "epoch": 3.746395176090186,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036490933028909383,
      "loss": 5.3486,
      "mean_token_accuracy": 0.2003444939851761,
      "num_tokens": 86875626.0,
      "step": 42870
    },
    {
      "entropy": 5.734955310821533,
      "epoch": 3.746832124442891,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036488076433174426,
      "loss": 5.2558,
      "mean_token_accuracy": 0.1927575871348381,
      "num_tokens": 86884944.0,
      "step": 42875
    },
    {
      "entropy": 5.779360961914063,
      "epoch": 3.7472690727955955,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036485219665043463,
      "loss": 5.3379,
      "mean_token_accuracy": 0.1978630468249321,
      "num_tokens": 86894570.0,
      "step": 42880
    },
    {
      "entropy": 5.782916116714477,
      "epoch": 3.7477060211483004,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003648236272457127,
      "loss": 5.2984,
      "mean_token_accuracy": 0.19885711520910263,
      "num_tokens": 86904093.0,
      "step": 42885
    },
    {
      "entropy": 5.791521215438843,
      "epoch": 3.748142969501005,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003647950561181265,
      "loss": 5.3159,
      "mean_token_accuracy": 0.19884065836668013,
      "num_tokens": 86915019.0,
      "step": 42890
    },
    {
      "entropy": 5.783283615112305,
      "epoch": 3.74857991785371,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036476648326822403,
      "loss": 5.2729,
      "mean_token_accuracy": 0.2024346709251404,
      "num_tokens": 86924695.0,
      "step": 42895
    },
    {
      "entropy": 5.747125864028931,
      "epoch": 3.7490168662064143,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003647379086965535,
      "loss": 5.2598,
      "mean_token_accuracy": 0.20333430767059327,
      "num_tokens": 86935030.0,
      "step": 42900
    },
    {
      "entropy": 5.760509252548218,
      "epoch": 3.749453814559119,
      "grad_norm": 1.125,
      "learning_rate": 0.00036470933240366276,
      "loss": 5.3942,
      "mean_token_accuracy": 0.19270469695329667,
      "num_tokens": 86945871.0,
      "step": 42905
    },
    {
      "entropy": 5.840439033508301,
      "epoch": 3.7498907629118237,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036468075439010007,
      "loss": 5.5211,
      "mean_token_accuracy": 0.1945922315120697,
      "num_tokens": 86958855.0,
      "step": 42910
    },
    {
      "entropy": 5.82876992225647,
      "epoch": 3.7503277112645286,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036465217465641355,
      "loss": 5.3082,
      "mean_token_accuracy": 0.20328305214643477,
      "num_tokens": 86969071.0,
      "step": 42915
    },
    {
      "entropy": 5.780021381378174,
      "epoch": 3.750764659617233,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036462359320315127,
      "loss": 5.314,
      "mean_token_accuracy": 0.1986316442489624,
      "num_tokens": 86978985.0,
      "step": 42920
    },
    {
      "entropy": 5.723165988922119,
      "epoch": 3.751201607969938,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003645950100308616,
      "loss": 5.2121,
      "mean_token_accuracy": 0.21303258687257767,
      "num_tokens": 86988527.0,
      "step": 42925
    },
    {
      "entropy": 5.7958056926727295,
      "epoch": 3.751638556322643,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003645664251400926,
      "loss": 5.3146,
      "mean_token_accuracy": 0.20187241286039354,
      "num_tokens": 86998218.0,
      "step": 42930
    },
    {
      "entropy": 5.659995651245117,
      "epoch": 3.7520755046753473,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036453783853139266,
      "loss": 5.1614,
      "mean_token_accuracy": 0.21525321900844574,
      "num_tokens": 87007147.0,
      "step": 42935
    },
    {
      "entropy": 5.669667530059814,
      "epoch": 3.752512453028052,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003645092502053101,
      "loss": 5.2487,
      "mean_token_accuracy": 0.20476232022047042,
      "num_tokens": 87016440.0,
      "step": 42940
    },
    {
      "entropy": 5.735062456130981,
      "epoch": 3.7529494013807567,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00036448066016239316,
      "loss": 5.2288,
      "mean_token_accuracy": 0.2069290816783905,
      "num_tokens": 87027969.0,
      "step": 42945
    },
    {
      "entropy": 5.854996204376221,
      "epoch": 3.7533863497334616,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003644520684031902,
      "loss": 5.3134,
      "mean_token_accuracy": 0.19784659445285796,
      "num_tokens": 87038403.0,
      "step": 42950
    },
    {
      "entropy": 5.773972749710083,
      "epoch": 3.753823298086166,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036442347492824983,
      "loss": 5.32,
      "mean_token_accuracy": 0.19926238805055618,
      "num_tokens": 87047484.0,
      "step": 42955
    },
    {
      "entropy": 5.787786817550659,
      "epoch": 3.754260246438871,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003643948797381203,
      "loss": 5.3073,
      "mean_token_accuracy": 0.19879609644412993,
      "num_tokens": 87058151.0,
      "step": 42960
    },
    {
      "entropy": 5.676926851272583,
      "epoch": 3.7546971947915755,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003643662828333502,
      "loss": 5.2471,
      "mean_token_accuracy": 0.20424319952726364,
      "num_tokens": 87068256.0,
      "step": 42965
    },
    {
      "entropy": 5.846097469329834,
      "epoch": 3.7551341431442804,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003643376842144879,
      "loss": 5.4029,
      "mean_token_accuracy": 0.19048615396022797,
      "num_tokens": 87078346.0,
      "step": 42970
    },
    {
      "entropy": 5.774329376220703,
      "epoch": 3.755571091496985,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036430908388208205,
      "loss": 5.2745,
      "mean_token_accuracy": 0.1943726122379303,
      "num_tokens": 87088086.0,
      "step": 42975
    },
    {
      "entropy": 5.743614435195923,
      "epoch": 3.75600803984969,
      "grad_norm": 1.125,
      "learning_rate": 0.0003642804818366811,
      "loss": 5.2996,
      "mean_token_accuracy": 0.20531834065914153,
      "num_tokens": 87098358.0,
      "step": 42980
    },
    {
      "entropy": 5.85037932395935,
      "epoch": 3.7564449882023947,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036425187807883377,
      "loss": 5.3843,
      "mean_token_accuracy": 0.19071035236120223,
      "num_tokens": 87108970.0,
      "step": 42985
    },
    {
      "entropy": 5.77495584487915,
      "epoch": 3.756881936555099,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036422327260908873,
      "loss": 5.3185,
      "mean_token_accuracy": 0.20664069503545762,
      "num_tokens": 87117950.0,
      "step": 42990
    },
    {
      "entropy": 5.844204998016357,
      "epoch": 3.7573188849078036,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003641946654279945,
      "loss": 5.4305,
      "mean_token_accuracy": 0.1928043082356453,
      "num_tokens": 87129177.0,
      "step": 42995
    },
    {
      "entropy": 5.803354644775391,
      "epoch": 3.7577558332605085,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003641660565360997,
      "loss": 5.3598,
      "mean_token_accuracy": 0.1984284371137619,
      "num_tokens": 87140418.0,
      "step": 43000
    },
    {
      "entropy": 5.81222939491272,
      "epoch": 3.7581927816132135,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036413744593395345,
      "loss": 5.3276,
      "mean_token_accuracy": 0.19826783388853073,
      "num_tokens": 87151624.0,
      "step": 43005
    },
    {
      "entropy": 5.810443973541259,
      "epoch": 3.758629729965918,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036410883362210416,
      "loss": 5.3334,
      "mean_token_accuracy": 0.1950523316860199,
      "num_tokens": 87160880.0,
      "step": 43010
    },
    {
      "entropy": 5.735592079162598,
      "epoch": 3.759066678318623,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036408021960110084,
      "loss": 5.2577,
      "mean_token_accuracy": 0.20196988582611083,
      "num_tokens": 87170673.0,
      "step": 43015
    },
    {
      "entropy": 5.731219148635864,
      "epoch": 3.7595036266713273,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003640516038714921,
      "loss": 5.3126,
      "mean_token_accuracy": 0.19888730496168136,
      "num_tokens": 87180751.0,
      "step": 43020
    },
    {
      "entropy": 5.80750789642334,
      "epoch": 3.7599405750240322,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036402298643382696,
      "loss": 5.324,
      "mean_token_accuracy": 0.19821199029684067,
      "num_tokens": 87191501.0,
      "step": 43025
    },
    {
      "entropy": 5.774884128570557,
      "epoch": 3.7603775233767367,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003639943672886544,
      "loss": 5.3175,
      "mean_token_accuracy": 0.20032646209001542,
      "num_tokens": 87201382.0,
      "step": 43030
    },
    {
      "entropy": 5.758829498291016,
      "epoch": 3.7608144717294416,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00036396574643652316,
      "loss": 5.3305,
      "mean_token_accuracy": 0.1946993887424469,
      "num_tokens": 87212454.0,
      "step": 43035
    },
    {
      "entropy": 5.786156463623047,
      "epoch": 3.7612514200821465,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036393712387798236,
      "loss": 5.3058,
      "mean_token_accuracy": 0.20138151198625565,
      "num_tokens": 87222439.0,
      "step": 43040
    },
    {
      "entropy": 5.7524254322052,
      "epoch": 3.761688368434851,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00036390849961358084,
      "loss": 5.269,
      "mean_token_accuracy": 0.2033987060189247,
      "num_tokens": 87232620.0,
      "step": 43045
    },
    {
      "entropy": 5.792064762115478,
      "epoch": 3.7621253167875555,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003638798736438677,
      "loss": 5.3169,
      "mean_token_accuracy": 0.19882808923721312,
      "num_tokens": 87242031.0,
      "step": 43050
    },
    {
      "entropy": 5.699888277053833,
      "epoch": 3.7625622651402604,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036385124596939204,
      "loss": 5.1754,
      "mean_token_accuracy": 0.20720046609640122,
      "num_tokens": 87251113.0,
      "step": 43055
    },
    {
      "entropy": 5.761004304885864,
      "epoch": 3.7629992134929653,
      "grad_norm": 0.96875,
      "learning_rate": 0.000363822616590703,
      "loss": 5.2764,
      "mean_token_accuracy": 0.1988348424434662,
      "num_tokens": 87261255.0,
      "step": 43060
    },
    {
      "entropy": 5.847371912002563,
      "epoch": 3.7634361618456698,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003637939855083495,
      "loss": 5.4299,
      "mean_token_accuracy": 0.18680068254470825,
      "num_tokens": 87271677.0,
      "step": 43065
    },
    {
      "entropy": 5.818808460235596,
      "epoch": 3.7638731101983747,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003637653527228809,
      "loss": 5.4073,
      "mean_token_accuracy": 0.1924215778708458,
      "num_tokens": 87280417.0,
      "step": 43070
    },
    {
      "entropy": 5.825795412063599,
      "epoch": 3.764310058551079,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036373671823484636,
      "loss": 5.391,
      "mean_token_accuracy": 0.19067210257053374,
      "num_tokens": 87291069.0,
      "step": 43075
    },
    {
      "entropy": 5.85130352973938,
      "epoch": 3.764747006903784,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036370808204479497,
      "loss": 5.4291,
      "mean_token_accuracy": 0.19099662750959395,
      "num_tokens": 87301586.0,
      "step": 43080
    },
    {
      "entropy": 5.835968160629273,
      "epoch": 3.7651839552564885,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003636794441532761,
      "loss": 5.3141,
      "mean_token_accuracy": 0.20058224648237227,
      "num_tokens": 87312292.0,
      "step": 43085
    },
    {
      "entropy": 5.762144327163696,
      "epoch": 3.7656209036091934,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036365080456083915,
      "loss": 5.3167,
      "mean_token_accuracy": 0.19559957087039948,
      "num_tokens": 87321896.0,
      "step": 43090
    },
    {
      "entropy": 5.7485918521881105,
      "epoch": 3.7660578519618984,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036362216326803323,
      "loss": 5.3027,
      "mean_token_accuracy": 0.20218078047037125,
      "num_tokens": 87331615.0,
      "step": 43095
    },
    {
      "entropy": 5.750226163864136,
      "epoch": 3.766494800314603,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003635935202754077,
      "loss": 5.3236,
      "mean_token_accuracy": 0.20117142498493196,
      "num_tokens": 87341345.0,
      "step": 43100
    },
    {
      "entropy": 5.800769948959351,
      "epoch": 3.7669317486673073,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036356487558351214,
      "loss": 5.3274,
      "mean_token_accuracy": 0.1996422678232193,
      "num_tokens": 87351015.0,
      "step": 43105
    },
    {
      "entropy": 5.720064401626587,
      "epoch": 3.767368697020012,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036353622919289586,
      "loss": 5.3254,
      "mean_token_accuracy": 0.20527481138706208,
      "num_tokens": 87360976.0,
      "step": 43110
    },
    {
      "entropy": 5.784945249557495,
      "epoch": 3.767805645372717,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00036350758110410825,
      "loss": 5.2929,
      "mean_token_accuracy": 0.1984490856528282,
      "num_tokens": 87370863.0,
      "step": 43115
    },
    {
      "entropy": 5.792631149291992,
      "epoch": 3.7682425937254216,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036347893131769894,
      "loss": 5.3017,
      "mean_token_accuracy": 0.1936858266592026,
      "num_tokens": 87380091.0,
      "step": 43120
    },
    {
      "entropy": 5.673592281341553,
      "epoch": 3.7686795420781265,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036345027983421724,
      "loss": 5.2861,
      "mean_token_accuracy": 0.20474276095628738,
      "num_tokens": 87391005.0,
      "step": 43125
    },
    {
      "entropy": 5.734602069854736,
      "epoch": 3.769116490430831,
      "grad_norm": 1.125,
      "learning_rate": 0.0003634216266542129,
      "loss": 5.2634,
      "mean_token_accuracy": 0.20764196068048477,
      "num_tokens": 87400586.0,
      "step": 43130
    },
    {
      "entropy": 5.811687707901001,
      "epoch": 3.769553438783536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036339297177823544,
      "loss": 5.3216,
      "mean_token_accuracy": 0.2065098062157631,
      "num_tokens": 87411054.0,
      "step": 43135
    },
    {
      "entropy": 5.759176397323609,
      "epoch": 3.7699903871362404,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003633643152068344,
      "loss": 5.3011,
      "mean_token_accuracy": 0.20120766311883925,
      "num_tokens": 87421413.0,
      "step": 43140
    },
    {
      "entropy": 5.781530332565308,
      "epoch": 3.7704273354889453,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00036333565694055957,
      "loss": 5.301,
      "mean_token_accuracy": 0.1997900441288948,
      "num_tokens": 87431592.0,
      "step": 43145
    },
    {
      "entropy": 5.7310200214385985,
      "epoch": 3.77086428384165,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036330699697996043,
      "loss": 5.3067,
      "mean_token_accuracy": 0.19679830223321915,
      "num_tokens": 87441708.0,
      "step": 43150
    },
    {
      "entropy": 5.772683048248291,
      "epoch": 3.7713012321943546,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036327833532558704,
      "loss": 5.3336,
      "mean_token_accuracy": 0.1986771121621132,
      "num_tokens": 87452083.0,
      "step": 43155
    },
    {
      "entropy": 5.758668851852417,
      "epoch": 3.771738180547059,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003632496719779888,
      "loss": 5.3071,
      "mean_token_accuracy": 0.20229581445455552,
      "num_tokens": 87460743.0,
      "step": 43160
    },
    {
      "entropy": 5.8479616165161135,
      "epoch": 3.772175128899764,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036322100693771554,
      "loss": 5.3442,
      "mean_token_accuracy": 0.19537707567214965,
      "num_tokens": 87470865.0,
      "step": 43165
    },
    {
      "entropy": 5.824261713027954,
      "epoch": 3.772612077252469,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036319234020531723,
      "loss": 5.3648,
      "mean_token_accuracy": 0.19664098620414733,
      "num_tokens": 87482063.0,
      "step": 43170
    },
    {
      "entropy": 5.836217260360717,
      "epoch": 3.7730490256051734,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003631636717813435,
      "loss": 5.3327,
      "mean_token_accuracy": 0.19983362704515456,
      "num_tokens": 87492284.0,
      "step": 43175
    },
    {
      "entropy": 5.817206621170044,
      "epoch": 3.7734859739578783,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003631350016663445,
      "loss": 5.3732,
      "mean_token_accuracy": 0.198237581551075,
      "num_tokens": 87501586.0,
      "step": 43180
    },
    {
      "entropy": 5.793729686737061,
      "epoch": 3.773922922310583,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036310632986086993,
      "loss": 5.39,
      "mean_token_accuracy": 0.19052881747484207,
      "num_tokens": 87510622.0,
      "step": 43185
    },
    {
      "entropy": 5.8073670864105225,
      "epoch": 3.7743598706632877,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003630776563654698,
      "loss": 5.3598,
      "mean_token_accuracy": 0.19424076676368712,
      "num_tokens": 87519480.0,
      "step": 43190
    },
    {
      "entropy": 5.84883131980896,
      "epoch": 3.774796819015992,
      "grad_norm": 1.0625,
      "learning_rate": 0.000363048981180694,
      "loss": 5.3401,
      "mean_token_accuracy": 0.1967175006866455,
      "num_tokens": 87529560.0,
      "step": 43195
    },
    {
      "entropy": 5.821262216567993,
      "epoch": 3.775233767368697,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036302030430709263,
      "loss": 5.4017,
      "mean_token_accuracy": 0.19120453745126725,
      "num_tokens": 87539969.0,
      "step": 43200
    },
    {
      "entropy": 5.752460670471192,
      "epoch": 3.775670715721402,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036299162574521585,
      "loss": 5.3367,
      "mean_token_accuracy": 0.19522331058979034,
      "num_tokens": 87550883.0,
      "step": 43205
    },
    {
      "entropy": 5.781354904174805,
      "epoch": 3.7761076640741065,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00036296294549561343,
      "loss": 5.2998,
      "mean_token_accuracy": 0.2043622389435768,
      "num_tokens": 87560569.0,
      "step": 43210
    },
    {
      "entropy": 5.821895790100098,
      "epoch": 3.776544612426811,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003629342635588356,
      "loss": 5.3841,
      "mean_token_accuracy": 0.1942053109407425,
      "num_tokens": 87570881.0,
      "step": 43215
    },
    {
      "entropy": 5.774883985519409,
      "epoch": 3.776981560779516,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00036290557993543253,
      "loss": 5.2998,
      "mean_token_accuracy": 0.21853165924549103,
      "num_tokens": 87581334.0,
      "step": 43220
    },
    {
      "entropy": 5.781919860839844,
      "epoch": 3.7774185091322208,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036287689462595447,
      "loss": 5.3175,
      "mean_token_accuracy": 0.19860877096652985,
      "num_tokens": 87591554.0,
      "step": 43225
    },
    {
      "entropy": 5.732016324996948,
      "epoch": 3.7778554574849252,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003628482076309515,
      "loss": 5.2383,
      "mean_token_accuracy": 0.20970374196767808,
      "num_tokens": 87600484.0,
      "step": 43230
    },
    {
      "entropy": 5.759493160247803,
      "epoch": 3.77829240583763,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036281951895097384,
      "loss": 5.3335,
      "mean_token_accuracy": 0.19991566091775895,
      "num_tokens": 87610632.0,
      "step": 43235
    },
    {
      "entropy": 5.699292087554932,
      "epoch": 3.7787293541903346,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00036279082858657174,
      "loss": 5.1646,
      "mean_token_accuracy": 0.21707479208707808,
      "num_tokens": 87620835.0,
      "step": 43240
    },
    {
      "entropy": 5.818719053268433,
      "epoch": 3.7791663025430395,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003627621365382956,
      "loss": 5.3708,
      "mean_token_accuracy": 0.193058380484581,
      "num_tokens": 87630437.0,
      "step": 43245
    },
    {
      "entropy": 5.812725639343261,
      "epoch": 3.779603250895744,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036273344280669573,
      "loss": 5.3935,
      "mean_token_accuracy": 0.19340317100286483,
      "num_tokens": 87640293.0,
      "step": 43250
    },
    {
      "entropy": 5.755433559417725,
      "epoch": 3.780040199248449,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036270474739232244,
      "loss": 5.2802,
      "mean_token_accuracy": 0.19976286590099335,
      "num_tokens": 87651026.0,
      "step": 43255
    },
    {
      "entropy": 5.807600355148315,
      "epoch": 3.7804771476011534,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036267605029572617,
      "loss": 5.3255,
      "mean_token_accuracy": 0.19756643176078797,
      "num_tokens": 87660957.0,
      "step": 43260
    },
    {
      "entropy": 5.809979581832886,
      "epoch": 3.7809140959538583,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003626473515174572,
      "loss": 5.297,
      "mean_token_accuracy": 0.20962247997522354,
      "num_tokens": 87671311.0,
      "step": 43265
    },
    {
      "entropy": 5.756803178787232,
      "epoch": 3.7813510443065628,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003626186510580661,
      "loss": 5.2399,
      "mean_token_accuracy": 0.20379242151975632,
      "num_tokens": 87681217.0,
      "step": 43270
    },
    {
      "entropy": 5.758460950851441,
      "epoch": 3.7817879926592677,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036258994891810357,
      "loss": 5.3171,
      "mean_token_accuracy": 0.20385794341564178,
      "num_tokens": 87691014.0,
      "step": 43275
    },
    {
      "entropy": 5.793578481674194,
      "epoch": 3.7822249410119726,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003625612450981197,
      "loss": 5.3887,
      "mean_token_accuracy": 0.1925251767039299,
      "num_tokens": 87701125.0,
      "step": 43280
    },
    {
      "entropy": 5.723571825027466,
      "epoch": 3.782661889364677,
      "grad_norm": 1.0,
      "learning_rate": 0.00036253253959866533,
      "loss": 5.1562,
      "mean_token_accuracy": 0.21433950811624528,
      "num_tokens": 87711451.0,
      "step": 43285
    },
    {
      "entropy": 5.8147440433502195,
      "epoch": 3.7830988377173815,
      "grad_norm": 1.109375,
      "learning_rate": 0.000362503832420291,
      "loss": 5.3581,
      "mean_token_accuracy": 0.19612788110971452,
      "num_tokens": 87722753.0,
      "step": 43290
    },
    {
      "entropy": 5.761569023132324,
      "epoch": 3.7835357860700864,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003624751235635473,
      "loss": 5.2852,
      "mean_token_accuracy": 0.20200299471616745,
      "num_tokens": 87732771.0,
      "step": 43295
    },
    {
      "entropy": 5.78529314994812,
      "epoch": 3.7839727344227914,
      "grad_norm": 0.984375,
      "learning_rate": 0.000362446413028985,
      "loss": 5.3911,
      "mean_token_accuracy": 0.19280657768249512,
      "num_tokens": 87742925.0,
      "step": 43300
    },
    {
      "entropy": 5.7990693092346195,
      "epoch": 3.784409682775496,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036241770081715455,
      "loss": 5.3237,
      "mean_token_accuracy": 0.19911002963781357,
      "num_tokens": 87752700.0,
      "step": 43305
    },
    {
      "entropy": 5.773849678039551,
      "epoch": 3.7848466311282007,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003623889869286067,
      "loss": 5.2703,
      "mean_token_accuracy": 0.2016654446721077,
      "num_tokens": 87764054.0,
      "step": 43310
    },
    {
      "entropy": 5.726472568511963,
      "epoch": 3.785283579480905,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003623602713638925,
      "loss": 5.2716,
      "mean_token_accuracy": 0.20868719816207887,
      "num_tokens": 87774861.0,
      "step": 43315
    },
    {
      "entropy": 5.77738733291626,
      "epoch": 3.78572052783361,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036233155412356244,
      "loss": 5.273,
      "mean_token_accuracy": 0.20625808984041213,
      "num_tokens": 87785322.0,
      "step": 43320
    },
    {
      "entropy": 5.819278955459595,
      "epoch": 3.7861574761863146,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036230283520816744,
      "loss": 5.32,
      "mean_token_accuracy": 0.20376133024692536,
      "num_tokens": 87795357.0,
      "step": 43325
    },
    {
      "entropy": 5.736589860916138,
      "epoch": 3.7865944245390195,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003622741146182583,
      "loss": 5.311,
      "mean_token_accuracy": 0.2094930186867714,
      "num_tokens": 87805642.0,
      "step": 43330
    },
    {
      "entropy": 5.770531463623047,
      "epoch": 3.7870313728917244,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003622453923543858,
      "loss": 5.24,
      "mean_token_accuracy": 0.20653128623962402,
      "num_tokens": 87815700.0,
      "step": 43335
    },
    {
      "entropy": 5.783266019821167,
      "epoch": 3.787468321244429,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000362216668417101,
      "loss": 5.3395,
      "mean_token_accuracy": 0.1997374787926674,
      "num_tokens": 87826121.0,
      "step": 43340
    },
    {
      "entropy": 5.761190938949585,
      "epoch": 3.7879052695971334,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00036218794280695494,
      "loss": 5.2973,
      "mean_token_accuracy": 0.19618868678808213,
      "num_tokens": 87838152.0,
      "step": 43345
    },
    {
      "entropy": 5.8186970233917235,
      "epoch": 3.7883422179498383,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003621592155244984,
      "loss": 5.3379,
      "mean_token_accuracy": 0.1970250740647316,
      "num_tokens": 87848657.0,
      "step": 43350
    },
    {
      "entropy": 5.851055049896241,
      "epoch": 3.788779166302543,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036213048657028233,
      "loss": 5.4043,
      "mean_token_accuracy": 0.1913243919610977,
      "num_tokens": 87858603.0,
      "step": 43355
    },
    {
      "entropy": 5.79928822517395,
      "epoch": 3.7892161146552477,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036210175594485796,
      "loss": 5.3073,
      "mean_token_accuracy": 0.2024122416973114,
      "num_tokens": 87868091.0,
      "step": 43360
    },
    {
      "entropy": 5.765394592285157,
      "epoch": 3.7896530630079526,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00036207302364877627,
      "loss": 5.2534,
      "mean_token_accuracy": 0.20193225741386414,
      "num_tokens": 87878978.0,
      "step": 43365
    },
    {
      "entropy": 5.788683605194092,
      "epoch": 3.790090011360657,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036204428968258836,
      "loss": 5.3436,
      "mean_token_accuracy": 0.19159597307443618,
      "num_tokens": 87888547.0,
      "step": 43370
    },
    {
      "entropy": 5.740214872360229,
      "epoch": 3.790526959713362,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036201555404684533,
      "loss": 5.291,
      "mean_token_accuracy": 0.20806059688329698,
      "num_tokens": 87898395.0,
      "step": 43375
    },
    {
      "entropy": 5.726042890548706,
      "epoch": 3.7909639080660664,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003619868167420984,
      "loss": 5.31,
      "mean_token_accuracy": 0.20030437111854554,
      "num_tokens": 87909127.0,
      "step": 43380
    },
    {
      "entropy": 5.874493026733399,
      "epoch": 3.7914008564187713,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036195807776889864,
      "loss": 5.3775,
      "mean_token_accuracy": 0.19394944459199906,
      "num_tokens": 87919554.0,
      "step": 43385
    },
    {
      "entropy": 5.806141090393067,
      "epoch": 3.7918378047714763,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036192933712779756,
      "loss": 5.2931,
      "mean_token_accuracy": 0.1994737058877945,
      "num_tokens": 87930728.0,
      "step": 43390
    },
    {
      "entropy": 5.671382093429566,
      "epoch": 3.7922747531241807,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036190059481934614,
      "loss": 5.2091,
      "mean_token_accuracy": 0.20997340530157088,
      "num_tokens": 87939928.0,
      "step": 43395
    },
    {
      "entropy": 5.764880752563476,
      "epoch": 3.792711701476885,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003618718508440958,
      "loss": 5.3418,
      "mean_token_accuracy": 0.1959706127643585,
      "num_tokens": 87949978.0,
      "step": 43400
    },
    {
      "entropy": 5.868348884582519,
      "epoch": 3.79314864982959,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003618431052025978,
      "loss": 5.3957,
      "mean_token_accuracy": 0.19237599521875381,
      "num_tokens": 87960157.0,
      "step": 43405
    },
    {
      "entropy": 5.817387914657592,
      "epoch": 3.793585598182295,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003618143578954036,
      "loss": 5.2781,
      "mean_token_accuracy": 0.20005892664194108,
      "num_tokens": 87971302.0,
      "step": 43410
    },
    {
      "entropy": 5.785513591766358,
      "epoch": 3.7940225465349995,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003617856089230645,
      "loss": 5.2463,
      "mean_token_accuracy": 0.20683218091726302,
      "num_tokens": 87980829.0,
      "step": 43415
    },
    {
      "entropy": 5.666175127029419,
      "epoch": 3.7944594948877044,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036175685828613197,
      "loss": 5.2509,
      "mean_token_accuracy": 0.20675614327192307,
      "num_tokens": 87991003.0,
      "step": 43420
    },
    {
      "entropy": 5.754777717590332,
      "epoch": 3.794896443240409,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003617281059851574,
      "loss": 5.3095,
      "mean_token_accuracy": 0.19841000735759734,
      "num_tokens": 88001475.0,
      "step": 43425
    },
    {
      "entropy": 5.852871942520141,
      "epoch": 3.795333391593114,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003616993520206923,
      "loss": 5.4093,
      "mean_token_accuracy": 0.19452106654644014,
      "num_tokens": 88012714.0,
      "step": 43430
    },
    {
      "entropy": 5.749449300765991,
      "epoch": 3.7957703399458183,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036167059639328824,
      "loss": 5.3462,
      "mean_token_accuracy": 0.20261748135089874,
      "num_tokens": 88023195.0,
      "step": 43435
    },
    {
      "entropy": 5.7507898807525635,
      "epoch": 3.796207288298523,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036164183910349676,
      "loss": 5.2728,
      "mean_token_accuracy": 0.20587846934795379,
      "num_tokens": 88032977.0,
      "step": 43440
    },
    {
      "entropy": 5.769747352600097,
      "epoch": 3.796644236651228,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003616130801518694,
      "loss": 5.2896,
      "mean_token_accuracy": 0.1994936555624008,
      "num_tokens": 88041761.0,
      "step": 43445
    },
    {
      "entropy": 5.738713693618775,
      "epoch": 3.7970811850039325,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003615843195389577,
      "loss": 5.267,
      "mean_token_accuracy": 0.20288171321153642,
      "num_tokens": 88050709.0,
      "step": 43450
    },
    {
      "entropy": 5.783014822006225,
      "epoch": 3.797518133356637,
      "grad_norm": 1.0,
      "learning_rate": 0.00036155555726531343,
      "loss": 5.3538,
      "mean_token_accuracy": 0.1998336836695671,
      "num_tokens": 88061381.0,
      "step": 43455
    },
    {
      "entropy": 5.854882860183716,
      "epoch": 3.797955081709342,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036152679333148816,
      "loss": 5.3422,
      "mean_token_accuracy": 0.20223041027784347,
      "num_tokens": 88070926.0,
      "step": 43460
    },
    {
      "entropy": 5.841819763183594,
      "epoch": 3.798392030062047,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00036149802773803374,
      "loss": 5.3969,
      "mean_token_accuracy": 0.1961440458893776,
      "num_tokens": 88081240.0,
      "step": 43465
    },
    {
      "entropy": 5.865791606903076,
      "epoch": 3.7988289784147513,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003614692604855018,
      "loss": 5.4294,
      "mean_token_accuracy": 0.18877863734960557,
      "num_tokens": 88092623.0,
      "step": 43470
    },
    {
      "entropy": 5.809430360794067,
      "epoch": 3.7992659267674562,
      "grad_norm": 1.125,
      "learning_rate": 0.0003614404915744441,
      "loss": 5.302,
      "mean_token_accuracy": 0.19731666892766953,
      "num_tokens": 88102532.0,
      "step": 43475
    },
    {
      "entropy": 5.839347076416016,
      "epoch": 3.7997028751201607,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036141172100541245,
      "loss": 5.4025,
      "mean_token_accuracy": 0.18968170285224914,
      "num_tokens": 88112210.0,
      "step": 43480
    },
    {
      "entropy": 5.828470611572266,
      "epoch": 3.8001398234728656,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036138294877895876,
      "loss": 5.3177,
      "mean_token_accuracy": 0.19702390879392623,
      "num_tokens": 88121800.0,
      "step": 43485
    },
    {
      "entropy": 5.7647357940673825,
      "epoch": 3.80057677182557,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003613541748956349,
      "loss": 5.2944,
      "mean_token_accuracy": 0.198004786670208,
      "num_tokens": 88131820.0,
      "step": 43490
    },
    {
      "entropy": 5.780892705917358,
      "epoch": 3.801013720178275,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003613253993559926,
      "loss": 5.3529,
      "mean_token_accuracy": 0.19661133885383605,
      "num_tokens": 88141799.0,
      "step": 43495
    },
    {
      "entropy": 5.817732906341552,
      "epoch": 3.80145066853098,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036129662216058395,
      "loss": 5.2883,
      "mean_token_accuracy": 0.2024422436952591,
      "num_tokens": 88151860.0,
      "step": 43500
    },
    {
      "entropy": 5.802628517150879,
      "epoch": 3.8018876168836844,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003612678433099609,
      "loss": 5.2592,
      "mean_token_accuracy": 0.201169690489769,
      "num_tokens": 88160580.0,
      "step": 43505
    },
    {
      "entropy": 5.7204173564910885,
      "epoch": 3.802324565236389,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036123906280467543,
      "loss": 5.358,
      "mean_token_accuracy": 0.19591988623142242,
      "num_tokens": 88170140.0,
      "step": 43510
    },
    {
      "entropy": 5.769259595870972,
      "epoch": 3.8027615135890938,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003612102806452795,
      "loss": 5.3897,
      "mean_token_accuracy": 0.19859427958726883,
      "num_tokens": 88180288.0,
      "step": 43515
    },
    {
      "entropy": 5.7854267120361325,
      "epoch": 3.8031984619417987,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003611814968323252,
      "loss": 5.3115,
      "mean_token_accuracy": 0.20193181484937667,
      "num_tokens": 88190046.0,
      "step": 43520
    },
    {
      "entropy": 5.771516227722168,
      "epoch": 3.803635410294503,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003611527113663647,
      "loss": 5.2616,
      "mean_token_accuracy": 0.20087483823299407,
      "num_tokens": 88199321.0,
      "step": 43525
    },
    {
      "entropy": 5.784922695159912,
      "epoch": 3.804072358647208,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036112392424794996,
      "loss": 5.3513,
      "mean_token_accuracy": 0.19210771918296815,
      "num_tokens": 88209277.0,
      "step": 43530
    },
    {
      "entropy": 5.790316009521485,
      "epoch": 3.8045093069999125,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003610951354776333,
      "loss": 5.3891,
      "mean_token_accuracy": 0.19114453494548797,
      "num_tokens": 88219664.0,
      "step": 43535
    },
    {
      "entropy": 5.778087902069092,
      "epoch": 3.8049462553526174,
      "grad_norm": 1.0,
      "learning_rate": 0.00036106634505596693,
      "loss": 5.3438,
      "mean_token_accuracy": 0.19936559498310089,
      "num_tokens": 88230058.0,
      "step": 43540
    },
    {
      "entropy": 5.777495431900024,
      "epoch": 3.805383203705322,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00036103755298350277,
      "loss": 5.2686,
      "mean_token_accuracy": 0.20323338955640793,
      "num_tokens": 88240861.0,
      "step": 43545
    },
    {
      "entropy": 5.816816520690918,
      "epoch": 3.805820152058027,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036100875926079335,
      "loss": 5.3764,
      "mean_token_accuracy": 0.19250108897686005,
      "num_tokens": 88250833.0,
      "step": 43550
    },
    {
      "entropy": 5.803835153579712,
      "epoch": 3.8062571004107313,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003609799638883908,
      "loss": 5.2373,
      "mean_token_accuracy": 0.2068156585097313,
      "num_tokens": 88260209.0,
      "step": 43555
    },
    {
      "entropy": 5.831411504745484,
      "epoch": 3.806694048763436,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003609511668668476,
      "loss": 5.3756,
      "mean_token_accuracy": 0.188571734726429,
      "num_tokens": 88269961.0,
      "step": 43560
    },
    {
      "entropy": 5.747266387939453,
      "epoch": 3.8071309971161407,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036092236819671586,
      "loss": 5.3453,
      "mean_token_accuracy": 0.1926243081688881,
      "num_tokens": 88279916.0,
      "step": 43565
    },
    {
      "entropy": 5.771701097488403,
      "epoch": 3.8075679454688456,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036089356787854816,
      "loss": 5.3717,
      "mean_token_accuracy": 0.1933842197060585,
      "num_tokens": 88290459.0,
      "step": 43570
    },
    {
      "entropy": 5.835747814178466,
      "epoch": 3.8080048938215505,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003608647659128968,
      "loss": 5.3002,
      "mean_token_accuracy": 0.20255809426307678,
      "num_tokens": 88300536.0,
      "step": 43575
    },
    {
      "entropy": 5.851814699172974,
      "epoch": 3.808441842174255,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036083596230031425,
      "loss": 5.364,
      "mean_token_accuracy": 0.19923774003982545,
      "num_tokens": 88310654.0,
      "step": 43580
    },
    {
      "entropy": 5.734544849395752,
      "epoch": 3.80887879052696,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003608071570413529,
      "loss": 5.2865,
      "mean_token_accuracy": 0.20691326707601548,
      "num_tokens": 88319831.0,
      "step": 43585
    },
    {
      "entropy": 5.843900489807129,
      "epoch": 3.8093157388796643,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003607783501365653,
      "loss": 5.4581,
      "mean_token_accuracy": 0.18953412920236587,
      "num_tokens": 88330931.0,
      "step": 43590
    },
    {
      "entropy": 5.841725015640259,
      "epoch": 3.8097526872323693,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003607495415865041,
      "loss": 5.3307,
      "mean_token_accuracy": 0.19948902279138564,
      "num_tokens": 88341699.0,
      "step": 43595
    },
    {
      "entropy": 5.827655458450318,
      "epoch": 3.8101896355850737,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036072073139172174,
      "loss": 5.3503,
      "mean_token_accuracy": 0.2002466842532158,
      "num_tokens": 88351375.0,
      "step": 43600
    },
    {
      "entropy": 5.79531078338623,
      "epoch": 3.8106265839377786,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036069191955277085,
      "loss": 5.3441,
      "mean_token_accuracy": 0.20334294438362122,
      "num_tokens": 88361865.0,
      "step": 43605
    },
    {
      "entropy": 5.779934072494507,
      "epoch": 3.811063532290483,
      "grad_norm": 1.046875,
      "learning_rate": 0.000360663106070204,
      "loss": 5.2909,
      "mean_token_accuracy": 0.20313978791236878,
      "num_tokens": 88371276.0,
      "step": 43610
    },
    {
      "entropy": 5.762072324752808,
      "epoch": 3.811500480643188,
      "grad_norm": 1.0,
      "learning_rate": 0.0003606342909445739,
      "loss": 5.2752,
      "mean_token_accuracy": 0.20631274729967117,
      "num_tokens": 88382098.0,
      "step": 43615
    },
    {
      "entropy": 5.816520023345947,
      "epoch": 3.8119374289958925,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003606054741764332,
      "loss": 5.2936,
      "mean_token_accuracy": 0.198983533680439,
      "num_tokens": 88391488.0,
      "step": 43620
    },
    {
      "entropy": 5.819920825958252,
      "epoch": 3.8123743773485974,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036057665576633466,
      "loss": 5.3223,
      "mean_token_accuracy": 0.20198715776205062,
      "num_tokens": 88401636.0,
      "step": 43625
    },
    {
      "entropy": 5.723612594604492,
      "epoch": 3.8128113257013023,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000360547835714831,
      "loss": 5.2563,
      "mean_token_accuracy": 0.2062033608555794,
      "num_tokens": 88412434.0,
      "step": 43630
    },
    {
      "entropy": 5.789831113815308,
      "epoch": 3.813248274054007,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003605190140224751,
      "loss": 5.3457,
      "mean_token_accuracy": 0.18825860023498536,
      "num_tokens": 88421995.0,
      "step": 43635
    },
    {
      "entropy": 5.788514566421509,
      "epoch": 3.8136852224067113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036049019068981966,
      "loss": 5.2324,
      "mean_token_accuracy": 0.21172960102558136,
      "num_tokens": 88431691.0,
      "step": 43640
    },
    {
      "entropy": 5.788782548904419,
      "epoch": 3.814122170759416,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003604613657174175,
      "loss": 5.3189,
      "mean_token_accuracy": 0.20510915219783782,
      "num_tokens": 88441284.0,
      "step": 43645
    },
    {
      "entropy": 5.712762689590454,
      "epoch": 3.814559119112121,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003604325391058216,
      "loss": 5.3009,
      "mean_token_accuracy": 0.20435905903577806,
      "num_tokens": 88450674.0,
      "step": 43650
    },
    {
      "entropy": 5.821035718917846,
      "epoch": 3.8149960674648256,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003604037108555849,
      "loss": 5.3593,
      "mean_token_accuracy": 0.19090043753385544,
      "num_tokens": 88461082.0,
      "step": 43655
    },
    {
      "entropy": 5.858132648468017,
      "epoch": 3.8154330158175305,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003603748809672602,
      "loss": 5.3949,
      "mean_token_accuracy": 0.19082180708646773,
      "num_tokens": 88471229.0,
      "step": 43660
    },
    {
      "entropy": 5.869896316528321,
      "epoch": 3.815869964170235,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036034604944140057,
      "loss": 5.391,
      "mean_token_accuracy": 0.19425344914197923,
      "num_tokens": 88481269.0,
      "step": 43665
    },
    {
      "entropy": 5.821843862533569,
      "epoch": 3.81630691252294,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036031721627855905,
      "loss": 5.2617,
      "mean_token_accuracy": 0.20317813903093337,
      "num_tokens": 88490754.0,
      "step": 43670
    },
    {
      "entropy": 5.735591554641724,
      "epoch": 3.8167438608756443,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036028838147928854,
      "loss": 5.3098,
      "mean_token_accuracy": 0.20604006797075272,
      "num_tokens": 88500487.0,
      "step": 43675
    },
    {
      "entropy": 5.72548542022705,
      "epoch": 3.8171808092283492,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036025954504414223,
      "loss": 5.2374,
      "mean_token_accuracy": 0.20774151980876923,
      "num_tokens": 88510857.0,
      "step": 43680
    },
    {
      "entropy": 5.777362966537476,
      "epoch": 3.817617757581054,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036023070697367314,
      "loss": 5.3092,
      "mean_token_accuracy": 0.20182986855506896,
      "num_tokens": 88522551.0,
      "step": 43685
    },
    {
      "entropy": 5.812146711349487,
      "epoch": 3.8180547059337586,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036020186726843447,
      "loss": 5.4189,
      "mean_token_accuracy": 0.194760699570179,
      "num_tokens": 88534069.0,
      "step": 43690
    },
    {
      "entropy": 5.778885364532471,
      "epoch": 3.818491654286463,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003601730259289793,
      "loss": 5.2273,
      "mean_token_accuracy": 0.2049483075737953,
      "num_tokens": 88544912.0,
      "step": 43695
    },
    {
      "entropy": 5.667371463775635,
      "epoch": 3.818928602639168,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003601441829558609,
      "loss": 5.1975,
      "mean_token_accuracy": 0.22601616382598877,
      "num_tokens": 88554445.0,
      "step": 43700
    },
    {
      "entropy": 5.751909875869751,
      "epoch": 3.819365550991873,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003601153383496324,
      "loss": 5.3292,
      "mean_token_accuracy": 0.20143994987010955,
      "num_tokens": 88563589.0,
      "step": 43705
    },
    {
      "entropy": 5.829052352905274,
      "epoch": 3.8198024993445774,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036008649211084716,
      "loss": 5.3932,
      "mean_token_accuracy": 0.19568243771791458,
      "num_tokens": 88574072.0,
      "step": 43710
    },
    {
      "entropy": 5.786689710617066,
      "epoch": 3.8202394476972823,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003600576442400584,
      "loss": 5.3153,
      "mean_token_accuracy": 0.19813265055418014,
      "num_tokens": 88584225.0,
      "step": 43715
    },
    {
      "entropy": 5.791597509384156,
      "epoch": 3.8206763960499868,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003600287947378195,
      "loss": 5.2278,
      "mean_token_accuracy": 0.21099233329296113,
      "num_tokens": 88594364.0,
      "step": 43720
    },
    {
      "entropy": 5.789405584335327,
      "epoch": 3.8211133444026917,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003599999436046837,
      "loss": 5.3725,
      "mean_token_accuracy": 0.19546314775943757,
      "num_tokens": 88605678.0,
      "step": 43725
    },
    {
      "entropy": 5.811369800567627,
      "epoch": 3.821550292755396,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035997109084120446,
      "loss": 5.2959,
      "mean_token_accuracy": 0.2043576180934906,
      "num_tokens": 88615824.0,
      "step": 43730
    },
    {
      "entropy": 5.800316095352173,
      "epoch": 3.821987241108101,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035994223644793516,
      "loss": 5.3318,
      "mean_token_accuracy": 0.19770106673240662,
      "num_tokens": 88625939.0,
      "step": 43735
    },
    {
      "entropy": 5.7809418678283695,
      "epoch": 3.822424189460806,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003599133804254293,
      "loss": 5.2746,
      "mean_token_accuracy": 0.2042969360947609,
      "num_tokens": 88635453.0,
      "step": 43740
    },
    {
      "entropy": 5.769077253341675,
      "epoch": 3.8228611378135104,
      "grad_norm": 1.0,
      "learning_rate": 0.0003598845227742402,
      "loss": 5.2474,
      "mean_token_accuracy": 0.20427243709564208,
      "num_tokens": 88646190.0,
      "step": 43745
    },
    {
      "entropy": 5.748137903213501,
      "epoch": 3.823298086166215,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035985566349492156,
      "loss": 5.2774,
      "mean_token_accuracy": 0.20225575268268586,
      "num_tokens": 88655957.0,
      "step": 43750
    },
    {
      "entropy": 5.824967718124389,
      "epoch": 3.82373503451892,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035982680258802685,
      "loss": 5.2722,
      "mean_token_accuracy": 0.2009318843483925,
      "num_tokens": 88665884.0,
      "step": 43755
    },
    {
      "entropy": 5.756330490112305,
      "epoch": 3.8241719828716247,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003597979400541096,
      "loss": 5.287,
      "mean_token_accuracy": 0.198964986205101,
      "num_tokens": 88677226.0,
      "step": 43760
    },
    {
      "entropy": 5.750030851364135,
      "epoch": 3.824608931224329,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003597690758937234,
      "loss": 5.2681,
      "mean_token_accuracy": 0.20474891662597655,
      "num_tokens": 88687447.0,
      "step": 43765
    },
    {
      "entropy": 5.739684343338013,
      "epoch": 3.825045879577034,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035974021010742196,
      "loss": 5.2771,
      "mean_token_accuracy": 0.20302165746688844,
      "num_tokens": 88696540.0,
      "step": 43770
    },
    {
      "entropy": 5.788246202468872,
      "epoch": 3.8254828279297386,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00035971134269575884,
      "loss": 5.2911,
      "mean_token_accuracy": 0.20151625126600264,
      "num_tokens": 88707945.0,
      "step": 43775
    },
    {
      "entropy": 5.862472677230835,
      "epoch": 3.8259197762824435,
      "grad_norm": 1.125,
      "learning_rate": 0.00035968247365928774,
      "loss": 5.3918,
      "mean_token_accuracy": 0.19268443882465364,
      "num_tokens": 88717398.0,
      "step": 43780
    },
    {
      "entropy": 5.8099977493286135,
      "epoch": 3.826356724635148,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035965360299856247,
      "loss": 5.2523,
      "mean_token_accuracy": 0.19608585089445113,
      "num_tokens": 88727030.0,
      "step": 43785
    },
    {
      "entropy": 5.754948472976684,
      "epoch": 3.826793672987853,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003596247307141368,
      "loss": 5.2264,
      "mean_token_accuracy": 0.21043471544981002,
      "num_tokens": 88737413.0,
      "step": 43790
    },
    {
      "entropy": 5.761193943023682,
      "epoch": 3.827230621340558,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003595958568065644,
      "loss": 5.3013,
      "mean_token_accuracy": 0.20753124505281448,
      "num_tokens": 88747577.0,
      "step": 43795
    },
    {
      "entropy": 5.781335258483887,
      "epoch": 3.8276675696932623,
      "grad_norm": 1.125,
      "learning_rate": 0.0003595669812763991,
      "loss": 5.3098,
      "mean_token_accuracy": 0.19318805634975433,
      "num_tokens": 88757023.0,
      "step": 43800
    },
    {
      "entropy": 5.743558931350708,
      "epoch": 3.8281045180459667,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003595381041241948,
      "loss": 5.253,
      "mean_token_accuracy": 0.21561974436044692,
      "num_tokens": 88767200.0,
      "step": 43805
    },
    {
      "entropy": 5.75216588973999,
      "epoch": 3.8285414663986717,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035950922535050534,
      "loss": 5.3327,
      "mean_token_accuracy": 0.20179086774587632,
      "num_tokens": 88777448.0,
      "step": 43810
    },
    {
      "entropy": 5.727777051925659,
      "epoch": 3.8289784147513766,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035948034495588473,
      "loss": 5.2265,
      "mean_token_accuracy": 0.20938412100076675,
      "num_tokens": 88787477.0,
      "step": 43815
    },
    {
      "entropy": 5.837803649902344,
      "epoch": 3.829415363104081,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003594514629408868,
      "loss": 5.3566,
      "mean_token_accuracy": 0.19641371816396713,
      "num_tokens": 88796971.0,
      "step": 43820
    },
    {
      "entropy": 5.719193840026856,
      "epoch": 3.829852311456786,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003594225793060656,
      "loss": 5.2806,
      "mean_token_accuracy": 0.2046870157122612,
      "num_tokens": 88807960.0,
      "step": 43825
    },
    {
      "entropy": 5.835083293914795,
      "epoch": 3.8302892598094904,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000359393694051975,
      "loss": 5.395,
      "mean_token_accuracy": 0.18984873592853546,
      "num_tokens": 88818874.0,
      "step": 43830
    },
    {
      "entropy": 5.80519061088562,
      "epoch": 3.8307262081621953,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003593648071791692,
      "loss": 5.2986,
      "mean_token_accuracy": 0.20028655529022216,
      "num_tokens": 88829216.0,
      "step": 43835
    },
    {
      "entropy": 5.859808540344238,
      "epoch": 3.8311631565149,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035933591868820213,
      "loss": 5.3698,
      "mean_token_accuracy": 0.19249035269021988,
      "num_tokens": 88839261.0,
      "step": 43840
    },
    {
      "entropy": 5.8034374713897705,
      "epoch": 3.8316001048676047,
      "grad_norm": 0.984375,
      "learning_rate": 0.00035930702857962804,
      "loss": 5.2701,
      "mean_token_accuracy": 0.2080129489302635,
      "num_tokens": 88849180.0,
      "step": 43845
    },
    {
      "entropy": 5.726749420166016,
      "epoch": 3.8320370532203096,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003592781368540009,
      "loss": 5.3532,
      "mean_token_accuracy": 0.20197911858558654,
      "num_tokens": 88859696.0,
      "step": 43850
    },
    {
      "entropy": 5.864733076095581,
      "epoch": 3.832474001573014,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035924924351187496,
      "loss": 5.3836,
      "mean_token_accuracy": 0.19752678275108337,
      "num_tokens": 88869932.0,
      "step": 43855
    },
    {
      "entropy": 5.815910148620605,
      "epoch": 3.8329109499257186,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035922034855380436,
      "loss": 5.3285,
      "mean_token_accuracy": 0.20054179131984712,
      "num_tokens": 88880946.0,
      "step": 43860
    },
    {
      "entropy": 5.780624294281006,
      "epoch": 3.8333478982784235,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035919145198034337,
      "loss": 5.3407,
      "mean_token_accuracy": 0.2002656266093254,
      "num_tokens": 88890409.0,
      "step": 43865
    },
    {
      "entropy": 5.779843425750732,
      "epoch": 3.8337848466311284,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035916255379204613,
      "loss": 5.4175,
      "mean_token_accuracy": 0.19157432913780212,
      "num_tokens": 88900872.0,
      "step": 43870
    },
    {
      "entropy": 5.855286598205566,
      "epoch": 3.834221794983833,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003591336539894671,
      "loss": 5.2564,
      "mean_token_accuracy": 0.20461922734975815,
      "num_tokens": 88909999.0,
      "step": 43875
    },
    {
      "entropy": 5.7378191471099855,
      "epoch": 3.834658743336538,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003591047525731605,
      "loss": 5.2544,
      "mean_token_accuracy": 0.20748024433851242,
      "num_tokens": 88920219.0,
      "step": 43880
    },
    {
      "entropy": 5.75292387008667,
      "epoch": 3.8350956916892422,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035907584954368067,
      "loss": 5.372,
      "mean_token_accuracy": 0.1989452749490738,
      "num_tokens": 88930115.0,
      "step": 43885
    },
    {
      "entropy": 5.837615823745727,
      "epoch": 3.835532640041947,
      "grad_norm": 1.125,
      "learning_rate": 0.00035904694490158196,
      "loss": 5.3364,
      "mean_token_accuracy": 0.20155636966228485,
      "num_tokens": 88940114.0,
      "step": 43890
    },
    {
      "entropy": 5.867182493209839,
      "epoch": 3.8359695883946516,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003590180386474188,
      "loss": 5.4186,
      "mean_token_accuracy": 0.19115301370620727,
      "num_tokens": 88951955.0,
      "step": 43895
    },
    {
      "entropy": 5.798640727996826,
      "epoch": 3.8364065367473565,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003589891307817457,
      "loss": 5.2379,
      "mean_token_accuracy": 0.2081568643450737,
      "num_tokens": 88961677.0,
      "step": 43900
    },
    {
      "entropy": 5.803599166870117,
      "epoch": 3.836843485100061,
      "grad_norm": 0.984375,
      "learning_rate": 0.000358960221305117,
      "loss": 5.2945,
      "mean_token_accuracy": 0.1981617823243141,
      "num_tokens": 88971888.0,
      "step": 43905
    },
    {
      "entropy": 5.836359882354737,
      "epoch": 3.837280433452766,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003589313102180873,
      "loss": 5.3896,
      "mean_token_accuracy": 0.19749129712581634,
      "num_tokens": 88980681.0,
      "step": 43910
    },
    {
      "entropy": 5.808224630355835,
      "epoch": 3.8377173818054704,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003589023975212111,
      "loss": 5.3908,
      "mean_token_accuracy": 0.1949414387345314,
      "num_tokens": 88991410.0,
      "step": 43915
    },
    {
      "entropy": 5.769586181640625,
      "epoch": 3.8381543301581753,
      "grad_norm": 0.953125,
      "learning_rate": 0.00035887348321504285,
      "loss": 5.2455,
      "mean_token_accuracy": 0.20991192013025284,
      "num_tokens": 89001743.0,
      "step": 43920
    },
    {
      "entropy": 5.711469268798828,
      "epoch": 3.8385912785108802,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035884456730013735,
      "loss": 5.2326,
      "mean_token_accuracy": 0.20930353105068206,
      "num_tokens": 89011138.0,
      "step": 43925
    },
    {
      "entropy": 5.7725859642028805,
      "epoch": 3.8390282268635847,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035881564977704913,
      "loss": 5.3286,
      "mean_token_accuracy": 0.19972403198480607,
      "num_tokens": 89021282.0,
      "step": 43930
    },
    {
      "entropy": 5.780422306060791,
      "epoch": 3.839465175216289,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035878673064633276,
      "loss": 5.3672,
      "mean_token_accuracy": 0.19788229912519456,
      "num_tokens": 89032656.0,
      "step": 43935
    },
    {
      "entropy": 5.854022645950318,
      "epoch": 3.839902123568994,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035875780990854303,
      "loss": 5.3824,
      "mean_token_accuracy": 0.1904473379254341,
      "num_tokens": 89042727.0,
      "step": 43940
    },
    {
      "entropy": 5.7629759311676025,
      "epoch": 3.840339071921699,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035872888756423456,
      "loss": 5.3005,
      "mean_token_accuracy": 0.19963733106851578,
      "num_tokens": 89052111.0,
      "step": 43945
    },
    {
      "entropy": 5.834768915176392,
      "epoch": 3.8407760202744035,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003586999636139622,
      "loss": 5.3508,
      "mean_token_accuracy": 0.191773921251297,
      "num_tokens": 89062423.0,
      "step": 43950
    },
    {
      "entropy": 5.7906327724456785,
      "epoch": 3.8412129686271084,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035867103805828067,
      "loss": 5.328,
      "mean_token_accuracy": 0.2012015774846077,
      "num_tokens": 89072457.0,
      "step": 43955
    },
    {
      "entropy": 5.816726732254028,
      "epoch": 3.841649916979813,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035864211089774487,
      "loss": 5.3451,
      "mean_token_accuracy": 0.1940385952591896,
      "num_tokens": 89081919.0,
      "step": 43960
    },
    {
      "entropy": 5.7622123718261715,
      "epoch": 3.8420868653325178,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003586131821329094,
      "loss": 5.2838,
      "mean_token_accuracy": 0.20044596642255783,
      "num_tokens": 89092177.0,
      "step": 43965
    },
    {
      "entropy": 5.781714820861817,
      "epoch": 3.8425238136852222,
      "grad_norm": 1.0,
      "learning_rate": 0.0003585842517643294,
      "loss": 5.3299,
      "mean_token_accuracy": 0.20349081158638,
      "num_tokens": 89102721.0,
      "step": 43970
    },
    {
      "entropy": 5.805011653900147,
      "epoch": 3.842960762037927,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035855531979255955,
      "loss": 5.2841,
      "mean_token_accuracy": 0.2053544744849205,
      "num_tokens": 89111841.0,
      "step": 43975
    },
    {
      "entropy": 5.790456295013428,
      "epoch": 3.843397710390632,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035852638621815497,
      "loss": 5.2901,
      "mean_token_accuracy": 0.20335128903388977,
      "num_tokens": 89122551.0,
      "step": 43980
    },
    {
      "entropy": 5.805319213867188,
      "epoch": 3.8438346587433365,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003584974510416706,
      "loss": 5.3762,
      "mean_token_accuracy": 0.19869826287031173,
      "num_tokens": 89133388.0,
      "step": 43985
    },
    {
      "entropy": 5.820262002944946,
      "epoch": 3.844271607096041,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035846851426366125,
      "loss": 5.3343,
      "mean_token_accuracy": 0.1965230718255043,
      "num_tokens": 89144435.0,
      "step": 43990
    },
    {
      "entropy": 5.786741542816162,
      "epoch": 3.844708555448746,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035843957588468207,
      "loss": 5.3296,
      "mean_token_accuracy": 0.2045712113380432,
      "num_tokens": 89154509.0,
      "step": 43995
    },
    {
      "entropy": 5.801753425598145,
      "epoch": 3.845145503801451,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003584106359052881,
      "loss": 5.4002,
      "mean_token_accuracy": 0.1952074483036995,
      "num_tokens": 89165352.0,
      "step": 44000
    },
    {
      "entropy": 5.772665691375733,
      "epoch": 3.8455824521541553,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003583816943260345,
      "loss": 5.2961,
      "mean_token_accuracy": 0.19734595119953155,
      "num_tokens": 89175185.0,
      "step": 44005
    },
    {
      "entropy": 5.821208047866821,
      "epoch": 3.84601940050686,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003583527511474763,
      "loss": 5.2942,
      "mean_token_accuracy": 0.1972964271903038,
      "num_tokens": 89183864.0,
      "step": 44010
    },
    {
      "entropy": 5.812565755844116,
      "epoch": 3.8464563488595647,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035832380637016846,
      "loss": 5.3417,
      "mean_token_accuracy": 0.19826022982597352,
      "num_tokens": 89194011.0,
      "step": 44015
    },
    {
      "entropy": 5.724667930603028,
      "epoch": 3.8468932972122696,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035829485999466657,
      "loss": 5.2966,
      "mean_token_accuracy": 0.2038285404443741,
      "num_tokens": 89204213.0,
      "step": 44020
    },
    {
      "entropy": 5.722606039047241,
      "epoch": 3.847330245564974,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003582659120215256,
      "loss": 5.2478,
      "mean_token_accuracy": 0.20982445478439332,
      "num_tokens": 89214160.0,
      "step": 44025
    },
    {
      "entropy": 5.734520339965821,
      "epoch": 3.847767193917679,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035823696245130073,
      "loss": 5.2637,
      "mean_token_accuracy": 0.2021468237042427,
      "num_tokens": 89223399.0,
      "step": 44030
    },
    {
      "entropy": 5.812994384765625,
      "epoch": 3.848204142270384,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035820801128454735,
      "loss": 5.2931,
      "mean_token_accuracy": 0.20234616994857788,
      "num_tokens": 89232703.0,
      "step": 44035
    },
    {
      "entropy": 5.835131740570068,
      "epoch": 3.8486410906230883,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003581790585218207,
      "loss": 5.3463,
      "mean_token_accuracy": 0.19583945572376252,
      "num_tokens": 89243302.0,
      "step": 44040
    },
    {
      "entropy": 5.791399717330933,
      "epoch": 3.849078038975793,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003581501041636761,
      "loss": 5.3526,
      "mean_token_accuracy": 0.19594552516937255,
      "num_tokens": 89253340.0,
      "step": 44045
    },
    {
      "entropy": 5.788732433319092,
      "epoch": 3.8495149873284977,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035812114821066903,
      "loss": 5.216,
      "mean_token_accuracy": 0.202625635266304,
      "num_tokens": 89263362.0,
      "step": 44050
    },
    {
      "entropy": 5.6848853588104244,
      "epoch": 3.8499519356812026,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003580921906633547,
      "loss": 5.1847,
      "mean_token_accuracy": 0.2136819466948509,
      "num_tokens": 89275386.0,
      "step": 44055
    },
    {
      "entropy": 5.816863775253296,
      "epoch": 3.850388884033907,
      "grad_norm": 1.125,
      "learning_rate": 0.0003580632315222886,
      "loss": 5.3303,
      "mean_token_accuracy": 0.1952592834830284,
      "num_tokens": 89285325.0,
      "step": 44060
    },
    {
      "entropy": 5.765636301040649,
      "epoch": 3.850825832386612,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003580342707880262,
      "loss": 5.2328,
      "mean_token_accuracy": 0.20440353006124495,
      "num_tokens": 89295850.0,
      "step": 44065
    },
    {
      "entropy": 5.7399664402008055,
      "epoch": 3.8512627807393165,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000358005308461123,
      "loss": 5.236,
      "mean_token_accuracy": 0.20883962661027908,
      "num_tokens": 89305403.0,
      "step": 44070
    },
    {
      "entropy": 5.700406408309936,
      "epoch": 3.8516997290920214,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035797634454213454,
      "loss": 5.1828,
      "mean_token_accuracy": 0.2077389895915985,
      "num_tokens": 89314551.0,
      "step": 44075
    },
    {
      "entropy": 5.727781581878662,
      "epoch": 3.852136677444726,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003579473790316162,
      "loss": 5.3235,
      "mean_token_accuracy": 0.19830652326345444,
      "num_tokens": 89323872.0,
      "step": 44080
    },
    {
      "entropy": 5.784640455245972,
      "epoch": 3.852573625797431,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035791841193012373,
      "loss": 5.2896,
      "mean_token_accuracy": 0.20761553943157196,
      "num_tokens": 89335114.0,
      "step": 44085
    },
    {
      "entropy": 5.813876914978027,
      "epoch": 3.8530105741501357,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003578894432382127,
      "loss": 5.2623,
      "mean_token_accuracy": 0.20166416466236115,
      "num_tokens": 89344382.0,
      "step": 44090
    },
    {
      "entropy": 5.731975030899048,
      "epoch": 3.85344752250284,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003578604729564387,
      "loss": 5.3254,
      "mean_token_accuracy": 0.1959819585084915,
      "num_tokens": 89355327.0,
      "step": 44095
    },
    {
      "entropy": 5.794822072982788,
      "epoch": 3.8538844708555446,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035783150108535737,
      "loss": 5.3017,
      "mean_token_accuracy": 0.20749431848526,
      "num_tokens": 89365361.0,
      "step": 44100
    },
    {
      "entropy": 5.779147148132324,
      "epoch": 3.8543214192082496,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035780252762552444,
      "loss": 5.3196,
      "mean_token_accuracy": 0.20116116553545,
      "num_tokens": 89375097.0,
      "step": 44105
    },
    {
      "entropy": 5.803750514984131,
      "epoch": 3.8547583675609545,
      "grad_norm": 1.125,
      "learning_rate": 0.00035777355257749573,
      "loss": 5.2663,
      "mean_token_accuracy": 0.2074946105480194,
      "num_tokens": 89384934.0,
      "step": 44110
    },
    {
      "entropy": 5.731619882583618,
      "epoch": 3.855195315913659,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035774457594182685,
      "loss": 5.1978,
      "mean_token_accuracy": 0.2139734745025635,
      "num_tokens": 89395493.0,
      "step": 44115
    },
    {
      "entropy": 5.711506986618042,
      "epoch": 3.855632264266364,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003577155977190736,
      "loss": 5.2437,
      "mean_token_accuracy": 0.21606750786304474,
      "num_tokens": 89405057.0,
      "step": 44120
    },
    {
      "entropy": 5.751554107666015,
      "epoch": 3.8560692126190683,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035768661790979193,
      "loss": 5.257,
      "mean_token_accuracy": 0.20926400423049926,
      "num_tokens": 89415328.0,
      "step": 44125
    },
    {
      "entropy": 5.731973123550415,
      "epoch": 3.8565061609717732,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003576576365145376,
      "loss": 5.2872,
      "mean_token_accuracy": 0.2079477846622467,
      "num_tokens": 89425138.0,
      "step": 44130
    },
    {
      "entropy": 5.740159177780152,
      "epoch": 3.8569431093244777,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035762865353386637,
      "loss": 5.2906,
      "mean_token_accuracy": 0.19913170486688614,
      "num_tokens": 89434972.0,
      "step": 44135
    },
    {
      "entropy": 5.749774312973022,
      "epoch": 3.8573800576771826,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003575996689683344,
      "loss": 5.32,
      "mean_token_accuracy": 0.20017373263835908,
      "num_tokens": 89445661.0,
      "step": 44140
    },
    {
      "entropy": 5.7455367088317875,
      "epoch": 3.8578170060298875,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035757068281849747,
      "loss": 5.3624,
      "mean_token_accuracy": 0.19790048003196717,
      "num_tokens": 89455970.0,
      "step": 44145
    },
    {
      "entropy": 5.806770896911621,
      "epoch": 3.858253954382592,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003575416950849115,
      "loss": 5.3382,
      "mean_token_accuracy": 0.19963753074407578,
      "num_tokens": 89466076.0,
      "step": 44150
    },
    {
      "entropy": 5.801160287857056,
      "epoch": 3.8586909027352965,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003575127057681326,
      "loss": 5.3053,
      "mean_token_accuracy": 0.20276542752981186,
      "num_tokens": 89475525.0,
      "step": 44155
    },
    {
      "entropy": 5.8001954555511475,
      "epoch": 3.8591278510880014,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003574837148687168,
      "loss": 5.3655,
      "mean_token_accuracy": 0.1954528346657753,
      "num_tokens": 89485913.0,
      "step": 44160
    },
    {
      "entropy": 5.767266798019409,
      "epoch": 3.8595647994407063,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035745472238722013,
      "loss": 5.2984,
      "mean_token_accuracy": 0.1972101539373398,
      "num_tokens": 89496321.0,
      "step": 44165
    },
    {
      "entropy": 5.768928575515747,
      "epoch": 3.8600017477934108,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035742572832419867,
      "loss": 5.3085,
      "mean_token_accuracy": 0.19912731796503066,
      "num_tokens": 89506335.0,
      "step": 44170
    },
    {
      "entropy": 5.798018455505371,
      "epoch": 3.8604386961461157,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035739673268020856,
      "loss": 5.3002,
      "mean_token_accuracy": 0.20131827443838118,
      "num_tokens": 89515683.0,
      "step": 44175
    },
    {
      "entropy": 5.72515082359314,
      "epoch": 3.86087564449882,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035736773545580594,
      "loss": 5.3368,
      "mean_token_accuracy": 0.2032852753996849,
      "num_tokens": 89525797.0,
      "step": 44180
    },
    {
      "entropy": 5.851973533630371,
      "epoch": 3.861312592851525,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035733873665154696,
      "loss": 5.3588,
      "mean_token_accuracy": 0.1972564399242401,
      "num_tokens": 89535312.0,
      "step": 44185
    },
    {
      "entropy": 5.682788276672364,
      "epoch": 3.8617495412042295,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003573097362679879,
      "loss": 5.1751,
      "mean_token_accuracy": 0.221379354596138,
      "num_tokens": 89545498.0,
      "step": 44190
    },
    {
      "entropy": 5.732202339172363,
      "epoch": 3.8621864895569344,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003572807343056849,
      "loss": 5.217,
      "mean_token_accuracy": 0.2103007659316063,
      "num_tokens": 89555421.0,
      "step": 44195
    },
    {
      "entropy": 5.810333967208862,
      "epoch": 3.8626234379096394,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003572517307651944,
      "loss": 5.2924,
      "mean_token_accuracy": 0.20620867758989334,
      "num_tokens": 89565549.0,
      "step": 44200
    },
    {
      "entropy": 5.787077188491821,
      "epoch": 3.863060386262344,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003572227256470725,
      "loss": 5.2891,
      "mean_token_accuracy": 0.2058503434062004,
      "num_tokens": 89575106.0,
      "step": 44205
    },
    {
      "entropy": 5.789406967163086,
      "epoch": 3.8634973346150483,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035719371895187574,
      "loss": 5.2525,
      "mean_token_accuracy": 0.19872958362102508,
      "num_tokens": 89584000.0,
      "step": 44210
    },
    {
      "entropy": 5.652596426010132,
      "epoch": 3.863934282967753,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003571647106801604,
      "loss": 5.289,
      "mean_token_accuracy": 0.20977911502122878,
      "num_tokens": 89594557.0,
      "step": 44215
    },
    {
      "entropy": 5.744245529174805,
      "epoch": 3.864371231320458,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003571357008324828,
      "loss": 5.2577,
      "mean_token_accuracy": 0.20519905984401704,
      "num_tokens": 89603710.0,
      "step": 44220
    },
    {
      "entropy": 5.805480432510376,
      "epoch": 3.8648081796731626,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035710668940939935,
      "loss": 5.3484,
      "mean_token_accuracy": 0.19278308898210525,
      "num_tokens": 89614880.0,
      "step": 44225
    },
    {
      "entropy": 5.830676651000976,
      "epoch": 3.8652451280258675,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035707767641146663,
      "loss": 5.3469,
      "mean_token_accuracy": 0.19800069630146028,
      "num_tokens": 89624377.0,
      "step": 44230
    },
    {
      "entropy": 5.744789505004883,
      "epoch": 3.865682076378572,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003570486618392411,
      "loss": 5.3294,
      "mean_token_accuracy": 0.19390667080879212,
      "num_tokens": 89634260.0,
      "step": 44235
    },
    {
      "entropy": 5.7684776306152346,
      "epoch": 3.866119024731277,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035701964569327926,
      "loss": 5.2821,
      "mean_token_accuracy": 0.19694947600364685,
      "num_tokens": 89643642.0,
      "step": 44240
    },
    {
      "entropy": 5.847441864013672,
      "epoch": 3.8665559730839814,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035699062797413754,
      "loss": 5.3576,
      "mean_token_accuracy": 0.19543086141347885,
      "num_tokens": 89653461.0,
      "step": 44245
    },
    {
      "entropy": 5.7901934623718265,
      "epoch": 3.8669929214366863,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003569616086823727,
      "loss": 5.3609,
      "mean_token_accuracy": 0.19422098845243455,
      "num_tokens": 89663108.0,
      "step": 44250
    },
    {
      "entropy": 5.844568061828613,
      "epoch": 3.8674298697893907,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003569325878185412,
      "loss": 5.3605,
      "mean_token_accuracy": 0.18913789242506027,
      "num_tokens": 89673682.0,
      "step": 44255
    },
    {
      "entropy": 5.842841291427613,
      "epoch": 3.8678668181420957,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003569035653831998,
      "loss": 5.3727,
      "mean_token_accuracy": 0.19760761708021163,
      "num_tokens": 89684795.0,
      "step": 44260
    },
    {
      "entropy": 5.811566352844238,
      "epoch": 3.8683037664948,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035687454137690514,
      "loss": 5.3292,
      "mean_token_accuracy": 0.195974200963974,
      "num_tokens": 89695134.0,
      "step": 44265
    },
    {
      "entropy": 5.765126085281372,
      "epoch": 3.868740714847505,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035684551580021374,
      "loss": 5.3121,
      "mean_token_accuracy": 0.20230402201414108,
      "num_tokens": 89705087.0,
      "step": 44270
    },
    {
      "entropy": 5.787012434005737,
      "epoch": 3.86917766320021,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035681648865368255,
      "loss": 5.2417,
      "mean_token_accuracy": 0.2035568028688431,
      "num_tokens": 89714920.0,
      "step": 44275
    },
    {
      "entropy": 5.78746337890625,
      "epoch": 3.8696146115529144,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035678745993786813,
      "loss": 5.2479,
      "mean_token_accuracy": 0.2039797693490982,
      "num_tokens": 89724022.0,
      "step": 44280
    },
    {
      "entropy": 5.791763639450073,
      "epoch": 3.870051559905619,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003567584296533275,
      "loss": 5.274,
      "mean_token_accuracy": 0.2023337259888649,
      "num_tokens": 89734179.0,
      "step": 44285
    },
    {
      "entropy": 5.7524830341339115,
      "epoch": 3.870488508258324,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003567293978006173,
      "loss": 5.2923,
      "mean_token_accuracy": 0.20503948032855987,
      "num_tokens": 89743810.0,
      "step": 44290
    },
    {
      "entropy": 5.7893674850463865,
      "epoch": 3.8709254566110287,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003567003643802944,
      "loss": 5.3225,
      "mean_token_accuracy": 0.20733165591955185,
      "num_tokens": 89753578.0,
      "step": 44295
    },
    {
      "entropy": 5.73199143409729,
      "epoch": 3.871362404963733,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035667132939291575,
      "loss": 5.248,
      "mean_token_accuracy": 0.20425612181425096,
      "num_tokens": 89763569.0,
      "step": 44300
    },
    {
      "entropy": 5.795423460006714,
      "epoch": 3.871799353316438,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003566422928390381,
      "loss": 5.3163,
      "mean_token_accuracy": 0.19943471252918243,
      "num_tokens": 89772759.0,
      "step": 44305
    },
    {
      "entropy": 5.702457237243652,
      "epoch": 3.8722363016691426,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003566132547192187,
      "loss": 5.2072,
      "mean_token_accuracy": 0.20952911376953126,
      "num_tokens": 89782499.0,
      "step": 44310
    },
    {
      "entropy": 5.750640821456909,
      "epoch": 3.8726732500218475,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035658421503401415,
      "loss": 5.3158,
      "mean_token_accuracy": 0.2020383134484291,
      "num_tokens": 89792831.0,
      "step": 44315
    },
    {
      "entropy": 5.79312162399292,
      "epoch": 3.873110198374552,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003565551737839816,
      "loss": 5.3009,
      "mean_token_accuracy": 0.20579757392406464,
      "num_tokens": 89802411.0,
      "step": 44320
    },
    {
      "entropy": 5.749050855636597,
      "epoch": 3.873547146727257,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003565261309696781,
      "loss": 5.3267,
      "mean_token_accuracy": 0.1958991214632988,
      "num_tokens": 89812157.0,
      "step": 44325
    },
    {
      "entropy": 5.732760620117188,
      "epoch": 3.8739840950799618,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003564970865916608,
      "loss": 5.2198,
      "mean_token_accuracy": 0.21921597719192504,
      "num_tokens": 89821849.0,
      "step": 44330
    },
    {
      "entropy": 5.743817853927612,
      "epoch": 3.8744210434326662,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035646804065048653,
      "loss": 5.2942,
      "mean_token_accuracy": 0.1980844810605049,
      "num_tokens": 89830993.0,
      "step": 44335
    },
    {
      "entropy": 5.717347812652588,
      "epoch": 3.8748579917853707,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035643899314671265,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20714132338762284,
      "num_tokens": 89841418.0,
      "step": 44340
    },
    {
      "entropy": 5.733987617492676,
      "epoch": 3.8752949401380756,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035640994408089616,
      "loss": 5.2569,
      "mean_token_accuracy": 0.21455292999744416,
      "num_tokens": 89851994.0,
      "step": 44345
    },
    {
      "entropy": 5.758320045471192,
      "epoch": 3.8757318884907805,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035638089345359426,
      "loss": 5.2925,
      "mean_token_accuracy": 0.20557199269533158,
      "num_tokens": 89861127.0,
      "step": 44350
    },
    {
      "entropy": 5.813248491287231,
      "epoch": 3.876168836843485,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035635184126536426,
      "loss": 5.3409,
      "mean_token_accuracy": 0.1981983095407486,
      "num_tokens": 89872346.0,
      "step": 44355
    },
    {
      "entropy": 5.843167066574097,
      "epoch": 3.87660578519619,
      "grad_norm": 1.0,
      "learning_rate": 0.0003563227875167633,
      "loss": 5.3631,
      "mean_token_accuracy": 0.19783878177404404,
      "num_tokens": 89882223.0,
      "step": 44360
    },
    {
      "entropy": 5.784473323822022,
      "epoch": 3.8770427335488944,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035629373220834857,
      "loss": 5.3105,
      "mean_token_accuracy": 0.19825534969568254,
      "num_tokens": 89892305.0,
      "step": 44365
    },
    {
      "entropy": 5.756222295761108,
      "epoch": 3.8774796819015993,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003562646753406775,
      "loss": 5.2626,
      "mean_token_accuracy": 0.2076100841164589,
      "num_tokens": 89901547.0,
      "step": 44370
    },
    {
      "entropy": 5.757818794250488,
      "epoch": 3.8779166302543038,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035623561691430734,
      "loss": 5.2943,
      "mean_token_accuracy": 0.20465860664844512,
      "num_tokens": 89912686.0,
      "step": 44375
    },
    {
      "entropy": 5.8225672245025635,
      "epoch": 3.8783535786070087,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003562065569297956,
      "loss": 5.314,
      "mean_token_accuracy": 0.202740316092968,
      "num_tokens": 89921728.0,
      "step": 44380
    },
    {
      "entropy": 5.740122938156128,
      "epoch": 3.8787905269597136,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003561774953876995,
      "loss": 5.2483,
      "mean_token_accuracy": 0.2067575126886368,
      "num_tokens": 89931741.0,
      "step": 44385
    },
    {
      "entropy": 5.748800897598267,
      "epoch": 3.879227475312418,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035614843228857644,
      "loss": 5.2773,
      "mean_token_accuracy": 0.20042144060134887,
      "num_tokens": 89941323.0,
      "step": 44390
    },
    {
      "entropy": 5.76639142036438,
      "epoch": 3.8796644236651225,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000356119367632984,
      "loss": 5.3583,
      "mean_token_accuracy": 0.19824391156435012,
      "num_tokens": 89950763.0,
      "step": 44395
    },
    {
      "entropy": 5.7875697135925295,
      "epoch": 3.8801013720178275,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003560903014214795,
      "loss": 5.3744,
      "mean_token_accuracy": 0.19384656697511674,
      "num_tokens": 89961371.0,
      "step": 44400
    },
    {
      "entropy": 5.855278539657593,
      "epoch": 3.8805383203705324,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003560612336546206,
      "loss": 5.3804,
      "mean_token_accuracy": 0.1973780944943428,
      "num_tokens": 89971880.0,
      "step": 44405
    },
    {
      "entropy": 5.782094144821167,
      "epoch": 3.880975268723237,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003560321643329648,
      "loss": 5.2358,
      "mean_token_accuracy": 0.20630068480968475,
      "num_tokens": 89982040.0,
      "step": 44410
    },
    {
      "entropy": 5.820514106750489,
      "epoch": 3.8814122170759418,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003560030934570696,
      "loss": 5.3641,
      "mean_token_accuracy": 0.19363474994897842,
      "num_tokens": 89992517.0,
      "step": 44415
    },
    {
      "entropy": 5.7590454578399655,
      "epoch": 3.881849165428646,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003559740210274925,
      "loss": 5.2781,
      "mean_token_accuracy": 0.20299446880817412,
      "num_tokens": 90002432.0,
      "step": 44420
    },
    {
      "entropy": 5.7032225131988525,
      "epoch": 3.882286113781351,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003559449470447914,
      "loss": 5.1937,
      "mean_token_accuracy": 0.2121090844273567,
      "num_tokens": 90012556.0,
      "step": 44425
    },
    {
      "entropy": 5.817321348190307,
      "epoch": 3.8827230621340556,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003559158715095238,
      "loss": 5.3651,
      "mean_token_accuracy": 0.1994666188955307,
      "num_tokens": 90022513.0,
      "step": 44430
    },
    {
      "entropy": 5.764922142028809,
      "epoch": 3.8831600104867605,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003558867944222473,
      "loss": 5.2911,
      "mean_token_accuracy": 0.20353008210659027,
      "num_tokens": 90032822.0,
      "step": 44435
    },
    {
      "entropy": 5.794762849807739,
      "epoch": 3.8835969588394654,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003558577157835197,
      "loss": 5.2584,
      "mean_token_accuracy": 0.20344043672084808,
      "num_tokens": 90042685.0,
      "step": 44440
    },
    {
      "entropy": 5.741289663314819,
      "epoch": 3.88403390719217,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003558286355938988,
      "loss": 5.3099,
      "mean_token_accuracy": 0.2080145701766014,
      "num_tokens": 90053118.0,
      "step": 44445
    },
    {
      "entropy": 5.693419885635376,
      "epoch": 3.8844708555448744,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003557995538539423,
      "loss": 5.2764,
      "mean_token_accuracy": 0.20242878943681716,
      "num_tokens": 90063041.0,
      "step": 44450
    },
    {
      "entropy": 5.833025979995727,
      "epoch": 3.8849078038975793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000355770470564208,
      "loss": 5.3347,
      "mean_token_accuracy": 0.20132698118686676,
      "num_tokens": 90072995.0,
      "step": 44455
    },
    {
      "entropy": 5.803652048110962,
      "epoch": 3.885344752250284,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003557413857252538,
      "loss": 5.2967,
      "mean_token_accuracy": 0.19880176037549974,
      "num_tokens": 90084369.0,
      "step": 44460
    },
    {
      "entropy": 5.801349782943726,
      "epoch": 3.8857817006029887,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003557122993376374,
      "loss": 5.3505,
      "mean_token_accuracy": 0.19363668113946914,
      "num_tokens": 90093207.0,
      "step": 44465
    },
    {
      "entropy": 5.733662223815918,
      "epoch": 3.8862186489556936,
      "grad_norm": 1.125,
      "learning_rate": 0.0003556832114019169,
      "loss": 5.2783,
      "mean_token_accuracy": 0.20627183020114898,
      "num_tokens": 90102452.0,
      "step": 44470
    },
    {
      "entropy": 5.7662118434906,
      "epoch": 3.886655597308398,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035565412191865015,
      "loss": 5.2667,
      "mean_token_accuracy": 0.20157750993967055,
      "num_tokens": 90112200.0,
      "step": 44475
    },
    {
      "entropy": 5.7596629619598385,
      "epoch": 3.887092545661103,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035562503088839496,
      "loss": 5.3446,
      "mean_token_accuracy": 0.1977856144309044,
      "num_tokens": 90122627.0,
      "step": 44480
    },
    {
      "entropy": 5.787665557861328,
      "epoch": 3.8875294940138074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003555959383117096,
      "loss": 5.2994,
      "mean_token_accuracy": 0.19549846947193145,
      "num_tokens": 90131303.0,
      "step": 44485
    },
    {
      "entropy": 5.784255599975586,
      "epoch": 3.8879664423665123,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003555668441891518,
      "loss": 5.2512,
      "mean_token_accuracy": 0.20236459523439407,
      "num_tokens": 90140904.0,
      "step": 44490
    },
    {
      "entropy": 5.7960748195648195,
      "epoch": 3.8884033907192173,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003555377485212797,
      "loss": 5.2548,
      "mean_token_accuracy": 0.19871461242437363,
      "num_tokens": 90150729.0,
      "step": 44495
    },
    {
      "entropy": 5.836073923110962,
      "epoch": 3.8888403390719217,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003555086513086514,
      "loss": 5.3557,
      "mean_token_accuracy": 0.1900378406047821,
      "num_tokens": 90160832.0,
      "step": 44500
    },
    {
      "entropy": 5.739452648162842,
      "epoch": 3.889277287424626,
      "grad_norm": 1.046875,
      "learning_rate": 0.000355479552551825,
      "loss": 5.2958,
      "mean_token_accuracy": 0.20473343580961229,
      "num_tokens": 90171606.0,
      "step": 44505
    },
    {
      "entropy": 5.689343881607056,
      "epoch": 3.889714235777331,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035545045225135864,
      "loss": 5.2552,
      "mean_token_accuracy": 0.20890145301818847,
      "num_tokens": 90181902.0,
      "step": 44510
    },
    {
      "entropy": 5.792512702941894,
      "epoch": 3.890151184130036,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003554213504078104,
      "loss": 5.3656,
      "mean_token_accuracy": 0.19871629923582076,
      "num_tokens": 90192380.0,
      "step": 44515
    },
    {
      "entropy": 5.757389831542969,
      "epoch": 3.8905881324827405,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003553922470217385,
      "loss": 5.2529,
      "mean_token_accuracy": 0.2105625331401825,
      "num_tokens": 90202469.0,
      "step": 44520
    },
    {
      "entropy": 5.813348388671875,
      "epoch": 3.8910250808354454,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035536314209370126,
      "loss": 5.2797,
      "mean_token_accuracy": 0.20944714844226836,
      "num_tokens": 90212706.0,
      "step": 44525
    },
    {
      "entropy": 5.754153680801392,
      "epoch": 3.89146202918815,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035533403562425676,
      "loss": 5.2891,
      "mean_token_accuracy": 0.2008061096072197,
      "num_tokens": 90222456.0,
      "step": 44530
    },
    {
      "entropy": 5.730535411834717,
      "epoch": 3.891898977540855,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035530492761396337,
      "loss": 5.2578,
      "mean_token_accuracy": 0.20479404777288437,
      "num_tokens": 90232960.0,
      "step": 44535
    },
    {
      "entropy": 5.738002109527588,
      "epoch": 3.8923359258935593,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00035527581806337946,
      "loss": 5.2661,
      "mean_token_accuracy": 0.2038431316614151,
      "num_tokens": 90242407.0,
      "step": 44540
    },
    {
      "entropy": 5.7808324813842775,
      "epoch": 3.892772874246264,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035524670697306324,
      "loss": 5.275,
      "mean_token_accuracy": 0.2038654178380966,
      "num_tokens": 90252211.0,
      "step": 44545
    },
    {
      "entropy": 5.767840337753296,
      "epoch": 3.8932098225989686,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003552175943435731,
      "loss": 5.3133,
      "mean_token_accuracy": 0.20213933736085893,
      "num_tokens": 90262047.0,
      "step": 44550
    },
    {
      "entropy": 5.720212411880493,
      "epoch": 3.8936467709516736,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035518848017546757,
      "loss": 5.2897,
      "mean_token_accuracy": 0.19888924360275267,
      "num_tokens": 90272870.0,
      "step": 44555
    },
    {
      "entropy": 5.7787047863006595,
      "epoch": 3.894083719304378,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003551593644693049,
      "loss": 5.3848,
      "mean_token_accuracy": 0.19126367270946504,
      "num_tokens": 90282581.0,
      "step": 44560
    },
    {
      "entropy": 5.816160011291504,
      "epoch": 3.894520667657083,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035513024722564365,
      "loss": 5.3292,
      "mean_token_accuracy": 0.20322499871253968,
      "num_tokens": 90294100.0,
      "step": 44565
    },
    {
      "entropy": 5.785943078994751,
      "epoch": 3.894957616009788,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035510112844504223,
      "loss": 5.278,
      "mean_token_accuracy": 0.20771701782941818,
      "num_tokens": 90305106.0,
      "step": 44570
    },
    {
      "entropy": 5.874212694168091,
      "epoch": 3.8953945643624923,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003550720081280592,
      "loss": 5.2959,
      "mean_token_accuracy": 0.2013613164424896,
      "num_tokens": 90315246.0,
      "step": 44575
    },
    {
      "entropy": 5.7664680004119875,
      "epoch": 3.895831512715197,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035504288627525313,
      "loss": 5.3074,
      "mean_token_accuracy": 0.20076515674591064,
      "num_tokens": 90324312.0,
      "step": 44580
    },
    {
      "entropy": 5.740685892105103,
      "epoch": 3.8962684610679017,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003550137628871825,
      "loss": 5.2347,
      "mean_token_accuracy": 0.21147608757019043,
      "num_tokens": 90333639.0,
      "step": 44585
    },
    {
      "entropy": 5.668576717376709,
      "epoch": 3.8967054094206066,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035498463796440605,
      "loss": 5.2197,
      "mean_token_accuracy": 0.21101531237363816,
      "num_tokens": 90343332.0,
      "step": 44590
    },
    {
      "entropy": 5.705508089065551,
      "epoch": 3.897142357773311,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035495551150748226,
      "loss": 5.2627,
      "mean_token_accuracy": 0.20598532110452653,
      "num_tokens": 90355337.0,
      "step": 44595
    },
    {
      "entropy": 5.830523872375489,
      "epoch": 3.897579306126016,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035492638351697,
      "loss": 5.3903,
      "mean_token_accuracy": 0.19446903169155122,
      "num_tokens": 90367126.0,
      "step": 44600
    },
    {
      "entropy": 5.777913379669189,
      "epoch": 3.8980162544787205,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003548972539934277,
      "loss": 5.3119,
      "mean_token_accuracy": 0.20196169316768647,
      "num_tokens": 90377408.0,
      "step": 44605
    },
    {
      "entropy": 5.813974046707154,
      "epoch": 3.8984532028314254,
      "grad_norm": 1.0,
      "learning_rate": 0.0003548681229374142,
      "loss": 5.2928,
      "mean_token_accuracy": 0.2041115790605545,
      "num_tokens": 90388451.0,
      "step": 44610
    },
    {
      "entropy": 5.786031484603882,
      "epoch": 3.89889015118413,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003548389903494883,
      "loss": 5.3028,
      "mean_token_accuracy": 0.2006497025489807,
      "num_tokens": 90399673.0,
      "step": 44615
    },
    {
      "entropy": 5.750596809387207,
      "epoch": 3.8993270995368348,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035480985623020873,
      "loss": 5.2371,
      "mean_token_accuracy": 0.20481122583150863,
      "num_tokens": 90409567.0,
      "step": 44620
    },
    {
      "entropy": 5.785013484954834,
      "epoch": 3.8997640478895397,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003547807205801342,
      "loss": 5.3734,
      "mean_token_accuracy": 0.1979500398039818,
      "num_tokens": 90419904.0,
      "step": 44625
    },
    {
      "entropy": 5.781566762924195,
      "epoch": 3.900200996242244,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003547515833998237,
      "loss": 5.2706,
      "mean_token_accuracy": 0.20069175362586975,
      "num_tokens": 90431445.0,
      "step": 44630
    },
    {
      "entropy": 5.783764266967774,
      "epoch": 3.9006379445949486,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003547224446898361,
      "loss": 5.2914,
      "mean_token_accuracy": 0.2034589469432831,
      "num_tokens": 90440933.0,
      "step": 44635
    },
    {
      "entropy": 5.75237340927124,
      "epoch": 3.9010748929476535,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035469330445073025,
      "loss": 5.3038,
      "mean_token_accuracy": 0.20448451340198517,
      "num_tokens": 90451145.0,
      "step": 44640
    },
    {
      "entropy": 5.804770231246948,
      "epoch": 3.9015118413003584,
      "grad_norm": 1.078125,
      "learning_rate": 0.000354664162683065,
      "loss": 5.3808,
      "mean_token_accuracy": 0.19289744645357132,
      "num_tokens": 90460355.0,
      "step": 44645
    },
    {
      "entropy": 5.825044345855713,
      "epoch": 3.901948789653063,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035463501938739936,
      "loss": 5.2868,
      "mean_token_accuracy": 0.20317529439926146,
      "num_tokens": 90471162.0,
      "step": 44650
    },
    {
      "entropy": 5.785550165176391,
      "epoch": 3.902385738005768,
      "grad_norm": 1.0,
      "learning_rate": 0.0003546058745642923,
      "loss": 5.3411,
      "mean_token_accuracy": 0.19733530879020691,
      "num_tokens": 90481807.0,
      "step": 44655
    },
    {
      "entropy": 5.747081661224366,
      "epoch": 3.9028226863584723,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035457672821430293,
      "loss": 5.2327,
      "mean_token_accuracy": 0.20376311093568802,
      "num_tokens": 90491478.0,
      "step": 44660
    },
    {
      "entropy": 5.658943176269531,
      "epoch": 3.903259634711177,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003545475803379902,
      "loss": 5.1828,
      "mean_token_accuracy": 0.21417573243379592,
      "num_tokens": 90501231.0,
      "step": 44665
    },
    {
      "entropy": 5.721306991577149,
      "epoch": 3.9036965830638817,
      "grad_norm": 1.0,
      "learning_rate": 0.00035451843093591316,
      "loss": 5.3518,
      "mean_token_accuracy": 0.19810845851898193,
      "num_tokens": 90512021.0,
      "step": 44670
    },
    {
      "entropy": 5.837132978439331,
      "epoch": 3.9041335314165866,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035448928000863096,
      "loss": 5.3839,
      "mean_token_accuracy": 0.20266618579626083,
      "num_tokens": 90522210.0,
      "step": 44675
    },
    {
      "entropy": 5.8622101783752445,
      "epoch": 3.9045704797692915,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003544601275567027,
      "loss": 5.3458,
      "mean_token_accuracy": 0.20199940502643585,
      "num_tokens": 90533024.0,
      "step": 44680
    },
    {
      "entropy": 5.6781693458557125,
      "epoch": 3.905007428121996,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003544309735806876,
      "loss": 5.264,
      "mean_token_accuracy": 0.20944373011589051,
      "num_tokens": 90541965.0,
      "step": 44685
    },
    {
      "entropy": 5.769038105010987,
      "epoch": 3.9054443764747004,
      "grad_norm": 1.125,
      "learning_rate": 0.0003544018180811448,
      "loss": 5.2456,
      "mean_token_accuracy": 0.2050122320652008,
      "num_tokens": 90552167.0,
      "step": 44690
    },
    {
      "entropy": 5.8055976867675785,
      "epoch": 3.9058813248274054,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003543726610586335,
      "loss": 5.2332,
      "mean_token_accuracy": 0.20170921832323074,
      "num_tokens": 90562383.0,
      "step": 44695
    },
    {
      "entropy": 5.8089948177337645,
      "epoch": 3.9063182731801103,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035434350251371295,
      "loss": 5.3038,
      "mean_token_accuracy": 0.2051260933279991,
      "num_tokens": 90572609.0,
      "step": 44700
    },
    {
      "entropy": 5.694352865219116,
      "epoch": 3.9067552215328147,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003543143424469425,
      "loss": 5.2938,
      "mean_token_accuracy": 0.20934423953294753,
      "num_tokens": 90582319.0,
      "step": 44705
    },
    {
      "entropy": 5.782286500930786,
      "epoch": 3.9071921698855197,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035428518085888134,
      "loss": 5.3261,
      "mean_token_accuracy": 0.20065145641565324,
      "num_tokens": 90592493.0,
      "step": 44710
    },
    {
      "entropy": 5.87460298538208,
      "epoch": 3.907629118238224,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035425601775008895,
      "loss": 5.3969,
      "mean_token_accuracy": 0.19753197878599166,
      "num_tokens": 90603504.0,
      "step": 44715
    },
    {
      "entropy": 5.841928195953369,
      "epoch": 3.908066066590929,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003542268531211245,
      "loss": 5.3628,
      "mean_token_accuracy": 0.1962682917714119,
      "num_tokens": 90613957.0,
      "step": 44720
    },
    {
      "entropy": 5.765921497344971,
      "epoch": 3.9085030149436335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003541976869725476,
      "loss": 5.269,
      "mean_token_accuracy": 0.20226182788610458,
      "num_tokens": 90623986.0,
      "step": 44725
    },
    {
      "entropy": 5.821413373947143,
      "epoch": 3.9089399632963384,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003541685193049175,
      "loss": 5.302,
      "mean_token_accuracy": 0.20526386946439742,
      "num_tokens": 90633879.0,
      "step": 44730
    },
    {
      "entropy": 5.793061637878418,
      "epoch": 3.9093769116490433,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003541393501187937,
      "loss": 5.324,
      "mean_token_accuracy": 0.1923380821943283,
      "num_tokens": 90644031.0,
      "step": 44735
    },
    {
      "entropy": 5.7963244915008545,
      "epoch": 3.909813860001748,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035411017941473576,
      "loss": 5.3308,
      "mean_token_accuracy": 0.202004337310791,
      "num_tokens": 90654550.0,
      "step": 44740
    },
    {
      "entropy": 5.777440309524536,
      "epoch": 3.9102508083544523,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035408100719330304,
      "loss": 5.3272,
      "mean_token_accuracy": 0.19346091002225876,
      "num_tokens": 90664117.0,
      "step": 44745
    },
    {
      "entropy": 5.815344619750976,
      "epoch": 3.910687756707157,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003540518334550552,
      "loss": 5.3349,
      "mean_token_accuracy": 0.19274982064962387,
      "num_tokens": 90674400.0,
      "step": 44750
    },
    {
      "entropy": 5.849542474746704,
      "epoch": 3.911124705059862,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035402265820055185,
      "loss": 5.349,
      "mean_token_accuracy": 0.19771458059549332,
      "num_tokens": 90684759.0,
      "step": 44755
    },
    {
      "entropy": 5.8635632514953615,
      "epoch": 3.9115616534125666,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035399348143035244,
      "loss": 5.3639,
      "mean_token_accuracy": 0.1958973988890648,
      "num_tokens": 90694346.0,
      "step": 44760
    },
    {
      "entropy": 5.778202390670776,
      "epoch": 3.9119986017652715,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035396430314501665,
      "loss": 5.3418,
      "mean_token_accuracy": 0.19664979726076126,
      "num_tokens": 90704239.0,
      "step": 44765
    },
    {
      "entropy": 5.752813673019409,
      "epoch": 3.912435550117976,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003539351233451042,
      "loss": 5.2853,
      "mean_token_accuracy": 0.20042333751916885,
      "num_tokens": 90714240.0,
      "step": 44770
    },
    {
      "entropy": 5.8132524490356445,
      "epoch": 3.912872498470681,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003539059420311747,
      "loss": 5.3421,
      "mean_token_accuracy": 0.19202695190906524,
      "num_tokens": 90724575.0,
      "step": 44775
    },
    {
      "entropy": 5.829574108123779,
      "epoch": 3.9133094468233853,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003538767592037879,
      "loss": 5.3185,
      "mean_token_accuracy": 0.19889962375164033,
      "num_tokens": 90735217.0,
      "step": 44780
    },
    {
      "entropy": 5.761868906021118,
      "epoch": 3.9137463951760902,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003538475748635035,
      "loss": 5.2991,
      "mean_token_accuracy": 0.20439224541187287,
      "num_tokens": 90745437.0,
      "step": 44785
    },
    {
      "entropy": 5.787009382247925,
      "epoch": 3.914183343528795,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003538183890108813,
      "loss": 5.3048,
      "mean_token_accuracy": 0.2013639658689499,
      "num_tokens": 90755286.0,
      "step": 44790
    },
    {
      "entropy": 5.782011556625366,
      "epoch": 3.9146202918814996,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003537892016464811,
      "loss": 5.3465,
      "mean_token_accuracy": 0.19358765184879304,
      "num_tokens": 90764734.0,
      "step": 44795
    },
    {
      "entropy": 5.833046865463257,
      "epoch": 3.915057240234204,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035376001277086276,
      "loss": 5.321,
      "mean_token_accuracy": 0.19723445922136307,
      "num_tokens": 90774569.0,
      "step": 44800
    },
    {
      "entropy": 5.865460968017578,
      "epoch": 3.915494188586909,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003537308223845861,
      "loss": 5.3789,
      "mean_token_accuracy": 0.1907857671380043,
      "num_tokens": 90785029.0,
      "step": 44805
    },
    {
      "entropy": 5.857933139801025,
      "epoch": 3.915931136939614,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035370163048821095,
      "loss": 5.3631,
      "mean_token_accuracy": 0.1980792313814163,
      "num_tokens": 90795382.0,
      "step": 44810
    },
    {
      "entropy": 5.736408042907715,
      "epoch": 3.9163680852923184,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035367243708229736,
      "loss": 5.2799,
      "mean_token_accuracy": 0.21022080332040788,
      "num_tokens": 90805354.0,
      "step": 44815
    },
    {
      "entropy": 5.779200601577759,
      "epoch": 3.9168050336450233,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035364324216740516,
      "loss": 5.3486,
      "mean_token_accuracy": 0.19536848366260529,
      "num_tokens": 90815967.0,
      "step": 44820
    },
    {
      "entropy": 5.836626386642456,
      "epoch": 3.9172419819977278,
      "grad_norm": 0.96875,
      "learning_rate": 0.00035361404574409436,
      "loss": 5.328,
      "mean_token_accuracy": 0.19918019622564315,
      "num_tokens": 90826717.0,
      "step": 44825
    },
    {
      "entropy": 5.765891075134277,
      "epoch": 3.9176789303504327,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000353584847812925,
      "loss": 5.2266,
      "mean_token_accuracy": 0.20849977135658265,
      "num_tokens": 90836592.0,
      "step": 44830
    },
    {
      "entropy": 5.759283638000488,
      "epoch": 3.918115878703137,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035355564837445707,
      "loss": 5.3429,
      "mean_token_accuracy": 0.19554296582937242,
      "num_tokens": 90847323.0,
      "step": 44835
    },
    {
      "entropy": 5.82909927368164,
      "epoch": 3.918552827055842,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035352644742925066,
      "loss": 5.4047,
      "mean_token_accuracy": 0.1889069512486458,
      "num_tokens": 90858471.0,
      "step": 44840
    },
    {
      "entropy": 5.822413969039917,
      "epoch": 3.918989775408547,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035349724497786577,
      "loss": 5.3522,
      "mean_token_accuracy": 0.19822238385677338,
      "num_tokens": 90868983.0,
      "step": 44845
    },
    {
      "entropy": 5.810620403289795,
      "epoch": 3.9194267237612515,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035346804102086266,
      "loss": 5.3152,
      "mean_token_accuracy": 0.19843723773956298,
      "num_tokens": 90879904.0,
      "step": 44850
    },
    {
      "entropy": 5.85993390083313,
      "epoch": 3.919863672113956,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035343883555880143,
      "loss": 5.3479,
      "mean_token_accuracy": 0.1952521249651909,
      "num_tokens": 90889923.0,
      "step": 44855
    },
    {
      "entropy": 5.831072759628296,
      "epoch": 3.920300620466661,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035340962859224217,
      "loss": 5.2988,
      "mean_token_accuracy": 0.19996020048856736,
      "num_tokens": 90899991.0,
      "step": 44860
    },
    {
      "entropy": 5.7228752136230465,
      "epoch": 3.9207375688193657,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003533804201217452,
      "loss": 5.2594,
      "mean_token_accuracy": 0.20164964497089385,
      "num_tokens": 90910359.0,
      "step": 44865
    },
    {
      "entropy": 5.741509675979614,
      "epoch": 3.92117451717207,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035335121014787063,
      "loss": 5.2636,
      "mean_token_accuracy": 0.20150666534900666,
      "num_tokens": 90920033.0,
      "step": 44870
    },
    {
      "entropy": 5.825690078735351,
      "epoch": 3.921611465524775,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003533219986711789,
      "loss": 5.3513,
      "mean_token_accuracy": 0.1977936863899231,
      "num_tokens": 90929301.0,
      "step": 44875
    },
    {
      "entropy": 5.779982328414917,
      "epoch": 3.9220484138774796,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003532927856922301,
      "loss": 5.2414,
      "mean_token_accuracy": 0.2073809489607811,
      "num_tokens": 90939197.0,
      "step": 44880
    },
    {
      "entropy": 5.80094723701477,
      "epoch": 3.9224853622301845,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035326357121158463,
      "loss": 5.3247,
      "mean_token_accuracy": 0.19850728362798692,
      "num_tokens": 90949013.0,
      "step": 44885
    },
    {
      "entropy": 5.763597822189331,
      "epoch": 3.922922310582889,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003532343552298029,
      "loss": 5.3099,
      "mean_token_accuracy": 0.1998721405863762,
      "num_tokens": 90960010.0,
      "step": 44890
    },
    {
      "entropy": 5.73783688545227,
      "epoch": 3.923359258935594,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003532051377474453,
      "loss": 5.3243,
      "mean_token_accuracy": 0.19933206737041473,
      "num_tokens": 90969768.0,
      "step": 44895
    },
    {
      "entropy": 5.752260541915893,
      "epoch": 3.9237962072882984,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003531759187650721,
      "loss": 5.307,
      "mean_token_accuracy": 0.20718273967504502,
      "num_tokens": 90980051.0,
      "step": 44900
    },
    {
      "entropy": 5.7951406955719,
      "epoch": 3.9242331556410033,
      "grad_norm": 1.125,
      "learning_rate": 0.0003531466982832438,
      "loss": 5.2789,
      "mean_token_accuracy": 0.20504388213157654,
      "num_tokens": 90990144.0,
      "step": 44905
    },
    {
      "entropy": 5.7146330833435055,
      "epoch": 3.9246701039937077,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003531174763025209,
      "loss": 5.1888,
      "mean_token_accuracy": 0.20830365568399428,
      "num_tokens": 91000353.0,
      "step": 44910
    },
    {
      "entropy": 5.780643701553345,
      "epoch": 3.9251070523464127,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003530882528234639,
      "loss": 5.2962,
      "mean_token_accuracy": 0.2043144404888153,
      "num_tokens": 91010819.0,
      "step": 44915
    },
    {
      "entropy": 5.830887222290039,
      "epoch": 3.9255440006991176,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035305902784663333,
      "loss": 5.3344,
      "mean_token_accuracy": 0.18989056050777436,
      "num_tokens": 91020454.0,
      "step": 44920
    },
    {
      "entropy": 5.827049922943115,
      "epoch": 3.925980949051822,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035302980137258964,
      "loss": 5.4167,
      "mean_token_accuracy": 0.1913527548313141,
      "num_tokens": 91030710.0,
      "step": 44925
    },
    {
      "entropy": 5.792471218109131,
      "epoch": 3.9264178974045265,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003530005734018935,
      "loss": 5.2967,
      "mean_token_accuracy": 0.2054697647690773,
      "num_tokens": 91039907.0,
      "step": 44930
    },
    {
      "entropy": 5.8232465267181395,
      "epoch": 3.9268548457572314,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035297134393510543,
      "loss": 5.3098,
      "mean_token_accuracy": 0.19589295834302903,
      "num_tokens": 91049377.0,
      "step": 44935
    },
    {
      "entropy": 5.772636795043946,
      "epoch": 3.9272917941099363,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003529421129727862,
      "loss": 5.3079,
      "mean_token_accuracy": 0.19708324819803238,
      "num_tokens": 91060341.0,
      "step": 44940
    },
    {
      "entropy": 5.784525346755982,
      "epoch": 3.927728742462641,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035291288051549644,
      "loss": 5.288,
      "mean_token_accuracy": 0.20320427268743516,
      "num_tokens": 91069737.0,
      "step": 44945
    },
    {
      "entropy": 5.794994306564331,
      "epoch": 3.9281656908153457,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035288364656379666,
      "loss": 5.3167,
      "mean_token_accuracy": 0.20180510431528093,
      "num_tokens": 91079290.0,
      "step": 44950
    },
    {
      "entropy": 5.738560581207276,
      "epoch": 3.92860263916805,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035285441111824784,
      "loss": 5.3002,
      "mean_token_accuracy": 0.20044557005167007,
      "num_tokens": 91090316.0,
      "step": 44955
    },
    {
      "entropy": 5.766907024383545,
      "epoch": 3.929039587520755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035282517417941056,
      "loss": 5.2676,
      "mean_token_accuracy": 0.2010502278804779,
      "num_tokens": 91100521.0,
      "step": 44960
    },
    {
      "entropy": 5.777335834503174,
      "epoch": 3.9294765358734596,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003527959357478457,
      "loss": 5.3218,
      "mean_token_accuracy": 0.20216814130544664,
      "num_tokens": 91110207.0,
      "step": 44965
    },
    {
      "entropy": 5.801444816589355,
      "epoch": 3.9299134842261645,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000352766695824114,
      "loss": 5.3358,
      "mean_token_accuracy": 0.1907573163509369,
      "num_tokens": 91120993.0,
      "step": 44970
    },
    {
      "entropy": 5.751843214035034,
      "epoch": 3.9303504325788694,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003527374544087764,
      "loss": 5.2505,
      "mean_token_accuracy": 0.202311410009861,
      "num_tokens": 91131769.0,
      "step": 44975
    },
    {
      "entropy": 5.795791816711426,
      "epoch": 3.930787380931574,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003527082115023936,
      "loss": 5.4057,
      "mean_token_accuracy": 0.19436436295509338,
      "num_tokens": 91141958.0,
      "step": 44980
    },
    {
      "entropy": 5.8211400508880615,
      "epoch": 3.9312243292842783,
      "grad_norm": 1.0,
      "learning_rate": 0.00035267896710552667,
      "loss": 5.283,
      "mean_token_accuracy": 0.20082185715436934,
      "num_tokens": 91151725.0,
      "step": 44985
    },
    {
      "entropy": 5.789218950271606,
      "epoch": 3.9316612776369833,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035264972121873633,
      "loss": 5.2835,
      "mean_token_accuracy": 0.2031555712223053,
      "num_tokens": 91161756.0,
      "step": 44990
    },
    {
      "entropy": 5.832932758331299,
      "epoch": 3.932098225989688,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035262047384258376,
      "loss": 5.3949,
      "mean_token_accuracy": 0.1929505616426468,
      "num_tokens": 91172160.0,
      "step": 44995
    },
    {
      "entropy": 5.829139947891235,
      "epoch": 3.9325351743423926,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003525912249776298,
      "loss": 5.3779,
      "mean_token_accuracy": 0.19059034287929535,
      "num_tokens": 91183030.0,
      "step": 45000
    },
    {
      "epoch": 3.9325351743423926,
      "eval_entropy": 5.567559715272645,
      "eval_loss": 5.362264156341553,
      "eval_mean_token_accuracy": 0.20625548067365607,
      "eval_num_tokens": 91183030.0,
      "eval_runtime": 28.3556,
      "eval_samples_per_second": 1305.35,
      "eval_steps_per_second": 163.178,
      "step": 45000
    },
    {
      "entropy": 5.722464847564697,
      "epoch": 3.9329721226950975,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035256197462443545,
      "loss": 5.2054,
      "mean_token_accuracy": 0.21801934540271758,
      "num_tokens": 91192590.0,
      "step": 45005
    },
    {
      "entropy": 5.753911733627319,
      "epoch": 3.933409071047802,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003525327227835618,
      "loss": 5.2476,
      "mean_token_accuracy": 0.20532068163156508,
      "num_tokens": 91202596.0,
      "step": 45010
    },
    {
      "entropy": 5.766079330444336,
      "epoch": 3.933846019400507,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035250346945556995,
      "loss": 5.3148,
      "mean_token_accuracy": 0.19739328771829606,
      "num_tokens": 91211922.0,
      "step": 45015
    },
    {
      "entropy": 5.767095518112183,
      "epoch": 3.9342829677532114,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003524742146410209,
      "loss": 5.3217,
      "mean_token_accuracy": 0.2027617335319519,
      "num_tokens": 91223325.0,
      "step": 45020
    },
    {
      "entropy": 5.781577730178833,
      "epoch": 3.9347199161059163,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003524449583404758,
      "loss": 5.3026,
      "mean_token_accuracy": 0.2043351948261261,
      "num_tokens": 91233404.0,
      "step": 45025
    },
    {
      "entropy": 5.759859037399292,
      "epoch": 3.9351568644586212,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035241570055449584,
      "loss": 5.3498,
      "mean_token_accuracy": 0.20083742290735246,
      "num_tokens": 91243816.0,
      "step": 45030
    },
    {
      "entropy": 5.846168756484985,
      "epoch": 3.9355938128113257,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00035238644128364217,
      "loss": 5.349,
      "mean_token_accuracy": 0.20126964896917343,
      "num_tokens": 91254023.0,
      "step": 45035
    },
    {
      "entropy": 5.768298482894897,
      "epoch": 3.93603076116403,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035235718052847604,
      "loss": 5.3036,
      "mean_token_accuracy": 0.20632355958223342,
      "num_tokens": 91264151.0,
      "step": 45040
    },
    {
      "entropy": 5.809586906433106,
      "epoch": 3.936467709516735,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035232791828955864,
      "loss": 5.2985,
      "mean_token_accuracy": 0.2028794139623642,
      "num_tokens": 91274088.0,
      "step": 45045
    },
    {
      "entropy": 5.624072504043579,
      "epoch": 3.93690465786944,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003522986545674512,
      "loss": 5.2497,
      "mean_token_accuracy": 0.21090161949396133,
      "num_tokens": 91284461.0,
      "step": 45050
    },
    {
      "entropy": 5.761558580398559,
      "epoch": 3.9373416062221445,
      "grad_norm": 1.25,
      "learning_rate": 0.0003522693893627151,
      "loss": 5.3168,
      "mean_token_accuracy": 0.2015551045536995,
      "num_tokens": 91293942.0,
      "step": 45055
    },
    {
      "entropy": 5.811024856567383,
      "epoch": 3.9377785545748494,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003522401226759116,
      "loss": 5.3495,
      "mean_token_accuracy": 0.19837386757135392,
      "num_tokens": 91305322.0,
      "step": 45060
    },
    {
      "entropy": 5.823342037200928,
      "epoch": 3.938215502927554,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003522108545076021,
      "loss": 5.3655,
      "mean_token_accuracy": 0.19441641718149186,
      "num_tokens": 91314947.0,
      "step": 45065
    },
    {
      "entropy": 5.819901609420777,
      "epoch": 3.9386524512802588,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000352181584858348,
      "loss": 5.301,
      "mean_token_accuracy": 0.19663951992988588,
      "num_tokens": 91324229.0,
      "step": 45070
    },
    {
      "entropy": 5.779207229614258,
      "epoch": 3.9390893996329632,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035215231372871054,
      "loss": 5.3258,
      "mean_token_accuracy": 0.1945930913090706,
      "num_tokens": 91334012.0,
      "step": 45075
    },
    {
      "entropy": 5.725587797164917,
      "epoch": 3.939526347985668,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003521230411192513,
      "loss": 5.2455,
      "mean_token_accuracy": 0.21118085980415344,
      "num_tokens": 91344067.0,
      "step": 45080
    },
    {
      "entropy": 5.667819166183472,
      "epoch": 3.939963296338373,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003520937670305318,
      "loss": 5.2317,
      "mean_token_accuracy": 0.20677698254585267,
      "num_tokens": 91353832.0,
      "step": 45085
    },
    {
      "entropy": 5.717036724090576,
      "epoch": 3.9404002446910775,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035206449146311345,
      "loss": 5.2054,
      "mean_token_accuracy": 0.20521843880414964,
      "num_tokens": 91362790.0,
      "step": 45090
    },
    {
      "entropy": 5.7927220344543455,
      "epoch": 3.940837193043782,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003520352144175577,
      "loss": 5.2928,
      "mean_token_accuracy": 0.20435260981321335,
      "num_tokens": 91372116.0,
      "step": 45095
    },
    {
      "entropy": 5.756730937957764,
      "epoch": 3.941274141396487,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003520059358944262,
      "loss": 5.249,
      "mean_token_accuracy": 0.2030319958925247,
      "num_tokens": 91382295.0,
      "step": 45100
    },
    {
      "entropy": 5.753274202346802,
      "epoch": 3.941711089749192,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035197665589428045,
      "loss": 5.2934,
      "mean_token_accuracy": 0.20444189608097077,
      "num_tokens": 91391568.0,
      "step": 45105
    },
    {
      "entropy": 5.74537878036499,
      "epoch": 3.9421480381018963,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035194737441768227,
      "loss": 5.2407,
      "mean_token_accuracy": 0.20993189066648482,
      "num_tokens": 91400254.0,
      "step": 45110
    },
    {
      "entropy": 5.814696550369263,
      "epoch": 3.942584986454601,
      "grad_norm": 1.0625,
      "learning_rate": 0.000351918091465193,
      "loss": 5.3239,
      "mean_token_accuracy": 0.19381391108036042,
      "num_tokens": 91409923.0,
      "step": 45115
    },
    {
      "entropy": 5.794309759140015,
      "epoch": 3.9430219348073057,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035188880703737444,
      "loss": 5.3357,
      "mean_token_accuracy": 0.19849587082862855,
      "num_tokens": 91420090.0,
      "step": 45120
    },
    {
      "entropy": 5.77896089553833,
      "epoch": 3.9434588831600106,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035185952113478826,
      "loss": 5.3846,
      "mean_token_accuracy": 0.19408017694950103,
      "num_tokens": 91429884.0,
      "step": 45125
    },
    {
      "entropy": 5.741040468215942,
      "epoch": 3.943895831512715,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003518302337579962,
      "loss": 5.2816,
      "mean_token_accuracy": 0.20150375813245774,
      "num_tokens": 91439517.0,
      "step": 45130
    },
    {
      "entropy": 5.849128532409668,
      "epoch": 3.94433277986542,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035180094490756006,
      "loss": 5.4272,
      "mean_token_accuracy": 0.191194149851799,
      "num_tokens": 91449673.0,
      "step": 45135
    },
    {
      "entropy": 5.829154396057129,
      "epoch": 3.944769728218125,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003517716545840415,
      "loss": 5.3646,
      "mean_token_accuracy": 0.1945726066827774,
      "num_tokens": 91460163.0,
      "step": 45140
    },
    {
      "entropy": 5.7398810386657715,
      "epoch": 3.9452066765708294,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003517423627880024,
      "loss": 5.3544,
      "mean_token_accuracy": 0.19995291233062745,
      "num_tokens": 91470735.0,
      "step": 45145
    },
    {
      "entropy": 5.70231065750122,
      "epoch": 3.945643624923534,
      "grad_norm": 1.0,
      "learning_rate": 0.0003517130695200045,
      "loss": 5.2624,
      "mean_token_accuracy": 0.20996433347463608,
      "num_tokens": 91480791.0,
      "step": 45150
    },
    {
      "entropy": 5.7675947666168215,
      "epoch": 3.9460805732762387,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035168377478060983,
      "loss": 5.1977,
      "mean_token_accuracy": 0.20839767754077912,
      "num_tokens": 91490061.0,
      "step": 45155
    },
    {
      "entropy": 5.744231796264648,
      "epoch": 3.9465175216289436,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003516544785703801,
      "loss": 5.2565,
      "mean_token_accuracy": 0.2069687470793724,
      "num_tokens": 91500185.0,
      "step": 45160
    },
    {
      "entropy": 5.769051551818848,
      "epoch": 3.946954469981648,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003516251808898774,
      "loss": 5.3053,
      "mean_token_accuracy": 0.19746237844228745,
      "num_tokens": 91510028.0,
      "step": 45165
    },
    {
      "entropy": 5.6946129322052,
      "epoch": 3.947391418334353,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035159588173966344,
      "loss": 5.2056,
      "mean_token_accuracy": 0.21761627197265626,
      "num_tokens": 91519333.0,
      "step": 45170
    },
    {
      "entropy": 5.7109198570251465,
      "epoch": 3.9478283666870575,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035156658112030036,
      "loss": 5.2139,
      "mean_token_accuracy": 0.20971193313598632,
      "num_tokens": 91528878.0,
      "step": 45175
    },
    {
      "entropy": 5.827957820892334,
      "epoch": 3.9482653150397624,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003515372790323502,
      "loss": 5.2914,
      "mean_token_accuracy": 0.194803287088871,
      "num_tokens": 91537817.0,
      "step": 45180
    },
    {
      "entropy": 5.757685136795044,
      "epoch": 3.948702263392467,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003515079754763749,
      "loss": 5.2446,
      "mean_token_accuracy": 0.20952966213226318,
      "num_tokens": 91548645.0,
      "step": 45185
    },
    {
      "entropy": 5.684130430221558,
      "epoch": 3.949139211745172,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035147867045293656,
      "loss": 5.2878,
      "mean_token_accuracy": 0.2055500790476799,
      "num_tokens": 91558695.0,
      "step": 45190
    },
    {
      "entropy": 5.77029504776001,
      "epoch": 3.9495761600978763,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035144936396259725,
      "loss": 5.3513,
      "mean_token_accuracy": 0.20086939185857772,
      "num_tokens": 91568711.0,
      "step": 45195
    },
    {
      "entropy": 5.756522464752197,
      "epoch": 3.950013108450581,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035142005600591905,
      "loss": 5.3295,
      "mean_token_accuracy": 0.20422399789094925,
      "num_tokens": 91578507.0,
      "step": 45200
    },
    {
      "entropy": 5.714606857299804,
      "epoch": 3.9504500568032856,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003513907465834642,
      "loss": 5.2288,
      "mean_token_accuracy": 0.21773778200149535,
      "num_tokens": 91588418.0,
      "step": 45205
    },
    {
      "entropy": 5.759281826019287,
      "epoch": 3.9508870051559906,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003513614356957947,
      "loss": 5.2827,
      "mean_token_accuracy": 0.20249475240707399,
      "num_tokens": 91597568.0,
      "step": 45210
    },
    {
      "entropy": 5.830329751968383,
      "epoch": 3.9513239535086955,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003513321233434729,
      "loss": 5.3812,
      "mean_token_accuracy": 0.19452255219221115,
      "num_tokens": 91607585.0,
      "step": 45215
    },
    {
      "entropy": 5.8476183891296385,
      "epoch": 3.9517609018614,
      "grad_norm": 0.953125,
      "learning_rate": 0.00035130280952706095,
      "loss": 5.3746,
      "mean_token_accuracy": 0.19434450715780258,
      "num_tokens": 91618593.0,
      "step": 45220
    },
    {
      "entropy": 5.7605767250061035,
      "epoch": 3.952197850214105,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003512734942471212,
      "loss": 5.3916,
      "mean_token_accuracy": 0.19231341332197188,
      "num_tokens": 91628788.0,
      "step": 45225
    },
    {
      "entropy": 5.780229091644287,
      "epoch": 3.9526347985668093,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003512441775042158,
      "loss": 5.3255,
      "mean_token_accuracy": 0.19253200590610503,
      "num_tokens": 91638340.0,
      "step": 45230
    },
    {
      "entropy": 5.873093032836914,
      "epoch": 3.9530717469195142,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003512148592989072,
      "loss": 5.4017,
      "mean_token_accuracy": 0.19266382455825806,
      "num_tokens": 91649510.0,
      "step": 45235
    },
    {
      "entropy": 5.768389940261841,
      "epoch": 3.9535086952722187,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003511855396317576,
      "loss": 5.2909,
      "mean_token_accuracy": 0.20870062559843064,
      "num_tokens": 91659751.0,
      "step": 45240
    },
    {
      "entropy": 5.735384368896485,
      "epoch": 3.9539456436249236,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003511562185033294,
      "loss": 5.2633,
      "mean_token_accuracy": 0.20337171703577042,
      "num_tokens": 91669395.0,
      "step": 45245
    },
    {
      "entropy": 5.712721061706543,
      "epoch": 3.954382591977628,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035112689591418507,
      "loss": 5.1853,
      "mean_token_accuracy": 0.20944264382123948,
      "num_tokens": 91679844.0,
      "step": 45250
    },
    {
      "entropy": 5.811263513565064,
      "epoch": 3.954819540330333,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035109757186488696,
      "loss": 5.4116,
      "mean_token_accuracy": 0.19489484131336213,
      "num_tokens": 91690868.0,
      "step": 45255
    },
    {
      "entropy": 5.762944984436035,
      "epoch": 3.9552564886830375,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035106824635599757,
      "loss": 5.2225,
      "mean_token_accuracy": 0.2061915874481201,
      "num_tokens": 91701785.0,
      "step": 45260
    },
    {
      "entropy": 5.8224176406860355,
      "epoch": 3.9556934370357424,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003510389193880793,
      "loss": 5.3089,
      "mean_token_accuracy": 0.1967039555311203,
      "num_tokens": 91711461.0,
      "step": 45265
    },
    {
      "entropy": 5.712564516067505,
      "epoch": 3.9561303853884473,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035100959096169473,
      "loss": 5.247,
      "mean_token_accuracy": 0.20746359080076218,
      "num_tokens": 91720590.0,
      "step": 45270
    },
    {
      "entropy": 5.741130685806274,
      "epoch": 3.9565673337411518,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003509802610774064,
      "loss": 5.3109,
      "mean_token_accuracy": 0.2087058261036873,
      "num_tokens": 91729798.0,
      "step": 45275
    },
    {
      "entropy": 5.8307044506073,
      "epoch": 3.9570042820938562,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003509509297357769,
      "loss": 5.3054,
      "mean_token_accuracy": 0.19221713542938232,
      "num_tokens": 91740498.0,
      "step": 45280
    },
    {
      "entropy": 5.837441396713257,
      "epoch": 3.957441230446561,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003509215969373687,
      "loss": 5.3152,
      "mean_token_accuracy": 0.20470752269029618,
      "num_tokens": 91750659.0,
      "step": 45285
    },
    {
      "entropy": 5.705283784866333,
      "epoch": 3.957878178799266,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035089226268274453,
      "loss": 5.2511,
      "mean_token_accuracy": 0.2033299759030342,
      "num_tokens": 91760426.0,
      "step": 45290
    },
    {
      "entropy": 5.752497339248658,
      "epoch": 3.9583151271519705,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035086292697246696,
      "loss": 5.266,
      "mean_token_accuracy": 0.20529307574033737,
      "num_tokens": 91769339.0,
      "step": 45295
    },
    {
      "entropy": 5.792366600036621,
      "epoch": 3.9587520755046754,
      "grad_norm": 1.25,
      "learning_rate": 0.00035083358980709877,
      "loss": 5.3361,
      "mean_token_accuracy": 0.19989053457975386,
      "num_tokens": 91779326.0,
      "step": 45300
    },
    {
      "entropy": 5.811782455444336,
      "epoch": 3.95918902385738,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003508042511872026,
      "loss": 5.2499,
      "mean_token_accuracy": 0.20153094977140426,
      "num_tokens": 91789626.0,
      "step": 45305
    },
    {
      "entropy": 5.703540229797364,
      "epoch": 3.959625972210085,
      "grad_norm": 1.078125,
      "learning_rate": 0.000350774911113341,
      "loss": 5.2846,
      "mean_token_accuracy": 0.20335458666086198,
      "num_tokens": 91799903.0,
      "step": 45310
    },
    {
      "entropy": 5.764790153503418,
      "epoch": 3.9600629205627893,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000350745569586077,
      "loss": 5.2604,
      "mean_token_accuracy": 0.2074039250612259,
      "num_tokens": 91810192.0,
      "step": 45315
    },
    {
      "entropy": 5.771958065032959,
      "epoch": 3.960499868915494,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003507162266059733,
      "loss": 5.2551,
      "mean_token_accuracy": 0.20123277157545089,
      "num_tokens": 91819213.0,
      "step": 45320
    },
    {
      "entropy": 5.819243431091309,
      "epoch": 3.960936817268199,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035068688217359276,
      "loss": 5.3869,
      "mean_token_accuracy": 0.19844255596399307,
      "num_tokens": 91829163.0,
      "step": 45325
    },
    {
      "entropy": 5.816220331192016,
      "epoch": 3.9613737656209036,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003506575362894981,
      "loss": 5.2995,
      "mean_token_accuracy": 0.20283593982458115,
      "num_tokens": 91838886.0,
      "step": 45330
    },
    {
      "entropy": 5.757699108123779,
      "epoch": 3.961810713973608,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035062818895425217,
      "loss": 5.246,
      "mean_token_accuracy": 0.20096812397241592,
      "num_tokens": 91848279.0,
      "step": 45335
    },
    {
      "entropy": 5.681399250030518,
      "epoch": 3.962247662326313,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003505988401684181,
      "loss": 5.2733,
      "mean_token_accuracy": 0.198733414709568,
      "num_tokens": 91858560.0,
      "step": 45340
    },
    {
      "entropy": 5.800277614593506,
      "epoch": 3.962684610679018,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003505694899325586,
      "loss": 5.2881,
      "mean_token_accuracy": 0.1959812819957733,
      "num_tokens": 91868381.0,
      "step": 45345
    },
    {
      "entropy": 5.812425231933593,
      "epoch": 3.9631215590317224,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003505401382472366,
      "loss": 5.3913,
      "mean_token_accuracy": 0.19418825209140778,
      "num_tokens": 91879512.0,
      "step": 45350
    },
    {
      "entropy": 5.732146501541138,
      "epoch": 3.9635585073844273,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035051078511301524,
      "loss": 5.254,
      "mean_token_accuracy": 0.20661930739879608,
      "num_tokens": 91888884.0,
      "step": 45355
    },
    {
      "entropy": 5.678847360610962,
      "epoch": 3.9639954557371317,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003504814305304575,
      "loss": 5.2325,
      "mean_token_accuracy": 0.209629026055336,
      "num_tokens": 91899029.0,
      "step": 45360
    },
    {
      "entropy": 5.798615884780884,
      "epoch": 3.9644324040898367,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00035045207450012635,
      "loss": 5.3736,
      "mean_token_accuracy": 0.19818245470523835,
      "num_tokens": 91909216.0,
      "step": 45365
    },
    {
      "entropy": 5.831394195556641,
      "epoch": 3.964869352442541,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00035042271702258483,
      "loss": 5.3297,
      "mean_token_accuracy": 0.1975288972258568,
      "num_tokens": 91920175.0,
      "step": 45370
    },
    {
      "entropy": 5.784224271774292,
      "epoch": 3.965306300795246,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003503933580983961,
      "loss": 5.2536,
      "mean_token_accuracy": 0.19810680747032167,
      "num_tokens": 91929916.0,
      "step": 45375
    },
    {
      "entropy": 5.739991188049316,
      "epoch": 3.965743249147951,
      "grad_norm": 1.0,
      "learning_rate": 0.00035036399772812323,
      "loss": 5.2327,
      "mean_token_accuracy": 0.2049515038728714,
      "num_tokens": 91939063.0,
      "step": 45380
    },
    {
      "entropy": 5.708801317214966,
      "epoch": 3.9661801975006554,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035033463591232946,
      "loss": 5.2942,
      "mean_token_accuracy": 0.20758705735206603,
      "num_tokens": 91949336.0,
      "step": 45385
    },
    {
      "entropy": 5.7289196968078615,
      "epoch": 3.96661714585336,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003503052726515778,
      "loss": 5.2409,
      "mean_token_accuracy": 0.2027894914150238,
      "num_tokens": 91958930.0,
      "step": 45390
    },
    {
      "entropy": 5.779714202880859,
      "epoch": 3.967054094206065,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003502759079464316,
      "loss": 5.3048,
      "mean_token_accuracy": 0.20057944357395172,
      "num_tokens": 91970761.0,
      "step": 45395
    },
    {
      "entropy": 5.817032766342163,
      "epoch": 3.9674910425587697,
      "grad_norm": 1.0,
      "learning_rate": 0.0003502465417974541,
      "loss": 5.355,
      "mean_token_accuracy": 0.19123406261205672,
      "num_tokens": 91981280.0,
      "step": 45400
    },
    {
      "entropy": 5.7769834995269775,
      "epoch": 3.967927990911474,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003502171742052084,
      "loss": 5.2199,
      "mean_token_accuracy": 0.20676645636558533,
      "num_tokens": 91990816.0,
      "step": 45405
    },
    {
      "entropy": 5.804506015777588,
      "epoch": 3.968364939264179,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035018780517025795,
      "loss": 5.3167,
      "mean_token_accuracy": 0.2021661266684532,
      "num_tokens": 92000183.0,
      "step": 45410
    },
    {
      "entropy": 5.753049612045288,
      "epoch": 3.9688018876168836,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035015843469316595,
      "loss": 5.344,
      "mean_token_accuracy": 0.1977086141705513,
      "num_tokens": 92010597.0,
      "step": 45415
    },
    {
      "entropy": 5.762103939056397,
      "epoch": 3.9692388359695885,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003501290627744959,
      "loss": 5.3554,
      "mean_token_accuracy": 0.20048696100711821,
      "num_tokens": 92021059.0,
      "step": 45420
    },
    {
      "entropy": 5.8149150848388675,
      "epoch": 3.969675784322293,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003500996894148109,
      "loss": 5.3055,
      "mean_token_accuracy": 0.2061001181602478,
      "num_tokens": 92030942.0,
      "step": 45425
    },
    {
      "entropy": 5.7536301612854,
      "epoch": 3.970112732674998,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035007031461467455,
      "loss": 5.2299,
      "mean_token_accuracy": 0.20968439280986786,
      "num_tokens": 92039821.0,
      "step": 45430
    },
    {
      "entropy": 5.748003149032593,
      "epoch": 3.970549681027703,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003500409383746503,
      "loss": 5.3077,
      "mean_token_accuracy": 0.20049989968538284,
      "num_tokens": 92049775.0,
      "step": 45435
    },
    {
      "entropy": 5.771305322647095,
      "epoch": 3.9709866293804073,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003500115606953015,
      "loss": 5.2493,
      "mean_token_accuracy": 0.2025871083140373,
      "num_tokens": 92059832.0,
      "step": 45440
    },
    {
      "entropy": 5.768004369735718,
      "epoch": 3.9714235777331117,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034998218157719163,
      "loss": 5.2684,
      "mean_token_accuracy": 0.19911076128482819,
      "num_tokens": 92069842.0,
      "step": 45445
    },
    {
      "entropy": 5.733288431167603,
      "epoch": 3.9718605260858166,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003499528010208842,
      "loss": 5.3307,
      "mean_token_accuracy": 0.20183704793453217,
      "num_tokens": 92079488.0,
      "step": 45450
    },
    {
      "entropy": 5.7438720703125,
      "epoch": 3.9722974744385215,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034992341902694284,
      "loss": 5.2796,
      "mean_token_accuracy": 0.20981639623641968,
      "num_tokens": 92088747.0,
      "step": 45455
    },
    {
      "entropy": 5.759666490554809,
      "epoch": 3.972734422791226,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034989403559593105,
      "loss": 5.265,
      "mean_token_accuracy": 0.20716967135667802,
      "num_tokens": 92099092.0,
      "step": 45460
    },
    {
      "entropy": 5.7226861953735355,
      "epoch": 3.973171371143931,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034986465072841237,
      "loss": 5.3376,
      "mean_token_accuracy": 0.20192731618881227,
      "num_tokens": 92108463.0,
      "step": 45465
    },
    {
      "entropy": 5.736295938491821,
      "epoch": 3.9736083194966354,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003498352644249506,
      "loss": 5.2444,
      "mean_token_accuracy": 0.21142036616802215,
      "num_tokens": 92117768.0,
      "step": 45470
    },
    {
      "entropy": 5.73481502532959,
      "epoch": 3.9740452678493403,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003498058766861091,
      "loss": 5.2941,
      "mean_token_accuracy": 0.20570998638868332,
      "num_tokens": 92127487.0,
      "step": 45475
    },
    {
      "entropy": 5.737604475021362,
      "epoch": 3.974482216202045,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003497764875124517,
      "loss": 5.1279,
      "mean_token_accuracy": 0.20596394687891006,
      "num_tokens": 92138154.0,
      "step": 45480
    },
    {
      "entropy": 5.823456144332885,
      "epoch": 3.9749191645547497,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003497470969045421,
      "loss": 5.2791,
      "mean_token_accuracy": 0.19830525368452073,
      "num_tokens": 92147591.0,
      "step": 45485
    },
    {
      "entropy": 5.760687160491943,
      "epoch": 3.9753561129074546,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034971770486294404,
      "loss": 5.2622,
      "mean_token_accuracy": 0.21026123464107513,
      "num_tokens": 92157621.0,
      "step": 45490
    },
    {
      "entropy": 5.792010736465454,
      "epoch": 3.975793061260159,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003496883113882212,
      "loss": 5.3913,
      "mean_token_accuracy": 0.1997600167989731,
      "num_tokens": 92168722.0,
      "step": 45495
    },
    {
      "entropy": 5.769952583312988,
      "epoch": 3.9762300096128635,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003496589164809374,
      "loss": 5.3386,
      "mean_token_accuracy": 0.20447733551263808,
      "num_tokens": 92179319.0,
      "step": 45500
    },
    {
      "entropy": 5.858365297317505,
      "epoch": 3.9766669579655685,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003496295201416565,
      "loss": 5.3394,
      "mean_token_accuracy": 0.19921338558197021,
      "num_tokens": 92191307.0,
      "step": 45505
    },
    {
      "entropy": 5.807950067520141,
      "epoch": 3.9771039063182734,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034960012237094236,
      "loss": 5.3154,
      "mean_token_accuracy": 0.20110504627227782,
      "num_tokens": 92201055.0,
      "step": 45510
    },
    {
      "entropy": 5.770984077453614,
      "epoch": 3.977540854670978,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00034957072316935865,
      "loss": 5.3237,
      "mean_token_accuracy": 0.19726345241069793,
      "num_tokens": 92211044.0,
      "step": 45515
    },
    {
      "entropy": 5.81788010597229,
      "epoch": 3.9779778030236828,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003495413225374695,
      "loss": 5.3079,
      "mean_token_accuracy": 0.20406400114297868,
      "num_tokens": 92221350.0,
      "step": 45520
    },
    {
      "entropy": 5.77415919303894,
      "epoch": 3.9784147513763872,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034951192047583867,
      "loss": 5.2389,
      "mean_token_accuracy": 0.20586840957403182,
      "num_tokens": 92230900.0,
      "step": 45525
    },
    {
      "entropy": 5.763245534896851,
      "epoch": 3.978851699729092,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034948251698503017,
      "loss": 5.3387,
      "mean_token_accuracy": 0.19693058282136916,
      "num_tokens": 92241715.0,
      "step": 45530
    },
    {
      "entropy": 5.822801828384399,
      "epoch": 3.9792886480817966,
      "grad_norm": 1.046875,
      "learning_rate": 0.000349453112065608,
      "loss": 5.3888,
      "mean_token_accuracy": 0.19513296335935593,
      "num_tokens": 92252007.0,
      "step": 45535
    },
    {
      "entropy": 5.8116601467132565,
      "epoch": 3.9797255964345015,
      "grad_norm": 2.5625,
      "learning_rate": 0.0003494237057181361,
      "loss": 5.3829,
      "mean_token_accuracy": 0.19682386964559556,
      "num_tokens": 92264377.0,
      "step": 45540
    },
    {
      "entropy": 5.824766349792481,
      "epoch": 3.980162544787206,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003493942979431786,
      "loss": 5.2954,
      "mean_token_accuracy": 0.20360403805971145,
      "num_tokens": 92275339.0,
      "step": 45545
    },
    {
      "entropy": 5.818782758712769,
      "epoch": 3.980599493139911,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034936488874129937,
      "loss": 5.2834,
      "mean_token_accuracy": 0.19454247802495955,
      "num_tokens": 92285923.0,
      "step": 45550
    },
    {
      "entropy": 5.771660375595093,
      "epoch": 3.9810364414926154,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003493354781130626,
      "loss": 5.3422,
      "mean_token_accuracy": 0.20195727795362473,
      "num_tokens": 92295148.0,
      "step": 45555
    },
    {
      "entropy": 5.80605149269104,
      "epoch": 3.9814733898453203,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00034930606605903256,
      "loss": 5.292,
      "mean_token_accuracy": 0.20131395012140274,
      "num_tokens": 92306074.0,
      "step": 45560
    },
    {
      "entropy": 5.719952392578125,
      "epoch": 3.981910338198025,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003492766525797731,
      "loss": 5.2394,
      "mean_token_accuracy": 0.20435271114110948,
      "num_tokens": 92315772.0,
      "step": 45565
    },
    {
      "entropy": 5.79303388595581,
      "epoch": 3.9823472865507297,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034924723767584847,
      "loss": 5.2866,
      "mean_token_accuracy": 0.20886735022068023,
      "num_tokens": 92326637.0,
      "step": 45570
    },
    {
      "entropy": 5.809984493255615,
      "epoch": 3.982784234903434,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034921782134782306,
      "loss": 5.322,
      "mean_token_accuracy": 0.20180429369211197,
      "num_tokens": 92337436.0,
      "step": 45575
    },
    {
      "entropy": 5.742823648452759,
      "epoch": 3.983221183256139,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00034918840359626085,
      "loss": 5.313,
      "mean_token_accuracy": 0.1969459369778633,
      "num_tokens": 92348548.0,
      "step": 45580
    },
    {
      "entropy": 5.756397914886475,
      "epoch": 3.983658131608844,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003491589844217263,
      "loss": 5.2786,
      "mean_token_accuracy": 0.20156058222055434,
      "num_tokens": 92359315.0,
      "step": 45585
    },
    {
      "entropy": 5.771241092681885,
      "epoch": 3.9840950799615484,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003491295638247835,
      "loss": 5.3478,
      "mean_token_accuracy": 0.20272220969200133,
      "num_tokens": 92369815.0,
      "step": 45590
    },
    {
      "entropy": 5.823926496505737,
      "epoch": 3.9845320283142533,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003491001418059968,
      "loss": 5.336,
      "mean_token_accuracy": 0.19632921516895294,
      "num_tokens": 92380658.0,
      "step": 45595
    },
    {
      "entropy": 5.839622735977173,
      "epoch": 3.984968976666958,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003490707183659306,
      "loss": 5.3338,
      "mean_token_accuracy": 0.20023045539855958,
      "num_tokens": 92389808.0,
      "step": 45600
    },
    {
      "entropy": 5.766720819473266,
      "epoch": 3.9854059250196627,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034904129350514913,
      "loss": 5.2904,
      "mean_token_accuracy": 0.19995625019073487,
      "num_tokens": 92399951.0,
      "step": 45605
    },
    {
      "entropy": 5.72490873336792,
      "epoch": 3.985842873372367,
      "grad_norm": 1.0,
      "learning_rate": 0.0003490118672242169,
      "loss": 5.3208,
      "mean_token_accuracy": 0.20165136456489563,
      "num_tokens": 92411157.0,
      "step": 45610
    },
    {
      "entropy": 5.79727725982666,
      "epoch": 3.986279821725072,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034898243952369827,
      "loss": 5.2278,
      "mean_token_accuracy": 0.20331313014030455,
      "num_tokens": 92421250.0,
      "step": 45615
    },
    {
      "entropy": 5.795769691467285,
      "epoch": 3.986716770077777,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003489530104041577,
      "loss": 5.4326,
      "mean_token_accuracy": 0.19533926248550415,
      "num_tokens": 92431664.0,
      "step": 45620
    },
    {
      "entropy": 5.795477771759034,
      "epoch": 3.9871537184304815,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003489235798661597,
      "loss": 5.3473,
      "mean_token_accuracy": 0.2029322698712349,
      "num_tokens": 92442362.0,
      "step": 45625
    },
    {
      "entropy": 5.8596807479858395,
      "epoch": 3.987590666783186,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034889414791026865,
      "loss": 5.2992,
      "mean_token_accuracy": 0.20313670784235,
      "num_tokens": 92452526.0,
      "step": 45630
    },
    {
      "entropy": 5.8216552734375,
      "epoch": 3.988027615135891,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034886471453704905,
      "loss": 5.3632,
      "mean_token_accuracy": 0.19353409856557846,
      "num_tokens": 92463418.0,
      "step": 45635
    },
    {
      "entropy": 5.784885263442993,
      "epoch": 3.988464563488596,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034883527974706566,
      "loss": 5.254,
      "mean_token_accuracy": 0.19995586723089218,
      "num_tokens": 92473814.0,
      "step": 45640
    },
    {
      "entropy": 5.758715677261352,
      "epoch": 3.9889015118413003,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034880584354088275,
      "loss": 5.3317,
      "mean_token_accuracy": 0.2072318285703659,
      "num_tokens": 92484234.0,
      "step": 45645
    },
    {
      "entropy": 5.830266332626342,
      "epoch": 3.989338460194005,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034877640591906526,
      "loss": 5.404,
      "mean_token_accuracy": 0.1926935762166977,
      "num_tokens": 92494746.0,
      "step": 45650
    },
    {
      "entropy": 5.820572185516357,
      "epoch": 3.9897754085467096,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034874696688217755,
      "loss": 5.3113,
      "mean_token_accuracy": 0.198506198823452,
      "num_tokens": 92505890.0,
      "step": 45655
    },
    {
      "entropy": 5.792422246932984,
      "epoch": 3.9902123568994146,
      "grad_norm": 0.96875,
      "learning_rate": 0.00034871752643078434,
      "loss": 5.3019,
      "mean_token_accuracy": 0.19979293644428253,
      "num_tokens": 92517593.0,
      "step": 45660
    },
    {
      "entropy": 5.785384559631348,
      "epoch": 3.990649305252119,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003486880845654504,
      "loss": 5.2685,
      "mean_token_accuracy": 0.19520262032747268,
      "num_tokens": 92528426.0,
      "step": 45665
    },
    {
      "entropy": 5.7779271602630615,
      "epoch": 3.991086253604824,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034865864128674035,
      "loss": 5.2912,
      "mean_token_accuracy": 0.19990044534206391,
      "num_tokens": 92538946.0,
      "step": 45670
    },
    {
      "entropy": 5.751228857040405,
      "epoch": 3.991523201957529,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003486291965952191,
      "loss": 5.335,
      "mean_token_accuracy": 0.2014627382159233,
      "num_tokens": 92548772.0,
      "step": 45675
    },
    {
      "entropy": 5.847142601013184,
      "epoch": 3.9919601503102333,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003485997504914511,
      "loss": 5.4136,
      "mean_token_accuracy": 0.19527264833450317,
      "num_tokens": 92560412.0,
      "step": 45680
    },
    {
      "entropy": 5.7394391059875485,
      "epoch": 3.992397098662938,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00034857030297600136,
      "loss": 5.2438,
      "mean_token_accuracy": 0.2101201370358467,
      "num_tokens": 92570936.0,
      "step": 45685
    },
    {
      "entropy": 5.778663969039917,
      "epoch": 3.9928340470156427,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003485408540494347,
      "loss": 5.2893,
      "mean_token_accuracy": 0.20058012306690215,
      "num_tokens": 92581431.0,
      "step": 45690
    },
    {
      "entropy": 5.830210971832275,
      "epoch": 3.9932709953683476,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003485114037123159,
      "loss": 5.3389,
      "mean_token_accuracy": 0.20532635897397994,
      "num_tokens": 92590787.0,
      "step": 45695
    },
    {
      "entropy": 5.747301149368286,
      "epoch": 3.993707943721052,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003484819519652099,
      "loss": 5.248,
      "mean_token_accuracy": 0.20260469764471054,
      "num_tokens": 92600473.0,
      "step": 45700
    },
    {
      "entropy": 5.818116950988769,
      "epoch": 3.994144892073757,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003484524988086815,
      "loss": 5.3703,
      "mean_token_accuracy": 0.19590949416160583,
      "num_tokens": 92611201.0,
      "step": 45705
    },
    {
      "entropy": 5.822778272628784,
      "epoch": 3.9945818404264615,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034842304424329576,
      "loss": 5.2725,
      "mean_token_accuracy": 0.19516904950141906,
      "num_tokens": 92620836.0,
      "step": 45710
    },
    {
      "entropy": 5.811301422119141,
      "epoch": 3.9950187887791664,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003483935882696175,
      "loss": 5.3385,
      "mean_token_accuracy": 0.19648140519857407,
      "num_tokens": 92630372.0,
      "step": 45715
    },
    {
      "entropy": 5.677077579498291,
      "epoch": 3.995455737131871,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003483641308882117,
      "loss": 5.2261,
      "mean_token_accuracy": 0.2051546037197113,
      "num_tokens": 92640281.0,
      "step": 45720
    },
    {
      "entropy": 5.800863742828369,
      "epoch": 3.9958926854845758,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003483346720996435,
      "loss": 5.2679,
      "mean_token_accuracy": 0.20368843376636506,
      "num_tokens": 92650505.0,
      "step": 45725
    },
    {
      "entropy": 5.779970455169678,
      "epoch": 3.9963296338372807,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003483052119044779,
      "loss": 5.311,
      "mean_token_accuracy": 0.20174129456281661,
      "num_tokens": 92660240.0,
      "step": 45730
    },
    {
      "entropy": 5.762682628631592,
      "epoch": 3.996766582189985,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003482757503032799,
      "loss": 5.2765,
      "mean_token_accuracy": 0.203689643740654,
      "num_tokens": 92670491.0,
      "step": 45735
    },
    {
      "entropy": 5.833797311782837,
      "epoch": 3.9972035305426896,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003482462872966146,
      "loss": 5.3018,
      "mean_token_accuracy": 0.20382059812545777,
      "num_tokens": 92680660.0,
      "step": 45740
    },
    {
      "entropy": 5.687645626068115,
      "epoch": 3.9976404788953945,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034821682288504704,
      "loss": 5.2072,
      "mean_token_accuracy": 0.2074928879737854,
      "num_tokens": 92690750.0,
      "step": 45745
    },
    {
      "entropy": 5.71488127708435,
      "epoch": 3.9980774272480994,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003481873570691426,
      "loss": 5.2251,
      "mean_token_accuracy": 0.2049037992954254,
      "num_tokens": 92702003.0,
      "step": 45750
    },
    {
      "entropy": 5.773447036743164,
      "epoch": 3.998514375600804,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034815788984946623,
      "loss": 5.2723,
      "mean_token_accuracy": 0.21110665053129196,
      "num_tokens": 92712480.0,
      "step": 45755
    },
    {
      "entropy": 5.721333456039429,
      "epoch": 3.998951323953509,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003481284212265832,
      "loss": 5.3395,
      "mean_token_accuracy": 0.19920795708894729,
      "num_tokens": 92723623.0,
      "step": 45760
    },
    {
      "entropy": 5.7871729850769045,
      "epoch": 3.9993882723062133,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003480989512010587,
      "loss": 5.2413,
      "mean_token_accuracy": 0.20620642006397247,
      "num_tokens": 92732778.0,
      "step": 45765
    },
    {
      "entropy": 5.833887195587158,
      "epoch": 3.999825220658918,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003480694797734581,
      "loss": 5.3389,
      "mean_token_accuracy": 0.19938234537839888,
      "num_tokens": 92742985.0,
      "step": 45770
    },
    {
      "entropy": 5.845783948898315,
      "epoch": 4.000262169011623,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034804000694434655,
      "loss": 5.2795,
      "mean_token_accuracy": 0.2021408721804619,
      "num_tokens": 92752370.0,
      "step": 45775
    },
    {
      "entropy": 5.792999315261841,
      "epoch": 4.000699117364327,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003480105327142894,
      "loss": 5.3143,
      "mean_token_accuracy": 0.2013792559504509,
      "num_tokens": 92763083.0,
      "step": 45780
    },
    {
      "entropy": 5.74084243774414,
      "epoch": 4.0011360657170325,
      "grad_norm": 1.0,
      "learning_rate": 0.00034798105708385194,
      "loss": 5.213,
      "mean_token_accuracy": 0.21051737815141677,
      "num_tokens": 92772492.0,
      "step": 45785
    },
    {
      "entropy": 5.776329612731933,
      "epoch": 4.001573014069737,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034795158005359954,
      "loss": 5.3071,
      "mean_token_accuracy": 0.1986828550696373,
      "num_tokens": 92783260.0,
      "step": 45790
    },
    {
      "entropy": 5.766127395629883,
      "epoch": 4.0020099624224414,
      "grad_norm": 1.5,
      "learning_rate": 0.00034792210162409767,
      "loss": 5.242,
      "mean_token_accuracy": 0.20638734251260757,
      "num_tokens": 92794269.0,
      "step": 45795
    },
    {
      "entropy": 5.796200180053711,
      "epoch": 4.002446910775147,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034789262179591167,
      "loss": 5.2134,
      "mean_token_accuracy": 0.20371314287185668,
      "num_tokens": 92804373.0,
      "step": 45800
    },
    {
      "entropy": 5.788658523559571,
      "epoch": 4.002883859127851,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000347863140569607,
      "loss": 5.2388,
      "mean_token_accuracy": 0.2018781915307045,
      "num_tokens": 92815626.0,
      "step": 45805
    },
    {
      "entropy": 5.823918581008911,
      "epoch": 4.003320807480556,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034783365794574907,
      "loss": 5.3183,
      "mean_token_accuracy": 0.19592168927192688,
      "num_tokens": 92825976.0,
      "step": 45810
    },
    {
      "entropy": 5.803660202026367,
      "epoch": 4.00375775583326,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003478041739249035,
      "loss": 5.2501,
      "mean_token_accuracy": 0.2065528929233551,
      "num_tokens": 92835677.0,
      "step": 45815
    },
    {
      "entropy": 5.767752933502197,
      "epoch": 4.004194704185966,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034777468850763564,
      "loss": 5.3062,
      "mean_token_accuracy": 0.2041911095380783,
      "num_tokens": 92846050.0,
      "step": 45820
    },
    {
      "entropy": 5.726237678527832,
      "epoch": 4.00463165253867,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034774520169451113,
      "loss": 5.1505,
      "mean_token_accuracy": 0.21279739141464232,
      "num_tokens": 92856016.0,
      "step": 45825
    },
    {
      "entropy": 5.745967626571655,
      "epoch": 4.0050686008913745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003477157134860955,
      "loss": 5.2959,
      "mean_token_accuracy": 0.20168893784284592,
      "num_tokens": 92866444.0,
      "step": 45830
    },
    {
      "entropy": 5.676704025268554,
      "epoch": 4.005505549244079,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034768622388295443,
      "loss": 5.1719,
      "mean_token_accuracy": 0.2153549537062645,
      "num_tokens": 92877329.0,
      "step": 45835
    },
    {
      "entropy": 5.804442739486694,
      "epoch": 4.005942497596784,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003476567328856536,
      "loss": 5.2599,
      "mean_token_accuracy": 0.2025332495570183,
      "num_tokens": 92886988.0,
      "step": 45840
    },
    {
      "entropy": 5.767997980117798,
      "epoch": 4.006379445949489,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003476272404947584,
      "loss": 5.219,
      "mean_token_accuracy": 0.20466239005327225,
      "num_tokens": 92896953.0,
      "step": 45845
    },
    {
      "entropy": 5.826470613479614,
      "epoch": 4.006816394302193,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034759774671083477,
      "loss": 5.3234,
      "mean_token_accuracy": 0.20161645859479904,
      "num_tokens": 92906938.0,
      "step": 45850
    },
    {
      "entropy": 5.75506911277771,
      "epoch": 4.007253342654899,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034756825153444825,
      "loss": 5.1654,
      "mean_token_accuracy": 0.2176588460803032,
      "num_tokens": 92916160.0,
      "step": 45855
    },
    {
      "entropy": 5.765103149414062,
      "epoch": 4.007690291007603,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034753875496616466,
      "loss": 5.2099,
      "mean_token_accuracy": 0.21083563864231109,
      "num_tokens": 92925610.0,
      "step": 45860
    },
    {
      "entropy": 5.779293584823608,
      "epoch": 4.008127239360308,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034750925700654985,
      "loss": 5.2615,
      "mean_token_accuracy": 0.20244917273521423,
      "num_tokens": 92935281.0,
      "step": 45865
    },
    {
      "entropy": 5.814653444290161,
      "epoch": 4.008564187713012,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034747975765616934,
      "loss": 5.3106,
      "mean_token_accuracy": 0.1971045032143593,
      "num_tokens": 92945258.0,
      "step": 45870
    },
    {
      "entropy": 5.766712284088134,
      "epoch": 4.009001136065717,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003474502569155892,
      "loss": 5.1907,
      "mean_token_accuracy": 0.20673797130584717,
      "num_tokens": 92955118.0,
      "step": 45875
    },
    {
      "entropy": 5.756727123260498,
      "epoch": 4.009438084418422,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034742075478537513,
      "loss": 5.2005,
      "mean_token_accuracy": 0.20647452622652054,
      "num_tokens": 92964579.0,
      "step": 45880
    },
    {
      "entropy": 5.794935321807861,
      "epoch": 4.009875032771126,
      "grad_norm": 1.046875,
      "learning_rate": 0.000347391251266093,
      "loss": 5.3005,
      "mean_token_accuracy": 0.19570521414279937,
      "num_tokens": 92974479.0,
      "step": 45885
    },
    {
      "entropy": 5.735324764251709,
      "epoch": 4.010311981123831,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003473617463583088,
      "loss": 5.2058,
      "mean_token_accuracy": 0.2030717447400093,
      "num_tokens": 92985392.0,
      "step": 45890
    },
    {
      "entropy": 5.831075620651245,
      "epoch": 4.010748929476536,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034733224006258835,
      "loss": 5.307,
      "mean_token_accuracy": 0.19758780300617218,
      "num_tokens": 92995742.0,
      "step": 45895
    },
    {
      "entropy": 5.801130533218384,
      "epoch": 4.011185877829241,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034730273237949767,
      "loss": 5.2377,
      "mean_token_accuracy": 0.2069494366645813,
      "num_tokens": 93006411.0,
      "step": 45900
    },
    {
      "entropy": 5.811720848083496,
      "epoch": 4.011622826181945,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034727322330960263,
      "loss": 5.3036,
      "mean_token_accuracy": 0.20671175718307494,
      "num_tokens": 93017369.0,
      "step": 45905
    },
    {
      "entropy": 5.728193616867065,
      "epoch": 4.01205977453465,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034724371285346935,
      "loss": 5.1871,
      "mean_token_accuracy": 0.2066642478108406,
      "num_tokens": 93027989.0,
      "step": 45910
    },
    {
      "entropy": 5.774593210220337,
      "epoch": 4.012496722887355,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034721420101166384,
      "loss": 5.244,
      "mean_token_accuracy": 0.20603490620851517,
      "num_tokens": 93037717.0,
      "step": 45915
    },
    {
      "entropy": 5.758599138259887,
      "epoch": 4.012933671240059,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003471846877847521,
      "loss": 5.2137,
      "mean_token_accuracy": 0.2040240079164505,
      "num_tokens": 93046879.0,
      "step": 45920
    },
    {
      "entropy": 5.784949398040771,
      "epoch": 4.013370619592764,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034715517317330017,
      "loss": 5.3163,
      "mean_token_accuracy": 0.19855143427848815,
      "num_tokens": 93058178.0,
      "step": 45925
    },
    {
      "entropy": 5.846062755584716,
      "epoch": 4.013807567945469,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003471256571778742,
      "loss": 5.2377,
      "mean_token_accuracy": 0.20254629254341125,
      "num_tokens": 93068001.0,
      "step": 45930
    },
    {
      "entropy": 5.717150640487671,
      "epoch": 4.014244516298174,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034709613979904037,
      "loss": 5.2138,
      "mean_token_accuracy": 0.20721083432435988,
      "num_tokens": 93077527.0,
      "step": 45935
    },
    {
      "entropy": 5.745383882522583,
      "epoch": 4.014681464650878,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034706662103736486,
      "loss": 5.2919,
      "mean_token_accuracy": 0.1972367838025093,
      "num_tokens": 93087242.0,
      "step": 45940
    },
    {
      "entropy": 5.843484830856323,
      "epoch": 4.015118413003583,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034703710089341365,
      "loss": 5.3188,
      "mean_token_accuracy": 0.19887037873268126,
      "num_tokens": 93097553.0,
      "step": 45945
    },
    {
      "entropy": 5.805345439910889,
      "epoch": 4.015555361356288,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003470075793677532,
      "loss": 5.1911,
      "mean_token_accuracy": 0.21067534387111664,
      "num_tokens": 93107433.0,
      "step": 45950
    },
    {
      "entropy": 5.698177337646484,
      "epoch": 4.0159923097089925,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003469780564609496,
      "loss": 5.1537,
      "mean_token_accuracy": 0.2105378031730652,
      "num_tokens": 93117643.0,
      "step": 45955
    },
    {
      "entropy": 5.707441711425782,
      "epoch": 4.016429258061697,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034694853217356924,
      "loss": 5.261,
      "mean_token_accuracy": 0.20888070166110992,
      "num_tokens": 93128134.0,
      "step": 45960
    },
    {
      "entropy": 5.790256977081299,
      "epoch": 4.016866206414401,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034691900650617825,
      "loss": 5.1981,
      "mean_token_accuracy": 0.2106435090303421,
      "num_tokens": 93138194.0,
      "step": 45965
    },
    {
      "entropy": 5.754315376281738,
      "epoch": 4.017303154767107,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000346889479459343,
      "loss": 5.2463,
      "mean_token_accuracy": 0.20525214821100235,
      "num_tokens": 93148015.0,
      "step": 45970
    },
    {
      "entropy": 5.702709627151489,
      "epoch": 4.017740103119811,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003468599510336299,
      "loss": 5.1936,
      "mean_token_accuracy": 0.20784238129854202,
      "num_tokens": 93157512.0,
      "step": 45975
    },
    {
      "entropy": 5.734858512878418,
      "epoch": 4.018177051472516,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034683042122960535,
      "loss": 5.2607,
      "mean_token_accuracy": 0.20710242986679078,
      "num_tokens": 93168863.0,
      "step": 45980
    },
    {
      "entropy": 5.734493637084961,
      "epoch": 4.018613999825221,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034680089004783564,
      "loss": 5.308,
      "mean_token_accuracy": 0.20285548716783525,
      "num_tokens": 93179375.0,
      "step": 45985
    },
    {
      "entropy": 5.830523538589477,
      "epoch": 4.0190509481779255,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003467713574888873,
      "loss": 5.2436,
      "mean_token_accuracy": 0.20423467010259627,
      "num_tokens": 93189517.0,
      "step": 45990
    },
    {
      "entropy": 5.795507526397705,
      "epoch": 4.01948789653063,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034674182355332663,
      "loss": 5.2525,
      "mean_token_accuracy": 0.2064846619963646,
      "num_tokens": 93199712.0,
      "step": 45995
    },
    {
      "entropy": 5.80151743888855,
      "epoch": 4.0199248448833345,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003467122882417202,
      "loss": 5.3104,
      "mean_token_accuracy": 0.20504960268735886,
      "num_tokens": 93209976.0,
      "step": 46000
    },
    {
      "entropy": 5.789269876480103,
      "epoch": 4.02036179323604,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003466827515546345,
      "loss": 5.3154,
      "mean_token_accuracy": 0.19999420791864395,
      "num_tokens": 93220003.0,
      "step": 46005
    },
    {
      "entropy": 5.7606213092803955,
      "epoch": 4.020798741588744,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034665321349263605,
      "loss": 5.2512,
      "mean_token_accuracy": 0.20180205255746841,
      "num_tokens": 93230220.0,
      "step": 46010
    },
    {
      "entropy": 5.769651985168457,
      "epoch": 4.021235689941449,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003466236740562915,
      "loss": 5.2525,
      "mean_token_accuracy": 0.20443236231803893,
      "num_tokens": 93241041.0,
      "step": 46015
    },
    {
      "entropy": 5.742683029174804,
      "epoch": 4.021672638294153,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003465941332461673,
      "loss": 5.1539,
      "mean_token_accuracy": 0.2125861793756485,
      "num_tokens": 93251200.0,
      "step": 46020
    },
    {
      "entropy": 5.7604248046875,
      "epoch": 4.022109586646859,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034656459106283013,
      "loss": 5.2819,
      "mean_token_accuracy": 0.20054540038108826,
      "num_tokens": 93260564.0,
      "step": 46025
    },
    {
      "entropy": 5.711347484588623,
      "epoch": 4.022546534999563,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003465350475068466,
      "loss": 5.1767,
      "mean_token_accuracy": 0.21524539291858674,
      "num_tokens": 93270101.0,
      "step": 46030
    },
    {
      "entropy": 5.775501298904419,
      "epoch": 4.0229834833522675,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003465055025787833,
      "loss": 5.2601,
      "mean_token_accuracy": 0.20459069311618805,
      "num_tokens": 93280551.0,
      "step": 46035
    },
    {
      "entropy": 5.770950508117676,
      "epoch": 4.023420431704973,
      "grad_norm": 1.125,
      "learning_rate": 0.0003464759562792071,
      "loss": 5.197,
      "mean_token_accuracy": 0.21322955787181855,
      "num_tokens": 93289658.0,
      "step": 46040
    },
    {
      "entropy": 5.772742366790771,
      "epoch": 4.023857380057677,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003464464086086845,
      "loss": 5.2786,
      "mean_token_accuracy": 0.20083337873220444,
      "num_tokens": 93300402.0,
      "step": 46045
    },
    {
      "entropy": 5.717894601821899,
      "epoch": 4.024294328410382,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034641685956778246,
      "loss": 5.1815,
      "mean_token_accuracy": 0.2101916939020157,
      "num_tokens": 93310254.0,
      "step": 46050
    },
    {
      "entropy": 5.6843743324279785,
      "epoch": 4.024731276763086,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003463873091570676,
      "loss": 5.2151,
      "mean_token_accuracy": 0.1997531995177269,
      "num_tokens": 93319745.0,
      "step": 46055
    },
    {
      "entropy": 5.705549097061157,
      "epoch": 4.025168225115792,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034635775737710666,
      "loss": 5.2202,
      "mean_token_accuracy": 0.2065220519900322,
      "num_tokens": 93329218.0,
      "step": 46060
    },
    {
      "entropy": 5.776245594024658,
      "epoch": 4.025605173468496,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003463282042284667,
      "loss": 5.2833,
      "mean_token_accuracy": 0.20457299649715424,
      "num_tokens": 93339140.0,
      "step": 46065
    },
    {
      "entropy": 5.778555965423584,
      "epoch": 4.026042121821201,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034629864971171426,
      "loss": 5.2573,
      "mean_token_accuracy": 0.20259866416454314,
      "num_tokens": 93348325.0,
      "step": 46070
    },
    {
      "entropy": 5.798482894897461,
      "epoch": 4.026479070173905,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003462690938274164,
      "loss": 5.2978,
      "mean_token_accuracy": 0.20202603936195374,
      "num_tokens": 93359050.0,
      "step": 46075
    },
    {
      "entropy": 5.788121032714844,
      "epoch": 4.02691601852661,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034623953657614,
      "loss": 5.2238,
      "mean_token_accuracy": 0.20547318905591966,
      "num_tokens": 93369518.0,
      "step": 46080
    },
    {
      "entropy": 5.700077724456787,
      "epoch": 4.027352966879315,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034620997795845195,
      "loss": 5.1595,
      "mean_token_accuracy": 0.21312186121940613,
      "num_tokens": 93380922.0,
      "step": 46085
    },
    {
      "entropy": 5.745960903167725,
      "epoch": 4.027789915232019,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003461804179749192,
      "loss": 5.2181,
      "mean_token_accuracy": 0.20774227678775786,
      "num_tokens": 93391196.0,
      "step": 46090
    },
    {
      "entropy": 5.788701438903809,
      "epoch": 4.028226863584725,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034615085662610874,
      "loss": 5.3094,
      "mean_token_accuracy": 0.19799134135246277,
      "num_tokens": 93401923.0,
      "step": 46095
    },
    {
      "entropy": 5.817461681365967,
      "epoch": 4.028663811937429,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003461212939125876,
      "loss": 5.2256,
      "mean_token_accuracy": 0.20459103137254714,
      "num_tokens": 93411940.0,
      "step": 46100
    },
    {
      "entropy": 5.82671709060669,
      "epoch": 4.029100760290134,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003460917298349227,
      "loss": 5.3227,
      "mean_token_accuracy": 0.19646211266517638,
      "num_tokens": 93422509.0,
      "step": 46105
    },
    {
      "entropy": 5.776410627365112,
      "epoch": 4.029537708642838,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003460621643936812,
      "loss": 5.2651,
      "mean_token_accuracy": 0.20464201420545577,
      "num_tokens": 93432426.0,
      "step": 46110
    },
    {
      "entropy": 5.792518281936646,
      "epoch": 4.0299746569955435,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034603259758943017,
      "loss": 5.2337,
      "mean_token_accuracy": 0.20729392915964126,
      "num_tokens": 93442572.0,
      "step": 46115
    },
    {
      "entropy": 5.77733154296875,
      "epoch": 4.030411605348248,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003460030294227367,
      "loss": 5.2079,
      "mean_token_accuracy": 0.2084834486246109,
      "num_tokens": 93452848.0,
      "step": 46120
    },
    {
      "entropy": 5.726694583892822,
      "epoch": 4.030848553700952,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000345973459894168,
      "loss": 5.2348,
      "mean_token_accuracy": 0.20498985797166824,
      "num_tokens": 93462793.0,
      "step": 46125
    },
    {
      "entropy": 5.7326561450958256,
      "epoch": 4.031285502053657,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000345943889004291,
      "loss": 5.2743,
      "mean_token_accuracy": 0.20551130175590515,
      "num_tokens": 93472634.0,
      "step": 46130
    },
    {
      "entropy": 5.743097925186158,
      "epoch": 4.031722450406362,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034591431675367305,
      "loss": 5.1945,
      "mean_token_accuracy": 0.20683964639902114,
      "num_tokens": 93482299.0,
      "step": 46135
    },
    {
      "entropy": 5.743162441253662,
      "epoch": 4.032159398759067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034588474314288135,
      "loss": 5.2339,
      "mean_token_accuracy": 0.21155254393815995,
      "num_tokens": 93492675.0,
      "step": 46140
    },
    {
      "entropy": 5.748600816726684,
      "epoch": 4.032596347111771,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034585516817248326,
      "loss": 5.1756,
      "mean_token_accuracy": 0.20503010153770446,
      "num_tokens": 93502256.0,
      "step": 46145
    },
    {
      "entropy": 5.769777154922485,
      "epoch": 4.0330332954644765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034582559184304583,
      "loss": 5.3106,
      "mean_token_accuracy": 0.19967083036899566,
      "num_tokens": 93512282.0,
      "step": 46150
    },
    {
      "entropy": 5.807619237899781,
      "epoch": 4.033470243817181,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003457960141551365,
      "loss": 5.3195,
      "mean_token_accuracy": 0.19740084558725357,
      "num_tokens": 93521916.0,
      "step": 46155
    },
    {
      "entropy": 5.7772767543792725,
      "epoch": 4.0339071921698855,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003457664351093224,
      "loss": 5.2347,
      "mean_token_accuracy": 0.2005910947918892,
      "num_tokens": 93531126.0,
      "step": 46160
    },
    {
      "entropy": 5.709876537322998,
      "epoch": 4.03434414052259,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000345736854706171,
      "loss": 5.1496,
      "mean_token_accuracy": 0.21447180658578874,
      "num_tokens": 93540368.0,
      "step": 46165
    },
    {
      "entropy": 5.744932079315186,
      "epoch": 4.034781088875295,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003457072729462498,
      "loss": 5.1802,
      "mean_token_accuracy": 0.20875655263662338,
      "num_tokens": 93550213.0,
      "step": 46170
    },
    {
      "entropy": 5.716238403320313,
      "epoch": 4.035218037228,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034567768983012593,
      "loss": 5.2266,
      "mean_token_accuracy": 0.20631600320339202,
      "num_tokens": 93560673.0,
      "step": 46175
    },
    {
      "entropy": 5.7428553104400635,
      "epoch": 4.035654985580704,
      "grad_norm": 1.078125,
      "learning_rate": 0.000345648105358367,
      "loss": 5.1728,
      "mean_token_accuracy": 0.21304628103971482,
      "num_tokens": 93571546.0,
      "step": 46180
    },
    {
      "entropy": 5.789531230926514,
      "epoch": 4.036091933933409,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034561851953154036,
      "loss": 5.2438,
      "mean_token_accuracy": 0.20045332759618759,
      "num_tokens": 93581440.0,
      "step": 46185
    },
    {
      "entropy": 5.719044971466064,
      "epoch": 4.036528882286114,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003455889323502135,
      "loss": 5.1593,
      "mean_token_accuracy": 0.212611386179924,
      "num_tokens": 93591686.0,
      "step": 46190
    },
    {
      "entropy": 5.718543291091919,
      "epoch": 4.0369658306388185,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034555934381495394,
      "loss": 5.2771,
      "mean_token_accuracy": 0.21129924803972244,
      "num_tokens": 93602043.0,
      "step": 46195
    },
    {
      "entropy": 5.819355535507202,
      "epoch": 4.037402778991523,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034552975392632926,
      "loss": 5.2246,
      "mean_token_accuracy": 0.20755399018526077,
      "num_tokens": 93611639.0,
      "step": 46200
    },
    {
      "entropy": 5.806118822097778,
      "epoch": 4.0378397273442275,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034550016268490683,
      "loss": 5.295,
      "mean_token_accuracy": 0.1978253111243248,
      "num_tokens": 93622236.0,
      "step": 46205
    },
    {
      "entropy": 5.697878456115722,
      "epoch": 4.038276675696933,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003454705700912544,
      "loss": 5.1349,
      "mean_token_accuracy": 0.2119095578789711,
      "num_tokens": 93632150.0,
      "step": 46210
    },
    {
      "entropy": 5.7765923023223875,
      "epoch": 4.038713624049637,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003454409761459395,
      "loss": 5.2986,
      "mean_token_accuracy": 0.19951745569705964,
      "num_tokens": 93644843.0,
      "step": 46215
    },
    {
      "entropy": 5.782945775985718,
      "epoch": 4.039150572402342,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003454113808495298,
      "loss": 5.2626,
      "mean_token_accuracy": 0.205076701939106,
      "num_tokens": 93654633.0,
      "step": 46220
    },
    {
      "entropy": 5.802406978607178,
      "epoch": 4.039587520755047,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034538178420259286,
      "loss": 5.2309,
      "mean_token_accuracy": 0.20419895499944687,
      "num_tokens": 93664158.0,
      "step": 46225
    },
    {
      "entropy": 5.718568801879883,
      "epoch": 4.040024469107752,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003453521862056965,
      "loss": 5.2153,
      "mean_token_accuracy": 0.2009284406900406,
      "num_tokens": 93673845.0,
      "step": 46230
    },
    {
      "entropy": 5.792005395889282,
      "epoch": 4.040461417460456,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034532258685940825,
      "loss": 5.2022,
      "mean_token_accuracy": 0.20558818131685258,
      "num_tokens": 93683871.0,
      "step": 46235
    },
    {
      "entropy": 5.777534246444702,
      "epoch": 4.0408983658131605,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000345292986164296,
      "loss": 5.2542,
      "mean_token_accuracy": 0.20393215119838715,
      "num_tokens": 93693652.0,
      "step": 46240
    },
    {
      "entropy": 5.794251775741577,
      "epoch": 4.041335314165866,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034526338412092745,
      "loss": 5.3,
      "mean_token_accuracy": 0.20534225851297377,
      "num_tokens": 93702903.0,
      "step": 46245
    },
    {
      "entropy": 5.706576871871948,
      "epoch": 4.04177226251857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003452337807298703,
      "loss": 5.1819,
      "mean_token_accuracy": 0.21060780733823775,
      "num_tokens": 93712328.0,
      "step": 46250
    },
    {
      "entropy": 5.697664070129394,
      "epoch": 4.042209210871275,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003452041759916925,
      "loss": 5.2154,
      "mean_token_accuracy": 0.21129674464464188,
      "num_tokens": 93722399.0,
      "step": 46255
    },
    {
      "entropy": 5.80123815536499,
      "epoch": 4.042646159223979,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003451745699069618,
      "loss": 5.2866,
      "mean_token_accuracy": 0.2047115683555603,
      "num_tokens": 93732034.0,
      "step": 46260
    },
    {
      "entropy": 5.727363872528076,
      "epoch": 4.043083107576685,
      "grad_norm": 1.125,
      "learning_rate": 0.0003451449624762461,
      "loss": 5.1437,
      "mean_token_accuracy": 0.21193058788776398,
      "num_tokens": 93741086.0,
      "step": 46265
    },
    {
      "entropy": 5.70851149559021,
      "epoch": 4.043520055929389,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003451153537001132,
      "loss": 5.1747,
      "mean_token_accuracy": 0.21051815748214722,
      "num_tokens": 93750297.0,
      "step": 46270
    },
    {
      "entropy": 5.73928599357605,
      "epoch": 4.043957004282094,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003450857435791311,
      "loss": 5.1835,
      "mean_token_accuracy": 0.20580843538045884,
      "num_tokens": 93760288.0,
      "step": 46275
    },
    {
      "entropy": 5.752780246734619,
      "epoch": 4.044393952634799,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034505613211386775,
      "loss": 5.2158,
      "mean_token_accuracy": 0.2052503392100334,
      "num_tokens": 93770025.0,
      "step": 46280
    },
    {
      "entropy": 5.706767892837524,
      "epoch": 4.044830900987503,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034502651930489104,
      "loss": 5.1539,
      "mean_token_accuracy": 0.21012303084135056,
      "num_tokens": 93779659.0,
      "step": 46285
    },
    {
      "entropy": 5.767877864837646,
      "epoch": 4.045267849340208,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034499690515276906,
      "loss": 5.2176,
      "mean_token_accuracy": 0.20599068254232406,
      "num_tokens": 93790820.0,
      "step": 46290
    },
    {
      "entropy": 5.652456617355346,
      "epoch": 4.045704797692912,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003449672896580697,
      "loss": 5.238,
      "mean_token_accuracy": 0.20899258852005004,
      "num_tokens": 93801703.0,
      "step": 46295
    },
    {
      "entropy": 5.70888123512268,
      "epoch": 4.046141746045618,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003449376728213611,
      "loss": 5.2187,
      "mean_token_accuracy": 0.20953814089298248,
      "num_tokens": 93812743.0,
      "step": 46300
    },
    {
      "entropy": 5.7448750019073485,
      "epoch": 4.046578694398322,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034490805464321123,
      "loss": 5.25,
      "mean_token_accuracy": 0.2072514995932579,
      "num_tokens": 93822825.0,
      "step": 46305
    },
    {
      "entropy": 5.771680021286011,
      "epoch": 4.047015642751027,
      "grad_norm": 1.0,
      "learning_rate": 0.00034487843512418827,
      "loss": 5.2311,
      "mean_token_accuracy": 0.19976333081722258,
      "num_tokens": 93832650.0,
      "step": 46310
    },
    {
      "entropy": 5.736601114273071,
      "epoch": 4.047452591103731,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034484881426486043,
      "loss": 5.2226,
      "mean_token_accuracy": 0.2120048940181732,
      "num_tokens": 93842689.0,
      "step": 46315
    },
    {
      "entropy": 5.771039152145386,
      "epoch": 4.0478895394564365,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034481919206579546,
      "loss": 5.1866,
      "mean_token_accuracy": 0.20917435437440873,
      "num_tokens": 93852464.0,
      "step": 46320
    },
    {
      "entropy": 5.7303053855896,
      "epoch": 4.048326487809141,
      "grad_norm": 1.03125,
      "learning_rate": 0.000344789568527562,
      "loss": 5.1624,
      "mean_token_accuracy": 0.21379658728837966,
      "num_tokens": 93862241.0,
      "step": 46325
    },
    {
      "entropy": 5.762551593780517,
      "epoch": 4.048763436161845,
      "grad_norm": 1.0,
      "learning_rate": 0.000344759943650728,
      "loss": 5.2888,
      "mean_token_accuracy": 0.19937712401151658,
      "num_tokens": 93872620.0,
      "step": 46330
    },
    {
      "entropy": 5.712474679946899,
      "epoch": 4.049200384514551,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003447303174358618,
      "loss": 5.1495,
      "mean_token_accuracy": 0.2155172422528267,
      "num_tokens": 93882536.0,
      "step": 46335
    },
    {
      "entropy": 5.800695133209229,
      "epoch": 4.049637332867255,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034470068988353146,
      "loss": 5.2081,
      "mean_token_accuracy": 0.20861902087926865,
      "num_tokens": 93892072.0,
      "step": 46340
    },
    {
      "entropy": 5.7831930160522464,
      "epoch": 4.05007428121996,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034467106099430537,
      "loss": 5.3345,
      "mean_token_accuracy": 0.19809054285287858,
      "num_tokens": 93902230.0,
      "step": 46345
    },
    {
      "entropy": 5.763327932357788,
      "epoch": 4.050511229572664,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034464143076875186,
      "loss": 5.2512,
      "mean_token_accuracy": 0.2080755576491356,
      "num_tokens": 93911617.0,
      "step": 46350
    },
    {
      "entropy": 5.711084032058716,
      "epoch": 4.0509481779253695,
      "grad_norm": 1.125,
      "learning_rate": 0.00034461179920743925,
      "loss": 5.0541,
      "mean_token_accuracy": 0.2191966861486435,
      "num_tokens": 93922131.0,
      "step": 46355
    },
    {
      "entropy": 5.72813081741333,
      "epoch": 4.051385126278074,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034458216631093585,
      "loss": 5.2418,
      "mean_token_accuracy": 0.20030282735824584,
      "num_tokens": 93932186.0,
      "step": 46360
    },
    {
      "entropy": 5.802563667297363,
      "epoch": 4.0518220746307785,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034455253207981004,
      "loss": 5.3078,
      "mean_token_accuracy": 0.20246020555496216,
      "num_tokens": 93942925.0,
      "step": 46365
    },
    {
      "entropy": 5.813523578643799,
      "epoch": 4.052259022983483,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034452289651463016,
      "loss": 5.2814,
      "mean_token_accuracy": 0.20247018188238144,
      "num_tokens": 93954179.0,
      "step": 46370
    },
    {
      "entropy": 5.808640813827514,
      "epoch": 4.052695971336188,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034449325961596476,
      "loss": 5.3026,
      "mean_token_accuracy": 0.2007174998521805,
      "num_tokens": 93964374.0,
      "step": 46375
    },
    {
      "entropy": 5.7611793041229244,
      "epoch": 4.053132919688893,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003444636213843822,
      "loss": 5.1959,
      "mean_token_accuracy": 0.2028034508228302,
      "num_tokens": 93974321.0,
      "step": 46380
    },
    {
      "entropy": 5.7527368545532225,
      "epoch": 4.053569868041597,
      "grad_norm": 1.0625,
      "learning_rate": 0.000344433981820451,
      "loss": 5.3234,
      "mean_token_accuracy": 0.20346688330173493,
      "num_tokens": 93984614.0,
      "step": 46385
    },
    {
      "entropy": 5.851608562469482,
      "epoch": 4.054006816394303,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003444043409247396,
      "loss": 5.3163,
      "mean_token_accuracy": 0.20095207244157792,
      "num_tokens": 93994797.0,
      "step": 46390
    },
    {
      "entropy": 5.771607685089111,
      "epoch": 4.054443764747007,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003443746986978166,
      "loss": 5.2584,
      "mean_token_accuracy": 0.20063008964061738,
      "num_tokens": 94005383.0,
      "step": 46395
    },
    {
      "entropy": 5.685872411727905,
      "epoch": 4.0548807130997115,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034434505514025057,
      "loss": 5.1407,
      "mean_token_accuracy": 0.22118010371923447,
      "num_tokens": 94013951.0,
      "step": 46400
    },
    {
      "entropy": 5.77421932220459,
      "epoch": 4.055317661452416,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034431541025260997,
      "loss": 5.2851,
      "mean_token_accuracy": 0.20365623980760575,
      "num_tokens": 94026044.0,
      "step": 46405
    },
    {
      "entropy": 5.767602968215942,
      "epoch": 4.055754609805121,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034428576403546353,
      "loss": 5.2442,
      "mean_token_accuracy": 0.20964139997959136,
      "num_tokens": 94035543.0,
      "step": 46410
    },
    {
      "entropy": 5.744489574432373,
      "epoch": 4.056191558157826,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003442561164893798,
      "loss": 5.2577,
      "mean_token_accuracy": 0.2006083369255066,
      "num_tokens": 94044480.0,
      "step": 46415
    },
    {
      "entropy": 5.728696298599243,
      "epoch": 4.05662850651053,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00034422646761492746,
      "loss": 5.2171,
      "mean_token_accuracy": 0.2107670336961746,
      "num_tokens": 94055233.0,
      "step": 46420
    },
    {
      "entropy": 5.783304262161255,
      "epoch": 4.057065454863235,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003441968174126752,
      "loss": 5.2652,
      "mean_token_accuracy": 0.20188159197568895,
      "num_tokens": 94067182.0,
      "step": 46425
    },
    {
      "entropy": 5.692991113662719,
      "epoch": 4.05750240321594,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003441671658831918,
      "loss": 5.2139,
      "mean_token_accuracy": 0.2119888812303543,
      "num_tokens": 94076799.0,
      "step": 46430
    },
    {
      "entropy": 5.731865310668946,
      "epoch": 4.057939351568645,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003441375130270458,
      "loss": 5.1715,
      "mean_token_accuracy": 0.21017423272132874,
      "num_tokens": 94086829.0,
      "step": 46435
    },
    {
      "entropy": 5.710104465484619,
      "epoch": 4.058376299921349,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034410785884480616,
      "loss": 5.2009,
      "mean_token_accuracy": 0.21319700926542282,
      "num_tokens": 94097124.0,
      "step": 46440
    },
    {
      "entropy": 5.753516292572021,
      "epoch": 4.058813248274054,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034407820333704153,
      "loss": 5.222,
      "mean_token_accuracy": 0.2073309227824211,
      "num_tokens": 94107714.0,
      "step": 46445
    },
    {
      "entropy": 5.6930561542510985,
      "epoch": 4.059250196626759,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003440485465043208,
      "loss": 5.2684,
      "mean_token_accuracy": 0.20466853976249694,
      "num_tokens": 94118139.0,
      "step": 46450
    },
    {
      "entropy": 5.760197114944458,
      "epoch": 4.059687144979463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034401888834721277,
      "loss": 5.2439,
      "mean_token_accuracy": 0.20441304743289948,
      "num_tokens": 94128452.0,
      "step": 46455
    },
    {
      "entropy": 5.780305337905884,
      "epoch": 4.060124093332168,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003439892288662863,
      "loss": 5.2491,
      "mean_token_accuracy": 0.20284321010112763,
      "num_tokens": 94138195.0,
      "step": 46460
    },
    {
      "entropy": 5.7555183410644535,
      "epoch": 4.060561041684873,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00034395956806211023,
      "loss": 5.2642,
      "mean_token_accuracy": 0.1986502915620804,
      "num_tokens": 94149261.0,
      "step": 46465
    },
    {
      "entropy": 5.80994725227356,
      "epoch": 4.060997990037578,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003439299059352535,
      "loss": 5.2393,
      "mean_token_accuracy": 0.20298687517642974,
      "num_tokens": 94159067.0,
      "step": 46470
    },
    {
      "entropy": 5.777424287796021,
      "epoch": 4.061434938390282,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003439002424862851,
      "loss": 5.2588,
      "mean_token_accuracy": 0.20163802057504654,
      "num_tokens": 94169463.0,
      "step": 46475
    },
    {
      "entropy": 5.762679576873779,
      "epoch": 4.061871886742987,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003438705777157739,
      "loss": 5.2283,
      "mean_token_accuracy": 0.20722174346446992,
      "num_tokens": 94179325.0,
      "step": 46480
    },
    {
      "entropy": 5.769746971130371,
      "epoch": 4.062308835095692,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000343840911624289,
      "loss": 5.2599,
      "mean_token_accuracy": 0.20212825238704682,
      "num_tokens": 94189428.0,
      "step": 46485
    },
    {
      "entropy": 5.694004249572754,
      "epoch": 4.062745783448396,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003438112442123993,
      "loss": 5.1987,
      "mean_token_accuracy": 0.21544961482286454,
      "num_tokens": 94198995.0,
      "step": 46490
    },
    {
      "entropy": 5.760121440887451,
      "epoch": 4.063182731801101,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003437815754806738,
      "loss": 5.2322,
      "mean_token_accuracy": 0.20371549874544143,
      "num_tokens": 94209456.0,
      "step": 46495
    },
    {
      "entropy": 5.7474006652832035,
      "epoch": 4.063619680153806,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034375190542968175,
      "loss": 5.2313,
      "mean_token_accuracy": 0.2135878786444664,
      "num_tokens": 94220390.0,
      "step": 46500
    },
    {
      "entropy": 5.7770920753479,
      "epoch": 4.064056628506511,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034372223405999206,
      "loss": 5.2878,
      "mean_token_accuracy": 0.20254841297864914,
      "num_tokens": 94230904.0,
      "step": 46505
    },
    {
      "entropy": 5.793235158920288,
      "epoch": 4.064493576859215,
      "grad_norm": 1.171875,
      "learning_rate": 0.000343692561372174,
      "loss": 5.2333,
      "mean_token_accuracy": 0.21032118648290635,
      "num_tokens": 94240817.0,
      "step": 46510
    },
    {
      "entropy": 5.743224048614502,
      "epoch": 4.06493052521192,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034366288736679653,
      "loss": 5.3059,
      "mean_token_accuracy": 0.20491744130849837,
      "num_tokens": 94251972.0,
      "step": 46515
    },
    {
      "entropy": 5.772879600524902,
      "epoch": 4.065367473564625,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003436332120444289,
      "loss": 5.2847,
      "mean_token_accuracy": 0.20450642704963684,
      "num_tokens": 94261450.0,
      "step": 46520
    },
    {
      "entropy": 5.80766863822937,
      "epoch": 4.0658044219173295,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034360353540564045,
      "loss": 5.2468,
      "mean_token_accuracy": 0.2083773136138916,
      "num_tokens": 94271174.0,
      "step": 46525
    },
    {
      "entropy": 5.752852725982666,
      "epoch": 4.066241370270034,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034357385745100013,
      "loss": 5.1614,
      "mean_token_accuracy": 0.21763156950473786,
      "num_tokens": 94281609.0,
      "step": 46530
    },
    {
      "entropy": 5.705964231491089,
      "epoch": 4.066678318622738,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034354417818107725,
      "loss": 5.245,
      "mean_token_accuracy": 0.20016942471265792,
      "num_tokens": 94292704.0,
      "step": 46535
    },
    {
      "entropy": 5.7116282939910885,
      "epoch": 4.067115266975444,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034351449759644116,
      "loss": 5.1911,
      "mean_token_accuracy": 0.20996023416519166,
      "num_tokens": 94302467.0,
      "step": 46540
    },
    {
      "entropy": 5.739780426025391,
      "epoch": 4.067552215328148,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003434848156976612,
      "loss": 5.2062,
      "mean_token_accuracy": 0.2123317763209343,
      "num_tokens": 94312365.0,
      "step": 46545
    },
    {
      "entropy": 5.727547931671142,
      "epoch": 4.067989163680853,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034345513248530647,
      "loss": 5.2168,
      "mean_token_accuracy": 0.20936757177114487,
      "num_tokens": 94323168.0,
      "step": 46550
    },
    {
      "entropy": 5.73716926574707,
      "epoch": 4.068426112033558,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003434254479599464,
      "loss": 5.265,
      "mean_token_accuracy": 0.20481614470481874,
      "num_tokens": 94333901.0,
      "step": 46555
    },
    {
      "entropy": 5.742246961593628,
      "epoch": 4.0688630603862626,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003433957621221504,
      "loss": 5.2303,
      "mean_token_accuracy": 0.21185799539089203,
      "num_tokens": 94344565.0,
      "step": 46560
    },
    {
      "entropy": 5.732172870635987,
      "epoch": 4.069300008738967,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003433660749724879,
      "loss": 5.2437,
      "mean_token_accuracy": 0.206267149746418,
      "num_tokens": 94354330.0,
      "step": 46565
    },
    {
      "entropy": 5.7347570896148685,
      "epoch": 4.0697369570916715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034333638651152826,
      "loss": 5.2667,
      "mean_token_accuracy": 0.1996844083070755,
      "num_tokens": 94364845.0,
      "step": 46570
    },
    {
      "entropy": 5.786379671096801,
      "epoch": 4.070173905444377,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034330669673984093,
      "loss": 5.2366,
      "mean_token_accuracy": 0.20525171160697936,
      "num_tokens": 94374932.0,
      "step": 46575
    },
    {
      "entropy": 5.761125278472901,
      "epoch": 4.070610853797081,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034327700565799535,
      "loss": 5.2603,
      "mean_token_accuracy": 0.21116049885749816,
      "num_tokens": 94385431.0,
      "step": 46580
    },
    {
      "entropy": 5.828136253356933,
      "epoch": 4.071047802149786,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034324731326656094,
      "loss": 5.3086,
      "mean_token_accuracy": 0.2008875399827957,
      "num_tokens": 94395727.0,
      "step": 46585
    },
    {
      "entropy": 5.721792268753052,
      "epoch": 4.07148475050249,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034321761956610734,
      "loss": 5.2661,
      "mean_token_accuracy": 0.20374234914779663,
      "num_tokens": 94406562.0,
      "step": 46590
    },
    {
      "entropy": 5.744830513000489,
      "epoch": 4.071921698855196,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034318792455720413,
      "loss": 5.218,
      "mean_token_accuracy": 0.21122362613677978,
      "num_tokens": 94416462.0,
      "step": 46595
    },
    {
      "entropy": 5.789189386367798,
      "epoch": 4.0723586472079,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034315822824042074,
      "loss": 5.2064,
      "mean_token_accuracy": 0.21240197718143464,
      "num_tokens": 94426913.0,
      "step": 46600
    },
    {
      "entropy": 5.774127388000489,
      "epoch": 4.0727955955606046,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034312853061632686,
      "loss": 5.193,
      "mean_token_accuracy": 0.21374954879283906,
      "num_tokens": 94436622.0,
      "step": 46605
    },
    {
      "entropy": 5.722853899002075,
      "epoch": 4.073232543913309,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000343098831685492,
      "loss": 5.2357,
      "mean_token_accuracy": 0.2062659442424774,
      "num_tokens": 94447027.0,
      "step": 46610
    },
    {
      "entropy": 5.790110111236572,
      "epoch": 4.073669492266014,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003430691314484859,
      "loss": 5.2405,
      "mean_token_accuracy": 0.20043047964572908,
      "num_tokens": 94456085.0,
      "step": 46615
    },
    {
      "entropy": 5.802597522735596,
      "epoch": 4.074106440618719,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003430394299058782,
      "loss": 5.2989,
      "mean_token_accuracy": 0.20529714077711106,
      "num_tokens": 94466497.0,
      "step": 46620
    },
    {
      "entropy": 5.755136108398437,
      "epoch": 4.074543388971423,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034300972705823856,
      "loss": 5.2589,
      "mean_token_accuracy": 0.19979682117700576,
      "num_tokens": 94477392.0,
      "step": 46625
    },
    {
      "entropy": 5.739276790618897,
      "epoch": 4.074980337324129,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034298002290613673,
      "loss": 5.2254,
      "mean_token_accuracy": 0.21396341621875764,
      "num_tokens": 94486620.0,
      "step": 46630
    },
    {
      "entropy": 5.702511119842529,
      "epoch": 4.075417285676833,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003429503174501424,
      "loss": 5.1342,
      "mean_token_accuracy": 0.21528539657592774,
      "num_tokens": 94497241.0,
      "step": 46635
    },
    {
      "entropy": 5.737957286834717,
      "epoch": 4.075854234029538,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003429206106908254,
      "loss": 5.2318,
      "mean_token_accuracy": 0.20946688950061798,
      "num_tokens": 94505885.0,
      "step": 46640
    },
    {
      "entropy": 5.733246755599976,
      "epoch": 4.076291182382242,
      "grad_norm": 1.125,
      "learning_rate": 0.00034289090262875553,
      "loss": 5.2032,
      "mean_token_accuracy": 0.21234239339828492,
      "num_tokens": 94515985.0,
      "step": 46645
    },
    {
      "entropy": 5.764606571197509,
      "epoch": 4.076728130734947,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003428611932645027,
      "loss": 5.2862,
      "mean_token_accuracy": 0.19665537178516387,
      "num_tokens": 94526321.0,
      "step": 46650
    },
    {
      "entropy": 5.739831638336182,
      "epoch": 4.077165079087652,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003428314825986364,
      "loss": 5.2022,
      "mean_token_accuracy": 0.20694276541471482,
      "num_tokens": 94536366.0,
      "step": 46655
    },
    {
      "entropy": 5.739323282241822,
      "epoch": 4.077602027440356,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003428017706317268,
      "loss": 5.2195,
      "mean_token_accuracy": 0.209042090177536,
      "num_tokens": 94546749.0,
      "step": 46660
    },
    {
      "entropy": 5.732594108581543,
      "epoch": 4.078038975793061,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034277205736434367,
      "loss": 5.2562,
      "mean_token_accuracy": 0.20599042773246765,
      "num_tokens": 94558013.0,
      "step": 46665
    },
    {
      "entropy": 5.770131397247314,
      "epoch": 4.078475924145766,
      "grad_norm": 1.125,
      "learning_rate": 0.000342742342797057,
      "loss": 5.2073,
      "mean_token_accuracy": 0.20817407816648484,
      "num_tokens": 94567417.0,
      "step": 46670
    },
    {
      "entropy": 5.667985820770264,
      "epoch": 4.078912872498471,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003427126269304367,
      "loss": 5.1501,
      "mean_token_accuracy": 0.2143148109316826,
      "num_tokens": 94577598.0,
      "step": 46675
    },
    {
      "entropy": 5.763435554504395,
      "epoch": 4.079349820851175,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003426829097650527,
      "loss": 5.2763,
      "mean_token_accuracy": 0.20556045174598694,
      "num_tokens": 94587260.0,
      "step": 46680
    },
    {
      "entropy": 5.812968730926514,
      "epoch": 4.0797867692038805,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034265319130147515,
      "loss": 5.228,
      "mean_token_accuracy": 0.2042794406414032,
      "num_tokens": 94598266.0,
      "step": 46685
    },
    {
      "entropy": 5.744569492340088,
      "epoch": 4.080223717556585,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003426234715402738,
      "loss": 5.2515,
      "mean_token_accuracy": 0.20881084352731705,
      "num_tokens": 94608468.0,
      "step": 46690
    },
    {
      "entropy": 5.733242321014404,
      "epoch": 4.080660665909289,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003425937504820189,
      "loss": 5.2692,
      "mean_token_accuracy": 0.2113519698381424,
      "num_tokens": 94618109.0,
      "step": 46695
    },
    {
      "entropy": 5.8061658382415775,
      "epoch": 4.081097614261994,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034256402812728046,
      "loss": 5.3194,
      "mean_token_accuracy": 0.1944749817252159,
      "num_tokens": 94628763.0,
      "step": 46700
    },
    {
      "entropy": 5.784065818786621,
      "epoch": 4.081534562614699,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003425343044766285,
      "loss": 5.2956,
      "mean_token_accuracy": 0.20255217254161834,
      "num_tokens": 94638607.0,
      "step": 46705
    },
    {
      "entropy": 5.778526496887207,
      "epoch": 4.081971510967404,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003425045795306333,
      "loss": 5.2378,
      "mean_token_accuracy": 0.20744287371635436,
      "num_tokens": 94649691.0,
      "step": 46710
    },
    {
      "entropy": 5.74263105392456,
      "epoch": 4.082408459320108,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034247485328986477,
      "loss": 5.2272,
      "mean_token_accuracy": 0.2099205583333969,
      "num_tokens": 94659754.0,
      "step": 46715
    },
    {
      "entropy": 5.782049751281738,
      "epoch": 4.082845407672813,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034244512575489324,
      "loss": 5.3249,
      "mean_token_accuracy": 0.19489568024873732,
      "num_tokens": 94670170.0,
      "step": 46720
    },
    {
      "entropy": 5.732559633255005,
      "epoch": 4.083282356025518,
      "grad_norm": 0.96875,
      "learning_rate": 0.00034241539692628887,
      "loss": 5.1261,
      "mean_token_accuracy": 0.21437157839536666,
      "num_tokens": 94680615.0,
      "step": 46725
    },
    {
      "entropy": 5.756051254272461,
      "epoch": 4.0837193043782225,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034238566680462184,
      "loss": 5.2447,
      "mean_token_accuracy": 0.2037610277533531,
      "num_tokens": 94690166.0,
      "step": 46730
    },
    {
      "entropy": 5.7319964408874515,
      "epoch": 4.084156252730927,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034235593539046243,
      "loss": 5.2473,
      "mean_token_accuracy": 0.2083444356918335,
      "num_tokens": 94699953.0,
      "step": 46735
    },
    {
      "entropy": 5.687215137481689,
      "epoch": 4.084593201083632,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003423262026843809,
      "loss": 5.1245,
      "mean_token_accuracy": 0.21875647753477095,
      "num_tokens": 94710030.0,
      "step": 46740
    },
    {
      "entropy": 5.710705137252807,
      "epoch": 4.085030149436337,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034229646868694745,
      "loss": 5.2502,
      "mean_token_accuracy": 0.2020377039909363,
      "num_tokens": 94719777.0,
      "step": 46745
    },
    {
      "entropy": 5.7437488555908205,
      "epoch": 4.085467097789041,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003422667333987326,
      "loss": 5.247,
      "mean_token_accuracy": 0.20320955067873,
      "num_tokens": 94729742.0,
      "step": 46750
    },
    {
      "entropy": 5.862812232971192,
      "epoch": 4.085904046141746,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034223699682030647,
      "loss": 5.3312,
      "mean_token_accuracy": 0.1982862636446953,
      "num_tokens": 94739685.0,
      "step": 46755
    },
    {
      "entropy": 5.6877847671508786,
      "epoch": 4.086340994494451,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003422072589522395,
      "loss": 5.0857,
      "mean_token_accuracy": 0.23395538330078125,
      "num_tokens": 94749133.0,
      "step": 46760
    },
    {
      "entropy": 5.707862472534179,
      "epoch": 4.086777942847156,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003421775197951021,
      "loss": 5.2788,
      "mean_token_accuracy": 0.20476533472537994,
      "num_tokens": 94758918.0,
      "step": 46765
    },
    {
      "entropy": 5.784018182754517,
      "epoch": 4.08721489119986,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034214777934946474,
      "loss": 5.2298,
      "mean_token_accuracy": 0.20733107030391693,
      "num_tokens": 94769879.0,
      "step": 46770
    },
    {
      "entropy": 5.803607368469239,
      "epoch": 4.0876518395525645,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003421180376158977,
      "loss": 5.2279,
      "mean_token_accuracy": 0.2051781475543976,
      "num_tokens": 94778668.0,
      "step": 46775
    },
    {
      "entropy": 5.757917070388794,
      "epoch": 4.08808878790527,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003420882945949715,
      "loss": 5.2956,
      "mean_token_accuracy": 0.20578900575637818,
      "num_tokens": 94789388.0,
      "step": 46780
    },
    {
      "entropy": 5.7099579811096195,
      "epoch": 4.088525736257974,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003420585502872568,
      "loss": 5.1767,
      "mean_token_accuracy": 0.22395751625299454,
      "num_tokens": 94799467.0,
      "step": 46785
    },
    {
      "entropy": 5.778035402297974,
      "epoch": 4.088962684610679,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003420288046933239,
      "loss": 5.2249,
      "mean_token_accuracy": 0.20910367965698243,
      "num_tokens": 94809050.0,
      "step": 46790
    },
    {
      "entropy": 5.741938352584839,
      "epoch": 4.089399632963384,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034199905781374327,
      "loss": 5.1723,
      "mean_token_accuracy": 0.2133411005139351,
      "num_tokens": 94818069.0,
      "step": 46795
    },
    {
      "entropy": 5.711814546585083,
      "epoch": 4.089836581316089,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034196930964908574,
      "loss": 5.2395,
      "mean_token_accuracy": 0.19879794716835023,
      "num_tokens": 94828288.0,
      "step": 46800
    },
    {
      "entropy": 5.723910999298096,
      "epoch": 4.090273529668793,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034193956019992175,
      "loss": 5.2353,
      "mean_token_accuracy": 0.2054239973425865,
      "num_tokens": 94839291.0,
      "step": 46805
    },
    {
      "entropy": 5.733066654205322,
      "epoch": 4.090710478021498,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003419098094668219,
      "loss": 5.1977,
      "mean_token_accuracy": 0.2131073072552681,
      "num_tokens": 94848582.0,
      "step": 46810
    },
    {
      "entropy": 5.775409269332886,
      "epoch": 4.091147426374203,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003418800574503569,
      "loss": 5.2635,
      "mean_token_accuracy": 0.20950798243284224,
      "num_tokens": 94859484.0,
      "step": 46815
    },
    {
      "entropy": 5.730039834976196,
      "epoch": 4.091584374726907,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034185030415109725,
      "loss": 5.1901,
      "mean_token_accuracy": 0.2043784439563751,
      "num_tokens": 94869581.0,
      "step": 46820
    },
    {
      "entropy": 5.712819194793701,
      "epoch": 4.092021323079612,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003418205495696138,
      "loss": 5.2057,
      "mean_token_accuracy": 0.21003184765577315,
      "num_tokens": 94880151.0,
      "step": 46825
    },
    {
      "entropy": 5.627088832855224,
      "epoch": 4.092458271432316,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034179079370647713,
      "loss": 5.1009,
      "mean_token_accuracy": 0.22511874735355378,
      "num_tokens": 94889302.0,
      "step": 46830
    },
    {
      "entropy": 5.77283821105957,
      "epoch": 4.092895219785022,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034176103656225815,
      "loss": 5.2596,
      "mean_token_accuracy": 0.20986385345458985,
      "num_tokens": 94900122.0,
      "step": 46835
    },
    {
      "entropy": 5.782526445388794,
      "epoch": 4.093332168137726,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003417312781375274,
      "loss": 5.2309,
      "mean_token_accuracy": 0.20870917439460754,
      "num_tokens": 94909815.0,
      "step": 46840
    },
    {
      "entropy": 5.748500633239746,
      "epoch": 4.093769116490431,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034170151843285586,
      "loss": 5.2309,
      "mean_token_accuracy": 0.21157913506031037,
      "num_tokens": 94919103.0,
      "step": 46845
    },
    {
      "entropy": 5.713701772689819,
      "epoch": 4.094206064843135,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034167175744881416,
      "loss": 5.1781,
      "mean_token_accuracy": 0.21061504632234573,
      "num_tokens": 94928611.0,
      "step": 46850
    },
    {
      "entropy": 5.71341381072998,
      "epoch": 4.0946430131958405,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003416419951859733,
      "loss": 5.221,
      "mean_token_accuracy": 0.210744009912014,
      "num_tokens": 94938178.0,
      "step": 46855
    },
    {
      "entropy": 5.764531421661377,
      "epoch": 4.095079961548545,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034161223164490403,
      "loss": 5.2499,
      "mean_token_accuracy": 0.2039560168981552,
      "num_tokens": 94947981.0,
      "step": 46860
    },
    {
      "entropy": 5.75673623085022,
      "epoch": 4.095516909901249,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00034158246682617716,
      "loss": 5.2471,
      "mean_token_accuracy": 0.1994270771741867,
      "num_tokens": 94958171.0,
      "step": 46865
    },
    {
      "entropy": 5.735060834884644,
      "epoch": 4.095953858253955,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003415527007303638,
      "loss": 5.1599,
      "mean_token_accuracy": 0.21352011263370513,
      "num_tokens": 94968262.0,
      "step": 46870
    },
    {
      "entropy": 5.748179578781128,
      "epoch": 4.096390806606659,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034152293335803477,
      "loss": 5.2199,
      "mean_token_accuracy": 0.20995606780052184,
      "num_tokens": 94977871.0,
      "step": 46875
    },
    {
      "entropy": 5.74490327835083,
      "epoch": 4.096827754959364,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003414931647097609,
      "loss": 5.2768,
      "mean_token_accuracy": 0.20665108114480973,
      "num_tokens": 94988072.0,
      "step": 46880
    },
    {
      "entropy": 5.7417045593261715,
      "epoch": 4.097264703312068,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034146339478611343,
      "loss": 5.189,
      "mean_token_accuracy": 0.21027769446372985,
      "num_tokens": 94998411.0,
      "step": 46885
    },
    {
      "entropy": 5.831594657897949,
      "epoch": 4.0977016516647735,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003414336235876631,
      "loss": 5.269,
      "mean_token_accuracy": 0.20559000521898269,
      "num_tokens": 95008338.0,
      "step": 46890
    },
    {
      "entropy": 5.72547287940979,
      "epoch": 4.098138600017478,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034140385111498117,
      "loss": 5.2141,
      "mean_token_accuracy": 0.19896989464759826,
      "num_tokens": 95019167.0,
      "step": 46895
    },
    {
      "entropy": 5.768980312347412,
      "epoch": 4.0985755483701825,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034137407736863854,
      "loss": 5.2217,
      "mean_token_accuracy": 0.20709829330444335,
      "num_tokens": 95029219.0,
      "step": 46900
    },
    {
      "entropy": 5.73866081237793,
      "epoch": 4.099012496722887,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003413443023492063,
      "loss": 5.1958,
      "mean_token_accuracy": 0.21325995326042174,
      "num_tokens": 95038430.0,
      "step": 46905
    },
    {
      "entropy": 5.754428768157959,
      "epoch": 4.099449445075592,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003413145260572556,
      "loss": 5.2173,
      "mean_token_accuracy": 0.21063722521066666,
      "num_tokens": 95048444.0,
      "step": 46910
    },
    {
      "entropy": 5.776828575134277,
      "epoch": 4.099886393428297,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034128474849335754,
      "loss": 5.3214,
      "mean_token_accuracy": 0.21160168647766114,
      "num_tokens": 95059427.0,
      "step": 46915
    },
    {
      "entropy": 5.749907970428467,
      "epoch": 4.100323341781001,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034125496965808325,
      "loss": 5.1879,
      "mean_token_accuracy": 0.20872896015644074,
      "num_tokens": 95068306.0,
      "step": 46920
    },
    {
      "entropy": 5.828069686889648,
      "epoch": 4.100760290133707,
      "grad_norm": 1.078125,
      "learning_rate": 0.000341225189552004,
      "loss": 5.3254,
      "mean_token_accuracy": 0.20235807895660402,
      "num_tokens": 95079314.0,
      "step": 46925
    },
    {
      "entropy": 5.786369276046753,
      "epoch": 4.101197238486411,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003411954081756908,
      "loss": 5.2356,
      "mean_token_accuracy": 0.20305826812982558,
      "num_tokens": 95089985.0,
      "step": 46930
    },
    {
      "entropy": 5.7734130859375,
      "epoch": 4.1016341868391155,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034116562552971504,
      "loss": 5.257,
      "mean_token_accuracy": 0.20387701243162154,
      "num_tokens": 95099694.0,
      "step": 46935
    },
    {
      "entropy": 5.711562776565552,
      "epoch": 4.10207113519182,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034113584161464793,
      "loss": 5.2414,
      "mean_token_accuracy": 0.20954986959695815,
      "num_tokens": 95110585.0,
      "step": 46940
    },
    {
      "entropy": 5.709229326248169,
      "epoch": 4.102508083544525,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003411060564310607,
      "loss": 5.188,
      "mean_token_accuracy": 0.21315813511610032,
      "num_tokens": 95120003.0,
      "step": 46945
    },
    {
      "entropy": 5.746845531463623,
      "epoch": 4.10294503189723,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034107626997952474,
      "loss": 5.1638,
      "mean_token_accuracy": 0.20981065034866334,
      "num_tokens": 95130697.0,
      "step": 46950
    },
    {
      "entropy": 5.813325357437134,
      "epoch": 4.103381980249934,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003410464822606112,
      "loss": 5.3485,
      "mean_token_accuracy": 0.19650222212076188,
      "num_tokens": 95141044.0,
      "step": 46955
    },
    {
      "entropy": 5.814549303054809,
      "epoch": 4.103818928602639,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003410166932748915,
      "loss": 5.2756,
      "mean_token_accuracy": 0.19611888378858566,
      "num_tokens": 95151046.0,
      "step": 46960
    },
    {
      "entropy": 5.76361985206604,
      "epoch": 4.104255876955344,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003409869030229371,
      "loss": 5.2256,
      "mean_token_accuracy": 0.2055207222700119,
      "num_tokens": 95160743.0,
      "step": 46965
    },
    {
      "entropy": 5.69626817703247,
      "epoch": 4.104692825308049,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003409571115053193,
      "loss": 5.1512,
      "mean_token_accuracy": 0.2073998987674713,
      "num_tokens": 95170437.0,
      "step": 46970
    },
    {
      "entropy": 5.73914361000061,
      "epoch": 4.105129773660753,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034092731872260965,
      "loss": 5.266,
      "mean_token_accuracy": 0.20684365928173065,
      "num_tokens": 95179734.0,
      "step": 46975
    },
    {
      "entropy": 5.786707830429077,
      "epoch": 4.105566722013458,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003408975246753793,
      "loss": 5.2614,
      "mean_token_accuracy": 0.20165977776050567,
      "num_tokens": 95190401.0,
      "step": 46980
    },
    {
      "entropy": 5.877245807647705,
      "epoch": 4.106003670366163,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003408677293642,
      "loss": 5.3214,
      "mean_token_accuracy": 0.18933287560939788,
      "num_tokens": 95200629.0,
      "step": 46985
    },
    {
      "entropy": 5.725918960571289,
      "epoch": 4.106440618718867,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003408379327896431,
      "loss": 5.2284,
      "mean_token_accuracy": 0.20751775354146956,
      "num_tokens": 95210858.0,
      "step": 46990
    },
    {
      "entropy": 5.757088947296142,
      "epoch": 4.106877567071572,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034080813495228026,
      "loss": 5.2897,
      "mean_token_accuracy": 0.19815145879983903,
      "num_tokens": 95221531.0,
      "step": 46995
    },
    {
      "entropy": 5.71357479095459,
      "epoch": 4.107314515424277,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034077833585268276,
      "loss": 5.2083,
      "mean_token_accuracy": 0.2088519290089607,
      "num_tokens": 95231689.0,
      "step": 47000
    },
    {
      "entropy": 5.783670234680176,
      "epoch": 4.107751463776982,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034074853549142235,
      "loss": 5.2629,
      "mean_token_accuracy": 0.20144194811582566,
      "num_tokens": 95242462.0,
      "step": 47005
    },
    {
      "entropy": 5.7018708229064945,
      "epoch": 4.108188412129686,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003407187338690706,
      "loss": 5.136,
      "mean_token_accuracy": 0.21093427538871765,
      "num_tokens": 95252063.0,
      "step": 47010
    },
    {
      "entropy": 5.764369869232178,
      "epoch": 4.108625360482391,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034068893098619906,
      "loss": 5.233,
      "mean_token_accuracy": 0.20220719575881957,
      "num_tokens": 95262047.0,
      "step": 47015
    },
    {
      "entropy": 5.637299394607544,
      "epoch": 4.109062308835096,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034065912684337954,
      "loss": 5.1183,
      "mean_token_accuracy": 0.2134103998541832,
      "num_tokens": 95272117.0,
      "step": 47020
    },
    {
      "entropy": 5.850068378448486,
      "epoch": 4.1094992571878,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034062932144118336,
      "loss": 5.3318,
      "mean_token_accuracy": 0.1990694135427475,
      "num_tokens": 95281477.0,
      "step": 47025
    },
    {
      "entropy": 5.756460952758789,
      "epoch": 4.109936205540505,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003405995147801825,
      "loss": 5.2216,
      "mean_token_accuracy": 0.21127833425998688,
      "num_tokens": 95291140.0,
      "step": 47030
    },
    {
      "entropy": 5.720225715637207,
      "epoch": 4.11037315389321,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034056970686094856,
      "loss": 5.2194,
      "mean_token_accuracy": 0.2103511646389961,
      "num_tokens": 95301256.0,
      "step": 47035
    },
    {
      "entropy": 5.806152868270874,
      "epoch": 4.110810102245915,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034053989768405334,
      "loss": 5.3082,
      "mean_token_accuracy": 0.20078431218862533,
      "num_tokens": 95311232.0,
      "step": 47040
    },
    {
      "entropy": 5.764930486679077,
      "epoch": 4.111247050598619,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003405100872500685,
      "loss": 5.2128,
      "mean_token_accuracy": 0.21002316772937774,
      "num_tokens": 95321578.0,
      "step": 47045
    },
    {
      "entropy": 5.763190746307373,
      "epoch": 4.111683998951324,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003404802755595658,
      "loss": 5.2372,
      "mean_token_accuracy": 0.2023852989077568,
      "num_tokens": 95330837.0,
      "step": 47050
    },
    {
      "entropy": 5.734825897216797,
      "epoch": 4.112120947304029,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003404504626131172,
      "loss": 5.1899,
      "mean_token_accuracy": 0.20838079005479812,
      "num_tokens": 95341778.0,
      "step": 47055
    },
    {
      "entropy": 5.7674181938171385,
      "epoch": 4.1125578956567335,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003404206484112944,
      "loss": 5.252,
      "mean_token_accuracy": 0.20576618760824203,
      "num_tokens": 95353629.0,
      "step": 47060
    },
    {
      "entropy": 5.7364706039428714,
      "epoch": 4.112994844009438,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034039083295466925,
      "loss": 5.3294,
      "mean_token_accuracy": 0.2006623238325119,
      "num_tokens": 95364778.0,
      "step": 47065
    },
    {
      "entropy": 5.751992082595825,
      "epoch": 4.113431792362142,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003403610162438137,
      "loss": 5.2341,
      "mean_token_accuracy": 0.2024528592824936,
      "num_tokens": 95375354.0,
      "step": 47070
    },
    {
      "entropy": 5.7897439956665036,
      "epoch": 4.113868740714848,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034033119827929964,
      "loss": 5.2485,
      "mean_token_accuracy": 0.20844241827726365,
      "num_tokens": 95385864.0,
      "step": 47075
    },
    {
      "entropy": 5.752073764801025,
      "epoch": 4.114305689067552,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003403013790616989,
      "loss": 5.235,
      "mean_token_accuracy": 0.206936839222908,
      "num_tokens": 95397443.0,
      "step": 47080
    },
    {
      "entropy": 5.769317007064819,
      "epoch": 4.114742637420257,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003402715585915835,
      "loss": 5.273,
      "mean_token_accuracy": 0.20348627418279647,
      "num_tokens": 95407322.0,
      "step": 47085
    },
    {
      "entropy": 5.773214387893677,
      "epoch": 4.115179585772962,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003402417368695255,
      "loss": 5.2296,
      "mean_token_accuracy": 0.20933317095041276,
      "num_tokens": 95417088.0,
      "step": 47090
    },
    {
      "entropy": 5.710595512390137,
      "epoch": 4.1156165341256665,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003402119138960967,
      "loss": 5.1851,
      "mean_token_accuracy": 0.21685346812009812,
      "num_tokens": 95426610.0,
      "step": 47095
    },
    {
      "entropy": 5.7655469417572025,
      "epoch": 4.116053482478371,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003401820896718692,
      "loss": 5.2565,
      "mean_token_accuracy": 0.2064650610089302,
      "num_tokens": 95436780.0,
      "step": 47100
    },
    {
      "entropy": 5.787498712539673,
      "epoch": 4.1164904308310755,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003401522641974151,
      "loss": 5.2444,
      "mean_token_accuracy": 0.2060854360461235,
      "num_tokens": 95446143.0,
      "step": 47105
    },
    {
      "entropy": 5.744511556625366,
      "epoch": 4.116927379183781,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034012243747330643,
      "loss": 5.1995,
      "mean_token_accuracy": 0.21280380338430405,
      "num_tokens": 95456609.0,
      "step": 47110
    },
    {
      "entropy": 5.688949632644653,
      "epoch": 4.117364327536485,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003400926095001153,
      "loss": 5.209,
      "mean_token_accuracy": 0.20774987041950227,
      "num_tokens": 95466788.0,
      "step": 47115
    },
    {
      "entropy": 5.714930868148803,
      "epoch": 4.11780127588919,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034006278027841385,
      "loss": 5.208,
      "mean_token_accuracy": 0.213109989464283,
      "num_tokens": 95476765.0,
      "step": 47120
    },
    {
      "entropy": 5.780306959152222,
      "epoch": 4.118238224241894,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034003294980877415,
      "loss": 5.2673,
      "mean_token_accuracy": 0.20715134888887404,
      "num_tokens": 95486604.0,
      "step": 47125
    },
    {
      "entropy": 5.695416879653931,
      "epoch": 4.1186751725946,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034000311809176837,
      "loss": 5.1639,
      "mean_token_accuracy": 0.21432343274354934,
      "num_tokens": 95496258.0,
      "step": 47130
    },
    {
      "entropy": 5.7371069431304935,
      "epoch": 4.119112120947304,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033997328512796863,
      "loss": 5.2084,
      "mean_token_accuracy": 0.20427343845367432,
      "num_tokens": 95506493.0,
      "step": 47135
    },
    {
      "entropy": 5.71157169342041,
      "epoch": 4.1195490693000085,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003399434509179475,
      "loss": 5.2203,
      "mean_token_accuracy": 0.2107591912150383,
      "num_tokens": 95516374.0,
      "step": 47140
    },
    {
      "entropy": 5.725535535812378,
      "epoch": 4.119986017652714,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00033991361546227675,
      "loss": 5.2546,
      "mean_token_accuracy": 0.20584553033113479,
      "num_tokens": 95525958.0,
      "step": 47145
    },
    {
      "entropy": 5.752416133880615,
      "epoch": 4.120422966005418,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033988377876152887,
      "loss": 5.1632,
      "mean_token_accuracy": 0.21571978330612182,
      "num_tokens": 95535703.0,
      "step": 47150
    },
    {
      "entropy": 5.70336389541626,
      "epoch": 4.120859914358123,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033985394081627605,
      "loss": 5.1805,
      "mean_token_accuracy": 0.2086760550737381,
      "num_tokens": 95545821.0,
      "step": 47155
    },
    {
      "entropy": 5.741788053512574,
      "epoch": 4.121296862710827,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003398241016270909,
      "loss": 5.1713,
      "mean_token_accuracy": 0.2086901769042015,
      "num_tokens": 95556095.0,
      "step": 47160
    },
    {
      "entropy": 5.828491878509522,
      "epoch": 4.121733811063533,
      "grad_norm": 1.125,
      "learning_rate": 0.0003397942611945453,
      "loss": 5.3055,
      "mean_token_accuracy": 0.19800201654434205,
      "num_tokens": 95566695.0,
      "step": 47165
    },
    {
      "entropy": 5.705971002578735,
      "epoch": 4.122170759416237,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003397644195192119,
      "loss": 5.2341,
      "mean_token_accuracy": 0.20388715863227844,
      "num_tokens": 95576525.0,
      "step": 47170
    },
    {
      "entropy": 5.750913190841675,
      "epoch": 4.122607707768942,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033973457660166297,
      "loss": 5.2118,
      "mean_token_accuracy": 0.20070724487304686,
      "num_tokens": 95586121.0,
      "step": 47175
    },
    {
      "entropy": 5.77387375831604,
      "epoch": 4.123044656121646,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003397047324424709,
      "loss": 5.2469,
      "mean_token_accuracy": 0.20588068813085555,
      "num_tokens": 95595694.0,
      "step": 47180
    },
    {
      "entropy": 5.7060870170593265,
      "epoch": 4.123481604474351,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033967488704220823,
      "loss": 5.1759,
      "mean_token_accuracy": 0.2129165858030319,
      "num_tokens": 95607114.0,
      "step": 47185
    },
    {
      "entropy": 5.749667453765869,
      "epoch": 4.123918552827056,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033964504040144734,
      "loss": 5.2275,
      "mean_token_accuracy": 0.20652742981910704,
      "num_tokens": 95617476.0,
      "step": 47190
    },
    {
      "entropy": 5.761335039138794,
      "epoch": 4.12435550117976,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003396151925207606,
      "loss": 5.2815,
      "mean_token_accuracy": 0.20692982971668245,
      "num_tokens": 95627205.0,
      "step": 47195
    },
    {
      "entropy": 5.7770082473754885,
      "epoch": 4.124792449532466,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003395853434007207,
      "loss": 5.2421,
      "mean_token_accuracy": 0.20842745006084443,
      "num_tokens": 95637798.0,
      "step": 47200
    },
    {
      "entropy": 5.7230369567871096,
      "epoch": 4.12522939788517,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033955549304190007,
      "loss": 5.16,
      "mean_token_accuracy": 0.21171444654464722,
      "num_tokens": 95647615.0,
      "step": 47205
    },
    {
      "entropy": 5.72765851020813,
      "epoch": 4.125666346237875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003395256414448712,
      "loss": 5.1813,
      "mean_token_accuracy": 0.208975388109684,
      "num_tokens": 95657019.0,
      "step": 47210
    },
    {
      "entropy": 5.712440538406372,
      "epoch": 4.126103294590579,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033949578861020675,
      "loss": 5.1757,
      "mean_token_accuracy": 0.2138697773218155,
      "num_tokens": 95667378.0,
      "step": 47215
    },
    {
      "entropy": 5.716198921203613,
      "epoch": 4.1265402429432845,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033946593453847927,
      "loss": 5.1912,
      "mean_token_accuracy": 0.2072631374001503,
      "num_tokens": 95678078.0,
      "step": 47220
    },
    {
      "entropy": 5.756351709365845,
      "epoch": 4.126977191295989,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003394360792302614,
      "loss": 5.2557,
      "mean_token_accuracy": 0.20604839473962783,
      "num_tokens": 95687450.0,
      "step": 47225
    },
    {
      "entropy": 5.772352600097657,
      "epoch": 4.127414139648693,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033940622268612583,
      "loss": 5.2721,
      "mean_token_accuracy": 0.2008348822593689,
      "num_tokens": 95698594.0,
      "step": 47230
    },
    {
      "entropy": 5.749399042129516,
      "epoch": 4.127851088001398,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033937636490664506,
      "loss": 5.2193,
      "mean_token_accuracy": 0.20435144305229186,
      "num_tokens": 95708708.0,
      "step": 47235
    },
    {
      "entropy": 5.742055082321167,
      "epoch": 4.128288036354103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003393465058923919,
      "loss": 5.2133,
      "mean_token_accuracy": 0.2077555313706398,
      "num_tokens": 95718367.0,
      "step": 47240
    },
    {
      "entropy": 5.769090938568115,
      "epoch": 4.128724984706808,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033931664564393907,
      "loss": 5.296,
      "mean_token_accuracy": 0.19617335945367814,
      "num_tokens": 95728371.0,
      "step": 47245
    },
    {
      "entropy": 5.774092769622802,
      "epoch": 4.129161933059512,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00033928678416185926,
      "loss": 5.2329,
      "mean_token_accuracy": 0.2002217873930931,
      "num_tokens": 95738170.0,
      "step": 47250
    },
    {
      "entropy": 5.647271537780762,
      "epoch": 4.129598881412217,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033925692144672526,
      "loss": 5.0919,
      "mean_token_accuracy": 0.22470621764659882,
      "num_tokens": 95748088.0,
      "step": 47255
    },
    {
      "entropy": 5.753188943862915,
      "epoch": 4.130035829764922,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033922705749910975,
      "loss": 5.2055,
      "mean_token_accuracy": 0.21159370392560958,
      "num_tokens": 95758033.0,
      "step": 47260
    },
    {
      "entropy": 5.780332136154175,
      "epoch": 4.1304727781176265,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003391971923195858,
      "loss": 5.3472,
      "mean_token_accuracy": 0.1999533548951149,
      "num_tokens": 95768261.0,
      "step": 47265
    },
    {
      "entropy": 5.765506935119629,
      "epoch": 4.130909726470331,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003391673259087259,
      "loss": 5.1759,
      "mean_token_accuracy": 0.20867911279201506,
      "num_tokens": 95777739.0,
      "step": 47270
    },
    {
      "entropy": 5.759853839874268,
      "epoch": 4.131346674823036,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003391374582671031,
      "loss": 5.2426,
      "mean_token_accuracy": 0.20662944614887238,
      "num_tokens": 95787683.0,
      "step": 47275
    },
    {
      "entropy": 5.796609497070312,
      "epoch": 4.131783623175741,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003391075893952903,
      "loss": 5.2853,
      "mean_token_accuracy": 0.20042867064476014,
      "num_tokens": 95798411.0,
      "step": 47280
    },
    {
      "entropy": 5.714182996749878,
      "epoch": 4.132220571528445,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003390777192938603,
      "loss": 5.1825,
      "mean_token_accuracy": 0.21181751042604446,
      "num_tokens": 95808869.0,
      "step": 47285
    },
    {
      "entropy": 5.77968578338623,
      "epoch": 4.13265751988115,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033904784796338607,
      "loss": 5.2059,
      "mean_token_accuracy": 0.21005183607339858,
      "num_tokens": 95818808.0,
      "step": 47290
    },
    {
      "entropy": 5.735262823104859,
      "epoch": 4.133094468233855,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003390179754044406,
      "loss": 5.255,
      "mean_token_accuracy": 0.20364274680614472,
      "num_tokens": 95828314.0,
      "step": 47295
    },
    {
      "entropy": 5.761210823059082,
      "epoch": 4.1335314165865595,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003389881016175968,
      "loss": 5.2292,
      "mean_token_accuracy": 0.20784345120191575,
      "num_tokens": 95837727.0,
      "step": 47300
    },
    {
      "entropy": 5.783284616470337,
      "epoch": 4.133968364939264,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033895822660342763,
      "loss": 5.2393,
      "mean_token_accuracy": 0.20648956298828125,
      "num_tokens": 95847871.0,
      "step": 47305
    },
    {
      "entropy": 5.694911670684815,
      "epoch": 4.1344053132919685,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033892835036250616,
      "loss": 5.2416,
      "mean_token_accuracy": 0.20589233487844466,
      "num_tokens": 95858147.0,
      "step": 47310
    },
    {
      "entropy": 5.734702491760254,
      "epoch": 4.134842261644674,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033889847289540547,
      "loss": 5.1678,
      "mean_token_accuracy": 0.20722950994968414,
      "num_tokens": 95867588.0,
      "step": 47315
    },
    {
      "entropy": 5.781117105484009,
      "epoch": 4.135279209997378,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033886859420269864,
      "loss": 5.2928,
      "mean_token_accuracy": 0.20053352862596513,
      "num_tokens": 95878193.0,
      "step": 47320
    },
    {
      "entropy": 5.75648078918457,
      "epoch": 4.135716158350083,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033883871428495865,
      "loss": 5.2925,
      "mean_token_accuracy": 0.20205021202564238,
      "num_tokens": 95888945.0,
      "step": 47325
    },
    {
      "entropy": 5.763395214080811,
      "epoch": 4.136153106702788,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003388088331427586,
      "loss": 5.2295,
      "mean_token_accuracy": 0.2084558218717575,
      "num_tokens": 95898476.0,
      "step": 47330
    },
    {
      "entropy": 5.719338130950928,
      "epoch": 4.136590055055493,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003387789507766718,
      "loss": 5.1894,
      "mean_token_accuracy": 0.21490229815244674,
      "num_tokens": 95907781.0,
      "step": 47335
    },
    {
      "entropy": 5.768698978424072,
      "epoch": 4.137027003408197,
      "grad_norm": 1.125,
      "learning_rate": 0.0003387490671872712,
      "loss": 5.2669,
      "mean_token_accuracy": 0.1977965459227562,
      "num_tokens": 95917558.0,
      "step": 47340
    },
    {
      "entropy": 5.749902725219727,
      "epoch": 4.1374639517609015,
      "grad_norm": 1.125,
      "learning_rate": 0.00033871918237513016,
      "loss": 5.2475,
      "mean_token_accuracy": 0.20297166258096694,
      "num_tokens": 95927862.0,
      "step": 47345
    },
    {
      "entropy": 5.715696096420288,
      "epoch": 4.137900900113607,
      "grad_norm": 1.125,
      "learning_rate": 0.00033868929634082187,
      "loss": 5.1901,
      "mean_token_accuracy": 0.20857318639755248,
      "num_tokens": 95937848.0,
      "step": 47350
    },
    {
      "entropy": 5.733705234527588,
      "epoch": 4.138337848466311,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003386594090849194,
      "loss": 5.2831,
      "mean_token_accuracy": 0.20679229497909546,
      "num_tokens": 95948013.0,
      "step": 47355
    },
    {
      "entropy": 5.7119481563568115,
      "epoch": 4.138774796819016,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003386295206079961,
      "loss": 5.1757,
      "mean_token_accuracy": 0.2175353243947029,
      "num_tokens": 95957483.0,
      "step": 47360
    },
    {
      "entropy": 5.720312690734863,
      "epoch": 4.13921174517172,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033859963091062537,
      "loss": 5.1634,
      "mean_token_accuracy": 0.21180816143751144,
      "num_tokens": 95966940.0,
      "step": 47365
    },
    {
      "entropy": 5.753624296188354,
      "epoch": 4.139648693524426,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033856973999338026,
      "loss": 5.1883,
      "mean_token_accuracy": 0.2059280604124069,
      "num_tokens": 95977634.0,
      "step": 47370
    },
    {
      "entropy": 5.740616750717163,
      "epoch": 4.14008564187713,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033853984785683433,
      "loss": 5.2716,
      "mean_token_accuracy": 0.20370829850435257,
      "num_tokens": 95987114.0,
      "step": 47375
    },
    {
      "entropy": 5.772904682159424,
      "epoch": 4.140522590229835,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003385099545015608,
      "loss": 5.2208,
      "mean_token_accuracy": 0.21220581084489823,
      "num_tokens": 95997051.0,
      "step": 47380
    },
    {
      "entropy": 5.7915623664855955,
      "epoch": 4.14095953858254,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033848005992813304,
      "loss": 5.2431,
      "mean_token_accuracy": 0.20204942524433137,
      "num_tokens": 96007160.0,
      "step": 47385
    },
    {
      "entropy": 5.683110952377319,
      "epoch": 4.141396486935244,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003384501641371244,
      "loss": 5.1694,
      "mean_token_accuracy": 0.2109895020723343,
      "num_tokens": 96017448.0,
      "step": 47390
    },
    {
      "entropy": 5.715445184707642,
      "epoch": 4.141833435287949,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003384202671291085,
      "loss": 5.1747,
      "mean_token_accuracy": 0.2174738973379135,
      "num_tokens": 96027061.0,
      "step": 47395
    },
    {
      "entropy": 5.702976083755493,
      "epoch": 4.142270383640653,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033839036890465855,
      "loss": 5.2326,
      "mean_token_accuracy": 0.20933985561132432,
      "num_tokens": 96036843.0,
      "step": 47400
    },
    {
      "entropy": 5.7311046600341795,
      "epoch": 4.142707331993359,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033836046946434815,
      "loss": 5.2104,
      "mean_token_accuracy": 0.2096848487854004,
      "num_tokens": 96047370.0,
      "step": 47405
    },
    {
      "entropy": 5.747100353240967,
      "epoch": 4.143144280346063,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003383305688087507,
      "loss": 5.2043,
      "mean_token_accuracy": 0.20979514718055725,
      "num_tokens": 96056759.0,
      "step": 47410
    },
    {
      "entropy": 5.718242931365967,
      "epoch": 4.143581228698768,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003383006669384398,
      "loss": 5.1792,
      "mean_token_accuracy": 0.21561333984136583,
      "num_tokens": 96067327.0,
      "step": 47415
    },
    {
      "entropy": 5.693504714965821,
      "epoch": 4.144018177051472,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033827076385398896,
      "loss": 5.1408,
      "mean_token_accuracy": 0.21323636919260025,
      "num_tokens": 96076818.0,
      "step": 47420
    },
    {
      "entropy": 5.853088998794556,
      "epoch": 4.1444551254041775,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003382408595559717,
      "loss": 5.2964,
      "mean_token_accuracy": 0.19735591858625412,
      "num_tokens": 96087490.0,
      "step": 47425
    },
    {
      "entropy": 5.80652437210083,
      "epoch": 4.144892073756882,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003382109540449616,
      "loss": 5.312,
      "mean_token_accuracy": 0.1960014820098877,
      "num_tokens": 96097042.0,
      "step": 47430
    },
    {
      "entropy": 5.767591428756714,
      "epoch": 4.145329022109586,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003381810473215323,
      "loss": 5.2814,
      "mean_token_accuracy": 0.19673639088869094,
      "num_tokens": 96107809.0,
      "step": 47435
    },
    {
      "entropy": 5.770219135284424,
      "epoch": 4.145765970462292,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003381511393862574,
      "loss": 5.2939,
      "mean_token_accuracy": 0.2074240744113922,
      "num_tokens": 96118731.0,
      "step": 47440
    },
    {
      "entropy": 5.7307356834411625,
      "epoch": 4.146202918814996,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003381212302397105,
      "loss": 5.2229,
      "mean_token_accuracy": 0.20855934023857117,
      "num_tokens": 96128832.0,
      "step": 47445
    },
    {
      "entropy": 5.827640008926392,
      "epoch": 4.146639867167701,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033809131988246546,
      "loss": 5.2147,
      "mean_token_accuracy": 0.2116180345416069,
      "num_tokens": 96139537.0,
      "step": 47450
    },
    {
      "entropy": 5.8035500049591064,
      "epoch": 4.147076815520405,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033806140831509567,
      "loss": 5.2349,
      "mean_token_accuracy": 0.2057618662714958,
      "num_tokens": 96150022.0,
      "step": 47455
    },
    {
      "entropy": 5.708599853515625,
      "epoch": 4.1475137638731105,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033803149553817514,
      "loss": 5.1445,
      "mean_token_accuracy": 0.20862346589565278,
      "num_tokens": 96160485.0,
      "step": 47460
    },
    {
      "entropy": 5.665042543411255,
      "epoch": 4.147950712225815,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003380015815522775,
      "loss": 5.1425,
      "mean_token_accuracy": 0.21226089149713517,
      "num_tokens": 96171068.0,
      "step": 47465
    },
    {
      "entropy": 5.743138599395752,
      "epoch": 4.1483876605785195,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003379716663579765,
      "loss": 5.2799,
      "mean_token_accuracy": 0.20310272872447968,
      "num_tokens": 96181448.0,
      "step": 47470
    },
    {
      "entropy": 5.750640344619751,
      "epoch": 4.148824608931224,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033794174995584593,
      "loss": 5.2708,
      "mean_token_accuracy": 0.20755527317523956,
      "num_tokens": 96191463.0,
      "step": 47475
    },
    {
      "entropy": 5.754293537139892,
      "epoch": 4.149261557283929,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033791183234645954,
      "loss": 5.2041,
      "mean_token_accuracy": 0.2090914472937584,
      "num_tokens": 96201559.0,
      "step": 47480
    },
    {
      "entropy": 5.847942304611206,
      "epoch": 4.149698505636634,
      "grad_norm": 0.984375,
      "learning_rate": 0.00033788191353039136,
      "loss": 5.317,
      "mean_token_accuracy": 0.19542235732078553,
      "num_tokens": 96213065.0,
      "step": 47485
    },
    {
      "entropy": 5.765965366363526,
      "epoch": 4.150135453989338,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003378519935082151,
      "loss": 5.2508,
      "mean_token_accuracy": 0.20450380593538284,
      "num_tokens": 96222952.0,
      "step": 47490
    },
    {
      "entropy": 5.8238029956817625,
      "epoch": 4.150572402342043,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033782207228050465,
      "loss": 5.2982,
      "mean_token_accuracy": 0.20075572431087493,
      "num_tokens": 96232745.0,
      "step": 47495
    },
    {
      "entropy": 5.7255912780761715,
      "epoch": 4.151009350694748,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033779214984783396,
      "loss": 5.1613,
      "mean_token_accuracy": 0.2130708634853363,
      "num_tokens": 96243371.0,
      "step": 47500
    },
    {
      "entropy": 5.760625839233398,
      "epoch": 4.1514462990474525,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003377622262107769,
      "loss": 5.2419,
      "mean_token_accuracy": 0.20260681957006454,
      "num_tokens": 96254019.0,
      "step": 47505
    },
    {
      "entropy": 5.774295663833618,
      "epoch": 4.151883247400157,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003377323013699075,
      "loss": 5.1946,
      "mean_token_accuracy": 0.21242360770702362,
      "num_tokens": 96264146.0,
      "step": 47510
    },
    {
      "entropy": 5.7470602035522464,
      "epoch": 4.152320195752862,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003377023753257997,
      "loss": 5.2878,
      "mean_token_accuracy": 0.2081464096903801,
      "num_tokens": 96274831.0,
      "step": 47515
    },
    {
      "entropy": 5.721555757522583,
      "epoch": 4.152757144105567,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003376724480790274,
      "loss": 5.2026,
      "mean_token_accuracy": 0.20653324574232101,
      "num_tokens": 96285692.0,
      "step": 47520
    },
    {
      "entropy": 5.705115699768067,
      "epoch": 4.153194092458271,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003376425196301647,
      "loss": 5.1581,
      "mean_token_accuracy": 0.21609934121370317,
      "num_tokens": 96295896.0,
      "step": 47525
    },
    {
      "entropy": 5.767329168319702,
      "epoch": 4.153631040810976,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033761258997978567,
      "loss": 5.1908,
      "mean_token_accuracy": 0.21169213056564332,
      "num_tokens": 96306287.0,
      "step": 47530
    },
    {
      "entropy": 5.67802586555481,
      "epoch": 4.154067989163681,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00033758265912846444,
      "loss": 5.1591,
      "mean_token_accuracy": 0.20909901112318038,
      "num_tokens": 96316296.0,
      "step": 47535
    },
    {
      "entropy": 5.709608268737793,
      "epoch": 4.154504937516386,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000337552727076775,
      "loss": 5.2338,
      "mean_token_accuracy": 0.19958046376705169,
      "num_tokens": 96325713.0,
      "step": 47540
    },
    {
      "entropy": 5.7711631774902346,
      "epoch": 4.15494188586909,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003375227938252914,
      "loss": 5.2578,
      "mean_token_accuracy": 0.21153353601694108,
      "num_tokens": 96335391.0,
      "step": 47545
    },
    {
      "entropy": 5.825844717025757,
      "epoch": 4.1553788342217945,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003374928593745878,
      "loss": 5.4036,
      "mean_token_accuracy": 0.1907055214047432,
      "num_tokens": 96345919.0,
      "step": 47550
    },
    {
      "entropy": 5.752786588668823,
      "epoch": 4.1558157825745,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033746292372523857,
      "loss": 5.2442,
      "mean_token_accuracy": 0.20674435347318648,
      "num_tokens": 96356296.0,
      "step": 47555
    },
    {
      "entropy": 5.779076719284058,
      "epoch": 4.156252730927204,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033743298687781767,
      "loss": 5.26,
      "mean_token_accuracy": 0.2052823483943939,
      "num_tokens": 96366780.0,
      "step": 47560
    },
    {
      "entropy": 5.786019802093506,
      "epoch": 4.156689679279909,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003374030488328993,
      "loss": 5.2234,
      "mean_token_accuracy": 0.20903733968734742,
      "num_tokens": 96376665.0,
      "step": 47565
    },
    {
      "entropy": 5.716207790374756,
      "epoch": 4.157126627632614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003373731095910578,
      "loss": 5.0846,
      "mean_token_accuracy": 0.22276715636253358,
      "num_tokens": 96386737.0,
      "step": 47570
    },
    {
      "entropy": 5.730334091186523,
      "epoch": 4.157563575985319,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033734316915286737,
      "loss": 5.2161,
      "mean_token_accuracy": 0.21027655601501466,
      "num_tokens": 96397022.0,
      "step": 47575
    },
    {
      "entropy": 5.668311882019043,
      "epoch": 4.158000524338023,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003373132275189023,
      "loss": 5.2002,
      "mean_token_accuracy": 0.20894185751676558,
      "num_tokens": 96406751.0,
      "step": 47580
    },
    {
      "entropy": 5.716224193572998,
      "epoch": 4.158437472690728,
      "grad_norm": 1.0,
      "learning_rate": 0.00033728328468973697,
      "loss": 5.2169,
      "mean_token_accuracy": 0.2120858073234558,
      "num_tokens": 96416987.0,
      "step": 47585
    },
    {
      "entropy": 5.78844952583313,
      "epoch": 4.158874421043433,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003372533406659455,
      "loss": 5.184,
      "mean_token_accuracy": 0.21304971575737,
      "num_tokens": 96427180.0,
      "step": 47590
    },
    {
      "entropy": 5.753362846374512,
      "epoch": 4.159311369396137,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003372233954481023,
      "loss": 5.2446,
      "mean_token_accuracy": 0.20733765959739686,
      "num_tokens": 96437220.0,
      "step": 47595
    },
    {
      "entropy": 5.711532974243164,
      "epoch": 4.159748317748842,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033719344903678183,
      "loss": 5.2025,
      "mean_token_accuracy": 0.2113091841340065,
      "num_tokens": 96447569.0,
      "step": 47600
    },
    {
      "entropy": 5.690315580368042,
      "epoch": 4.160185266101546,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003371635014325585,
      "loss": 5.2153,
      "mean_token_accuracy": 0.2063892215490341,
      "num_tokens": 96457128.0,
      "step": 47605
    },
    {
      "entropy": 5.714135456085205,
      "epoch": 4.160622214454252,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033713355263600655,
      "loss": 5.2095,
      "mean_token_accuracy": 0.20705727785825728,
      "num_tokens": 96466678.0,
      "step": 47610
    },
    {
      "entropy": 5.765192604064941,
      "epoch": 4.161059162806956,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033710360264770046,
      "loss": 5.2989,
      "mean_token_accuracy": 0.20039586424827577,
      "num_tokens": 96477506.0,
      "step": 47615
    },
    {
      "entropy": 5.779310798645019,
      "epoch": 4.161496111159661,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003370736514682148,
      "loss": 5.2749,
      "mean_token_accuracy": 0.2004241332411766,
      "num_tokens": 96487811.0,
      "step": 47620
    },
    {
      "entropy": 5.7924436092376705,
      "epoch": 4.161933059512366,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033704369909812393,
      "loss": 5.2175,
      "mean_token_accuracy": 0.20635579228401185,
      "num_tokens": 96497363.0,
      "step": 47625
    },
    {
      "entropy": 5.715611457824707,
      "epoch": 4.1623700078650705,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003370137455380024,
      "loss": 5.1994,
      "mean_token_accuracy": 0.21291467398405076,
      "num_tokens": 96507390.0,
      "step": 47630
    },
    {
      "entropy": 5.882424449920654,
      "epoch": 4.162806956217775,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033698379078842477,
      "loss": 5.3845,
      "mean_token_accuracy": 0.1905989244580269,
      "num_tokens": 96518705.0,
      "step": 47635
    },
    {
      "entropy": 5.784201383590698,
      "epoch": 4.163243904570479,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033695383484996555,
      "loss": 5.2502,
      "mean_token_accuracy": 0.20738596618175506,
      "num_tokens": 96529296.0,
      "step": 47640
    },
    {
      "entropy": 5.737231683731079,
      "epoch": 4.163680852923185,
      "grad_norm": 1.125,
      "learning_rate": 0.00033692387772319934,
      "loss": 5.1796,
      "mean_token_accuracy": 0.2116026610136032,
      "num_tokens": 96538959.0,
      "step": 47645
    },
    {
      "entropy": 5.76048321723938,
      "epoch": 4.164117801275889,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003368939194087006,
      "loss": 5.2602,
      "mean_token_accuracy": 0.20977508574724196,
      "num_tokens": 96548893.0,
      "step": 47650
    },
    {
      "entropy": 5.757289838790894,
      "epoch": 4.164554749628594,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003368639599070442,
      "loss": 5.1998,
      "mean_token_accuracy": 0.20840396136045455,
      "num_tokens": 96557390.0,
      "step": 47655
    },
    {
      "entropy": 5.767095470428467,
      "epoch": 4.164991697981298,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003368339992188045,
      "loss": 5.251,
      "mean_token_accuracy": 0.1993142619729042,
      "num_tokens": 96568010.0,
      "step": 47660
    },
    {
      "entropy": 5.750211906433106,
      "epoch": 4.165428646334004,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003368040373445564,
      "loss": 5.2439,
      "mean_token_accuracy": 0.2060219019651413,
      "num_tokens": 96578636.0,
      "step": 47665
    },
    {
      "entropy": 5.764035987854004,
      "epoch": 4.165865594686708,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033677407428487445,
      "loss": 5.3146,
      "mean_token_accuracy": 0.1941012054681778,
      "num_tokens": 96589172.0,
      "step": 47670
    },
    {
      "entropy": 5.77745361328125,
      "epoch": 4.1663025430394125,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033674411004033333,
      "loss": 5.3181,
      "mean_token_accuracy": 0.2003816694021225,
      "num_tokens": 96599473.0,
      "step": 47675
    },
    {
      "entropy": 5.785663986206055,
      "epoch": 4.166739491392118,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003367141446115079,
      "loss": 5.3136,
      "mean_token_accuracy": 0.19999590814113616,
      "num_tokens": 96609215.0,
      "step": 47680
    },
    {
      "entropy": 5.728609418869018,
      "epoch": 4.167176439744822,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003366841779989728,
      "loss": 5.2072,
      "mean_token_accuracy": 0.20912400931119918,
      "num_tokens": 96619742.0,
      "step": 47685
    },
    {
      "entropy": 5.803304815292359,
      "epoch": 4.167613388097527,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003366542102033028,
      "loss": 5.3113,
      "mean_token_accuracy": 0.19998765140771865,
      "num_tokens": 96629899.0,
      "step": 47690
    },
    {
      "entropy": 5.781625461578369,
      "epoch": 4.168050336450231,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003366242412250728,
      "loss": 5.2102,
      "mean_token_accuracy": 0.20836726129055022,
      "num_tokens": 96640308.0,
      "step": 47695
    },
    {
      "entropy": 5.7924253940582275,
      "epoch": 4.168487284802937,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003365942710648576,
      "loss": 5.2671,
      "mean_token_accuracy": 0.202874593436718,
      "num_tokens": 96649929.0,
      "step": 47700
    },
    {
      "entropy": 5.7008765697479244,
      "epoch": 4.168924233155641,
      "grad_norm": 1.03125,
      "learning_rate": 0.000336564299723232,
      "loss": 5.1569,
      "mean_token_accuracy": 0.21320269256830215,
      "num_tokens": 96659794.0,
      "step": 47705
    },
    {
      "entropy": 5.812801837921143,
      "epoch": 4.169361181508346,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033653432720077075,
      "loss": 5.3417,
      "mean_token_accuracy": 0.20238335579633712,
      "num_tokens": 96670452.0,
      "step": 47710
    },
    {
      "entropy": 5.741295003890992,
      "epoch": 4.16979812986105,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000336504353498049,
      "loss": 5.183,
      "mean_token_accuracy": 0.21265308260917665,
      "num_tokens": 96679926.0,
      "step": 47715
    },
    {
      "entropy": 5.76411805152893,
      "epoch": 4.170235078213755,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00033647437861564146,
      "loss": 5.2835,
      "mean_token_accuracy": 0.20500184148550032,
      "num_tokens": 96690089.0,
      "step": 47720
    },
    {
      "entropy": 5.828068733215332,
      "epoch": 4.17067202656646,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003364444025541231,
      "loss": 5.3698,
      "mean_token_accuracy": 0.1982852429151535,
      "num_tokens": 96700991.0,
      "step": 47725
    },
    {
      "entropy": 5.815298271179199,
      "epoch": 4.171108974919164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033641442531406895,
      "loss": 5.325,
      "mean_token_accuracy": 0.19856300055980683,
      "num_tokens": 96711077.0,
      "step": 47730
    },
    {
      "entropy": 5.792220258712769,
      "epoch": 4.17154592327187,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033638444689605386,
      "loss": 5.2229,
      "mean_token_accuracy": 0.20257839560508728,
      "num_tokens": 96721590.0,
      "step": 47735
    },
    {
      "entropy": 5.809258031845093,
      "epoch": 4.171982871624574,
      "grad_norm": 1.0625,
      "learning_rate": 0.000336354467300653,
      "loss": 5.2465,
      "mean_token_accuracy": 0.21273929327726365,
      "num_tokens": 96732385.0,
      "step": 47740
    },
    {
      "entropy": 5.728327989578247,
      "epoch": 4.172419819977279,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033632448652844133,
      "loss": 5.2872,
      "mean_token_accuracy": 0.20843659490346908,
      "num_tokens": 96742305.0,
      "step": 47745
    },
    {
      "entropy": 5.828824138641357,
      "epoch": 4.172856768329983,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033629450457999384,
      "loss": 5.3985,
      "mean_token_accuracy": 0.19282852709293366,
      "num_tokens": 96753573.0,
      "step": 47750
    },
    {
      "entropy": 5.772874546051026,
      "epoch": 4.1732937166826884,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003362645214558856,
      "loss": 5.1071,
      "mean_token_accuracy": 0.215509769320488,
      "num_tokens": 96763647.0,
      "step": 47755
    },
    {
      "entropy": 5.743780565261841,
      "epoch": 4.173730665035393,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003362345371566918,
      "loss": 5.2721,
      "mean_token_accuracy": 0.20263365060091018,
      "num_tokens": 96772966.0,
      "step": 47760
    },
    {
      "entropy": 5.801036643981933,
      "epoch": 4.174167613388097,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003362045516829874,
      "loss": 5.2388,
      "mean_token_accuracy": 0.20609713196754456,
      "num_tokens": 96783465.0,
      "step": 47765
    },
    {
      "entropy": 5.771075439453125,
      "epoch": 4.174604561740802,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00033617456503534773,
      "loss": 5.2118,
      "mean_token_accuracy": 0.21108569651842118,
      "num_tokens": 96793460.0,
      "step": 47770
    },
    {
      "entropy": 5.787150526046753,
      "epoch": 4.175041510093507,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033614457721434774,
      "loss": 5.2966,
      "mean_token_accuracy": 0.20030826032161714,
      "num_tokens": 96803699.0,
      "step": 47775
    },
    {
      "entropy": 5.832696723937988,
      "epoch": 4.175478458446212,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003361145882205628,
      "loss": 5.2969,
      "mean_token_accuracy": 0.20224573612213134,
      "num_tokens": 96814605.0,
      "step": 47780
    },
    {
      "entropy": 5.780944585800171,
      "epoch": 4.175915406798916,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033608459805456806,
      "loss": 5.2438,
      "mean_token_accuracy": 0.20334162414073945,
      "num_tokens": 96824737.0,
      "step": 47785
    },
    {
      "entropy": 5.7824419975280765,
      "epoch": 4.1763523551516215,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003360546067169386,
      "loss": 5.2546,
      "mean_token_accuracy": 0.2073807418346405,
      "num_tokens": 96834771.0,
      "step": 47790
    },
    {
      "entropy": 5.782817411422729,
      "epoch": 4.176789303504326,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033602461420824994,
      "loss": 5.2826,
      "mean_token_accuracy": 0.19987950325012208,
      "num_tokens": 96844106.0,
      "step": 47795
    },
    {
      "entropy": 5.764507722854614,
      "epoch": 4.1772262518570304,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033599462052907707,
      "loss": 5.1718,
      "mean_token_accuracy": 0.20969458371400834,
      "num_tokens": 96854296.0,
      "step": 47800
    },
    {
      "entropy": 5.717115449905395,
      "epoch": 4.177663200209735,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033596462567999546,
      "loss": 5.1857,
      "mean_token_accuracy": 0.2130592182278633,
      "num_tokens": 96864351.0,
      "step": 47805
    },
    {
      "entropy": 5.771771430969238,
      "epoch": 4.17810014856244,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003359346296615804,
      "loss": 5.2845,
      "mean_token_accuracy": 0.2063409775495529,
      "num_tokens": 96874789.0,
      "step": 47810
    },
    {
      "entropy": 5.723705673217774,
      "epoch": 4.178537096915145,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003359046324744072,
      "loss": 5.1755,
      "mean_token_accuracy": 0.2129555970430374,
      "num_tokens": 96885141.0,
      "step": 47815
    },
    {
      "entropy": 5.797661209106446,
      "epoch": 4.178974045267849,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033587463411905127,
      "loss": 5.2089,
      "mean_token_accuracy": 0.21262694895267487,
      "num_tokens": 96894556.0,
      "step": 47820
    },
    {
      "entropy": 5.704473066329956,
      "epoch": 4.179410993620554,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000335844634596088,
      "loss": 5.1145,
      "mean_token_accuracy": 0.22022313475608826,
      "num_tokens": 96904355.0,
      "step": 47825
    },
    {
      "entropy": 5.714384460449219,
      "epoch": 4.179847941973259,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033581463390609264,
      "loss": 5.1875,
      "mean_token_accuracy": 0.2148823231458664,
      "num_tokens": 96913866.0,
      "step": 47830
    },
    {
      "entropy": 5.7776295185089115,
      "epoch": 4.1802848903259635,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003357846320496408,
      "loss": 5.2441,
      "mean_token_accuracy": 0.20753653198480607,
      "num_tokens": 96924142.0,
      "step": 47835
    },
    {
      "entropy": 5.774512767791748,
      "epoch": 4.180721838678668,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033575462902730785,
      "loss": 5.1768,
      "mean_token_accuracy": 0.21260992139577867,
      "num_tokens": 96934197.0,
      "step": 47840
    },
    {
      "entropy": 5.731493425369263,
      "epoch": 4.181158787031373,
      "grad_norm": 1.125,
      "learning_rate": 0.00033572462483966934,
      "loss": 5.2301,
      "mean_token_accuracy": 0.21062200367450715,
      "num_tokens": 96942932.0,
      "step": 47845
    },
    {
      "entropy": 5.749538707733154,
      "epoch": 4.181595735384078,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033569461948730064,
      "loss": 5.1898,
      "mean_token_accuracy": 0.21129475086927413,
      "num_tokens": 96951859.0,
      "step": 47850
    },
    {
      "entropy": 5.748414611816406,
      "epoch": 4.182032683736782,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033566461297077733,
      "loss": 5.2349,
      "mean_token_accuracy": 0.20720428377389907,
      "num_tokens": 96962206.0,
      "step": 47855
    },
    {
      "entropy": 5.724806308746338,
      "epoch": 4.182469632089487,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033563460529067497,
      "loss": 5.2635,
      "mean_token_accuracy": 0.21174207478761672,
      "num_tokens": 96971302.0,
      "step": 47860
    },
    {
      "entropy": 5.629686450958252,
      "epoch": 4.182906580442192,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0003356045964475691,
      "loss": 5.1624,
      "mean_token_accuracy": 0.21573818773031234,
      "num_tokens": 96981446.0,
      "step": 47865
    },
    {
      "entropy": 5.792446327209473,
      "epoch": 4.183343528794897,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003355745864420354,
      "loss": 5.2912,
      "mean_token_accuracy": 0.2013075664639473,
      "num_tokens": 96990383.0,
      "step": 47870
    },
    {
      "entropy": 5.725125885009765,
      "epoch": 4.183780477147601,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033554457527464927,
      "loss": 5.2372,
      "mean_token_accuracy": 0.21086713075637817,
      "num_tokens": 97001497.0,
      "step": 47875
    },
    {
      "entropy": 5.787036752700805,
      "epoch": 4.1842174255003055,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003355145629459865,
      "loss": 5.269,
      "mean_token_accuracy": 0.21083782762289047,
      "num_tokens": 97011824.0,
      "step": 47880
    },
    {
      "entropy": 5.775169754028321,
      "epoch": 4.184654373853011,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003354845494566227,
      "loss": 5.2173,
      "mean_token_accuracy": 0.205662602186203,
      "num_tokens": 97022882.0,
      "step": 47885
    },
    {
      "entropy": 5.681792449951172,
      "epoch": 4.185091322205715,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003354545348071337,
      "loss": 5.1329,
      "mean_token_accuracy": 0.218423755466938,
      "num_tokens": 97032576.0,
      "step": 47890
    },
    {
      "entropy": 5.699679899215698,
      "epoch": 4.18552827055842,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033542451899809485,
      "loss": 5.2327,
      "mean_token_accuracy": 0.2073699042201042,
      "num_tokens": 97042602.0,
      "step": 47895
    },
    {
      "entropy": 5.757630681991577,
      "epoch": 4.185965218911125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003353945020300821,
      "loss": 5.2041,
      "mean_token_accuracy": 0.21495998203754424,
      "num_tokens": 97052669.0,
      "step": 47900
    },
    {
      "entropy": 5.7316309928894045,
      "epoch": 4.18640216726383,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033536448390367114,
      "loss": 5.158,
      "mean_token_accuracy": 0.21349920928478242,
      "num_tokens": 97062068.0,
      "step": 47905
    },
    {
      "entropy": 5.770053148269653,
      "epoch": 4.186839115616534,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033533446461943777,
      "loss": 5.2036,
      "mean_token_accuracy": 0.21104210913181304,
      "num_tokens": 97071214.0,
      "step": 47910
    },
    {
      "entropy": 5.710597372055053,
      "epoch": 4.187276063969239,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003353044441779578,
      "loss": 5.2582,
      "mean_token_accuracy": 0.20221720784902572,
      "num_tokens": 97080974.0,
      "step": 47915
    },
    {
      "entropy": 5.8314203262329105,
      "epoch": 4.187713012321944,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033527442257980696,
      "loss": 5.3266,
      "mean_token_accuracy": 0.1958453595638275,
      "num_tokens": 97092507.0,
      "step": 47920
    },
    {
      "entropy": 5.816764497756958,
      "epoch": 4.188149960674648,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033524439982556106,
      "loss": 5.2933,
      "mean_token_accuracy": 0.19607650935649873,
      "num_tokens": 97101342.0,
      "step": 47925
    },
    {
      "entropy": 5.776048946380615,
      "epoch": 4.188586909027353,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000335214375915796,
      "loss": 5.2051,
      "mean_token_accuracy": 0.20671217441558837,
      "num_tokens": 97112174.0,
      "step": 47930
    },
    {
      "entropy": 5.734047985076904,
      "epoch": 4.189023857380057,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003351843508510878,
      "loss": 5.2345,
      "mean_token_accuracy": 0.20748777240514754,
      "num_tokens": 97122557.0,
      "step": 47935
    },
    {
      "entropy": 5.766507720947265,
      "epoch": 4.189460805732763,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033515432463201213,
      "loss": 5.2186,
      "mean_token_accuracy": 0.21049606055021286,
      "num_tokens": 97132565.0,
      "step": 47940
    },
    {
      "entropy": 5.771609735488892,
      "epoch": 4.189897754085467,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000335124297259145,
      "loss": 5.2017,
      "mean_token_accuracy": 0.2078954041004181,
      "num_tokens": 97141367.0,
      "step": 47945
    },
    {
      "entropy": 5.782004976272583,
      "epoch": 4.190334702438172,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003350942687330623,
      "loss": 5.2412,
      "mean_token_accuracy": 0.2019062250852585,
      "num_tokens": 97151859.0,
      "step": 47950
    },
    {
      "entropy": 5.8050895690917965,
      "epoch": 4.190771650790876,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033506423905434,
      "loss": 5.3007,
      "mean_token_accuracy": 0.198605777323246,
      "num_tokens": 97161670.0,
      "step": 47955
    },
    {
      "entropy": 5.719483852386475,
      "epoch": 4.1912085991435815,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003350342082235543,
      "loss": 5.1434,
      "mean_token_accuracy": 0.21744581758975984,
      "num_tokens": 97171089.0,
      "step": 47960
    },
    {
      "entropy": 5.759927415847779,
      "epoch": 4.191645547496286,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033500417624128087,
      "loss": 5.3135,
      "mean_token_accuracy": 0.20023958384990692,
      "num_tokens": 97181934.0,
      "step": 47965
    },
    {
      "entropy": 5.7572578430175785,
      "epoch": 4.19208249584899,
      "grad_norm": 1.140625,
      "learning_rate": 0.000334974143108096,
      "loss": 5.2125,
      "mean_token_accuracy": 0.21104494333267212,
      "num_tokens": 97191150.0,
      "step": 47970
    },
    {
      "entropy": 5.803983974456787,
      "epoch": 4.192519444201696,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003349441088245756,
      "loss": 5.2855,
      "mean_token_accuracy": 0.20499981045722962,
      "num_tokens": 97202015.0,
      "step": 47975
    },
    {
      "entropy": 5.74813141822815,
      "epoch": 4.1929563925544,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003349140733912957,
      "loss": 5.2417,
      "mean_token_accuracy": 0.20687844902276992,
      "num_tokens": 97211418.0,
      "step": 47980
    },
    {
      "entropy": 5.811210966110229,
      "epoch": 4.193393340907105,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003348840368088326,
      "loss": 5.358,
      "mean_token_accuracy": 0.1945490673184395,
      "num_tokens": 97222557.0,
      "step": 47985
    },
    {
      "entropy": 5.813355541229248,
      "epoch": 4.193830289259809,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003348539990777623,
      "loss": 5.3656,
      "mean_token_accuracy": 0.19975760132074355,
      "num_tokens": 97231966.0,
      "step": 47990
    },
    {
      "entropy": 5.794587326049805,
      "epoch": 4.1942672376125145,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003348239601986608,
      "loss": 5.2033,
      "mean_token_accuracy": 0.20543840676546096,
      "num_tokens": 97241843.0,
      "step": 47995
    },
    {
      "entropy": 5.679954147338867,
      "epoch": 4.194704185965219,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003347939201721045,
      "loss": 5.1918,
      "mean_token_accuracy": 0.21124981045722963,
      "num_tokens": 97251987.0,
      "step": 48000
    },
    {
      "epoch": 4.194704185965219,
      "eval_entropy": 5.45683185853787,
      "eval_loss": 5.329683303833008,
      "eval_mean_token_accuracy": 0.21050717517723913,
      "eval_num_tokens": 97251987.0,
      "eval_runtime": 28.2413,
      "eval_samples_per_second": 1310.636,
      "eval_steps_per_second": 163.838,
      "step": 48000
    },
    {
      "entropy": 5.717936706542969,
      "epoch": 4.1951411343179235,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033476387899866944,
      "loss": 5.2492,
      "mean_token_accuracy": 0.20513129830360413,
      "num_tokens": 97262489.0,
      "step": 48005
    },
    {
      "entropy": 5.862109565734864,
      "epoch": 4.195578082670628,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033473383667893185,
      "loss": 5.2601,
      "mean_token_accuracy": 0.20731114000082015,
      "num_tokens": 97272380.0,
      "step": 48010
    },
    {
      "entropy": 5.708059167861938,
      "epoch": 4.196015031023333,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033470379321346805,
      "loss": 5.1942,
      "mean_token_accuracy": 0.21176115572452545,
      "num_tokens": 97282313.0,
      "step": 48015
    },
    {
      "entropy": 5.825455856323242,
      "epoch": 4.196451979376038,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033467374860285414,
      "loss": 5.2841,
      "mean_token_accuracy": 0.20089250355958937,
      "num_tokens": 97292720.0,
      "step": 48020
    },
    {
      "entropy": 5.753923130035401,
      "epoch": 4.196888927728742,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033464370284766654,
      "loss": 5.2081,
      "mean_token_accuracy": 0.20995998233556748,
      "num_tokens": 97301988.0,
      "step": 48025
    },
    {
      "entropy": 5.6828360080719,
      "epoch": 4.197325876081448,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033461365594848134,
      "loss": 5.2497,
      "mean_token_accuracy": 0.20777015537023544,
      "num_tokens": 97312594.0,
      "step": 48030
    },
    {
      "entropy": 5.793582344055176,
      "epoch": 4.197762824434152,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003345836079058751,
      "loss": 5.2932,
      "mean_token_accuracy": 0.2008165568113327,
      "num_tokens": 97321727.0,
      "step": 48035
    },
    {
      "entropy": 5.6716561794281,
      "epoch": 4.1981997727868565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000334553558720424,
      "loss": 5.09,
      "mean_token_accuracy": 0.22189500778913498,
      "num_tokens": 97331180.0,
      "step": 48040
    },
    {
      "entropy": 5.629448270797729,
      "epoch": 4.198636721139561,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003345235083927044,
      "loss": 5.0777,
      "mean_token_accuracy": 0.22708400189876557,
      "num_tokens": 97340432.0,
      "step": 48045
    },
    {
      "entropy": 5.667933320999145,
      "epoch": 4.199073669492266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033449345692329276,
      "loss": 5.1499,
      "mean_token_accuracy": 0.20534356385469438,
      "num_tokens": 97350493.0,
      "step": 48050
    },
    {
      "entropy": 5.726819276809692,
      "epoch": 4.199510617844971,
      "grad_norm": 1.125,
      "learning_rate": 0.0003344634043127655,
      "loss": 5.2235,
      "mean_token_accuracy": 0.21835844218730927,
      "num_tokens": 97360484.0,
      "step": 48055
    },
    {
      "entropy": 5.754037284851075,
      "epoch": 4.199947566197675,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003344333505616989,
      "loss": 5.2105,
      "mean_token_accuracy": 0.21274515390396118,
      "num_tokens": 97371575.0,
      "step": 48060
    },
    {
      "entropy": 5.668487501144409,
      "epoch": 4.20038451455038,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003344032956706695,
      "loss": 5.1202,
      "mean_token_accuracy": 0.21636077016592026,
      "num_tokens": 97381806.0,
      "step": 48065
    },
    {
      "entropy": 5.758576202392578,
      "epoch": 4.200821462903085,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033437323964025377,
      "loss": 5.1827,
      "mean_token_accuracy": 0.211322557926178,
      "num_tokens": 97391679.0,
      "step": 48070
    },
    {
      "entropy": 5.784478378295899,
      "epoch": 4.20125841125579,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033434318247102816,
      "loss": 5.2914,
      "mean_token_accuracy": 0.20095385164022445,
      "num_tokens": 97401683.0,
      "step": 48075
    },
    {
      "entropy": 5.7025768756866455,
      "epoch": 4.201695359608494,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003343131241635693,
      "loss": 5.1645,
      "mean_token_accuracy": 0.21424417346715927,
      "num_tokens": 97411650.0,
      "step": 48080
    },
    {
      "entropy": 5.793331575393677,
      "epoch": 4.202132307961199,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003342830647184536,
      "loss": 5.2536,
      "mean_token_accuracy": 0.2070830225944519,
      "num_tokens": 97421246.0,
      "step": 48085
    },
    {
      "entropy": 5.7637251853942875,
      "epoch": 4.202569256313904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003342530041362576,
      "loss": 5.2851,
      "mean_token_accuracy": 0.20556629896163942,
      "num_tokens": 97431287.0,
      "step": 48090
    },
    {
      "entropy": 5.668624782562256,
      "epoch": 4.203006204666608,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033422294241755796,
      "loss": 5.1042,
      "mean_token_accuracy": 0.21665276885032653,
      "num_tokens": 97442283.0,
      "step": 48095
    },
    {
      "entropy": 5.779133892059326,
      "epoch": 4.203443153019313,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003341928795629313,
      "loss": 5.1796,
      "mean_token_accuracy": 0.21404631137847902,
      "num_tokens": 97451993.0,
      "step": 48100
    },
    {
      "entropy": 5.752556562423706,
      "epoch": 4.203880101372018,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003341628155729542,
      "loss": 5.2172,
      "mean_token_accuracy": 0.20751077085733413,
      "num_tokens": 97462441.0,
      "step": 48105
    },
    {
      "entropy": 5.740087604522705,
      "epoch": 4.204317049724723,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033413275044820326,
      "loss": 5.1758,
      "mean_token_accuracy": 0.21020736694335937,
      "num_tokens": 97472827.0,
      "step": 48110
    },
    {
      "entropy": 5.822997808456421,
      "epoch": 4.204753998077427,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033410268418925513,
      "loss": 5.3108,
      "mean_token_accuracy": 0.19905846118927,
      "num_tokens": 97483696.0,
      "step": 48115
    },
    {
      "entropy": 5.644427728652954,
      "epoch": 4.205190946430132,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003340726167966867,
      "loss": 5.161,
      "mean_token_accuracy": 0.2170985922217369,
      "num_tokens": 97493846.0,
      "step": 48120
    },
    {
      "entropy": 5.729534339904785,
      "epoch": 4.205627894782837,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003340425482710743,
      "loss": 5.212,
      "mean_token_accuracy": 0.21128261387348174,
      "num_tokens": 97504216.0,
      "step": 48125
    },
    {
      "entropy": 5.80805230140686,
      "epoch": 4.206064843135541,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003340124786129951,
      "loss": 5.3053,
      "mean_token_accuracy": 0.19835573434829712,
      "num_tokens": 97515464.0,
      "step": 48130
    },
    {
      "entropy": 5.764473867416382,
      "epoch": 4.206501791488246,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033398240782302554,
      "loss": 5.2915,
      "mean_token_accuracy": 0.2025603398680687,
      "num_tokens": 97525702.0,
      "step": 48135
    },
    {
      "entropy": 5.749943208694458,
      "epoch": 4.20693873984095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003339523359017425,
      "loss": 5.2715,
      "mean_token_accuracy": 0.21122009307146072,
      "num_tokens": 97536712.0,
      "step": 48140
    },
    {
      "entropy": 5.809886407852173,
      "epoch": 4.207375688193656,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003339222628497228,
      "loss": 5.2431,
      "mean_token_accuracy": 0.2004813715815544,
      "num_tokens": 97544976.0,
      "step": 48145
    },
    {
      "entropy": 5.767768096923828,
      "epoch": 4.20781263654636,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003338921886675432,
      "loss": 5.3107,
      "mean_token_accuracy": 0.19681439250707627,
      "num_tokens": 97555140.0,
      "step": 48150
    },
    {
      "entropy": 5.738437747955322,
      "epoch": 4.208249584899065,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033386211335578053,
      "loss": 5.1485,
      "mean_token_accuracy": 0.21025555282831193,
      "num_tokens": 97564578.0,
      "step": 48155
    },
    {
      "entropy": 5.780811023712158,
      "epoch": 4.20868653325177,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003338320369150117,
      "loss": 5.2395,
      "mean_token_accuracy": 0.2068064585328102,
      "num_tokens": 97574585.0,
      "step": 48160
    },
    {
      "entropy": 5.709344148635864,
      "epoch": 4.2091234816044745,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033380195934581356,
      "loss": 5.1936,
      "mean_token_accuracy": 0.21558786779642106,
      "num_tokens": 97586133.0,
      "step": 48165
    },
    {
      "entropy": 5.760355043411255,
      "epoch": 4.209560429957179,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003337718806487631,
      "loss": 5.205,
      "mean_token_accuracy": 0.20810637772083282,
      "num_tokens": 97595940.0,
      "step": 48170
    },
    {
      "entropy": 5.664193058013916,
      "epoch": 4.209997378309883,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003337418008244371,
      "loss": 5.1578,
      "mean_token_accuracy": 0.2180177539587021,
      "num_tokens": 97605656.0,
      "step": 48175
    },
    {
      "entropy": 5.789137506484986,
      "epoch": 4.210434326662589,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033371171987341266,
      "loss": 5.2907,
      "mean_token_accuracy": 0.19788089245557786,
      "num_tokens": 97615256.0,
      "step": 48180
    },
    {
      "entropy": 5.758964920043946,
      "epoch": 4.210871275015293,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033368163779626654,
      "loss": 5.2556,
      "mean_token_accuracy": 0.2110874906182289,
      "num_tokens": 97624820.0,
      "step": 48185
    },
    {
      "entropy": 5.82016429901123,
      "epoch": 4.211308223367998,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00033365155459357595,
      "loss": 5.2651,
      "mean_token_accuracy": 0.20776187032461166,
      "num_tokens": 97636221.0,
      "step": 48190
    },
    {
      "entropy": 5.726201152801513,
      "epoch": 4.211745171720702,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003336214702659179,
      "loss": 5.2529,
      "mean_token_accuracy": 0.2049372300505638,
      "num_tokens": 97646460.0,
      "step": 48195
    },
    {
      "entropy": 5.711814117431641,
      "epoch": 4.2121821200734075,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033359138481386914,
      "loss": 5.2038,
      "mean_token_accuracy": 0.20604861825704573,
      "num_tokens": 97655762.0,
      "step": 48200
    },
    {
      "entropy": 5.725775814056396,
      "epoch": 4.212619068426112,
      "grad_norm": 1.109375,
      "learning_rate": 0.000333561298238007,
      "loss": 5.1753,
      "mean_token_accuracy": 0.2179071217775345,
      "num_tokens": 97665916.0,
      "step": 48205
    },
    {
      "entropy": 5.815834426879883,
      "epoch": 4.2130560167788165,
      "grad_norm": 1.125,
      "learning_rate": 0.00033353121053890844,
      "loss": 5.2806,
      "mean_token_accuracy": 0.20321095138788223,
      "num_tokens": 97675529.0,
      "step": 48210
    },
    {
      "entropy": 5.716463327407837,
      "epoch": 4.213492965131522,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033350112171715067,
      "loss": 5.1974,
      "mean_token_accuracy": 0.20980744957923889,
      "num_tokens": 97684860.0,
      "step": 48215
    },
    {
      "entropy": 5.719722604751587,
      "epoch": 4.213929913484226,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033347103177331066,
      "loss": 5.2192,
      "mean_token_accuracy": 0.20817384570837022,
      "num_tokens": 97695271.0,
      "step": 48220
    },
    {
      "entropy": 5.779172325134278,
      "epoch": 4.214366861836931,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003334409407079657,
      "loss": 5.191,
      "mean_token_accuracy": 0.20805077999830246,
      "num_tokens": 97706346.0,
      "step": 48225
    },
    {
      "entropy": 5.741833257675171,
      "epoch": 4.214803810189635,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003334108485216928,
      "loss": 5.1953,
      "mean_token_accuracy": 0.20676409155130387,
      "num_tokens": 97716667.0,
      "step": 48230
    },
    {
      "entropy": 5.736965131759644,
      "epoch": 4.215240758542341,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003333807552150693,
      "loss": 5.2671,
      "mean_token_accuracy": 0.2001111924648285,
      "num_tokens": 97725868.0,
      "step": 48235
    },
    {
      "entropy": 5.769966983795166,
      "epoch": 4.215677706895045,
      "grad_norm": 1.125,
      "learning_rate": 0.0003333506607886722,
      "loss": 5.1974,
      "mean_token_accuracy": 0.2095757693052292,
      "num_tokens": 97735308.0,
      "step": 48240
    },
    {
      "entropy": 5.799214267730713,
      "epoch": 4.2161146552477495,
      "grad_norm": 1.109375,
      "learning_rate": 0.000333320565243079,
      "loss": 5.2059,
      "mean_token_accuracy": 0.2107425943017006,
      "num_tokens": 97745345.0,
      "step": 48245
    },
    {
      "entropy": 5.731277942657471,
      "epoch": 4.216551603600454,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003332904685788667,
      "loss": 5.2451,
      "mean_token_accuracy": 0.207790307700634,
      "num_tokens": 97754858.0,
      "step": 48250
    },
    {
      "entropy": 5.691017055511475,
      "epoch": 4.216988551953159,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003332603707966127,
      "loss": 5.207,
      "mean_token_accuracy": 0.20778943598270416,
      "num_tokens": 97764681.0,
      "step": 48255
    },
    {
      "entropy": 5.758703184127808,
      "epoch": 4.217425500305864,
      "grad_norm": 1.125,
      "learning_rate": 0.0003332302718968942,
      "loss": 5.2255,
      "mean_token_accuracy": 0.20798033475875854,
      "num_tokens": 97775051.0,
      "step": 48260
    },
    {
      "entropy": 5.748432636260986,
      "epoch": 4.217862448658568,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003332001718802886,
      "loss": 5.2097,
      "mean_token_accuracy": 0.2143061488866806,
      "num_tokens": 97787263.0,
      "step": 48265
    },
    {
      "entropy": 5.85691237449646,
      "epoch": 4.218299397011274,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003331700707473733,
      "loss": 5.2563,
      "mean_token_accuracy": 0.20661619454622268,
      "num_tokens": 97797507.0,
      "step": 48270
    },
    {
      "entropy": 5.810731410980225,
      "epoch": 4.218736345363978,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033313996849872543,
      "loss": 5.2258,
      "mean_token_accuracy": 0.21119388788938523,
      "num_tokens": 97807304.0,
      "step": 48275
    },
    {
      "entropy": 5.651919269561768,
      "epoch": 4.219173293716683,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003331098651349225,
      "loss": 5.1677,
      "mean_token_accuracy": 0.21658274233341218,
      "num_tokens": 97817552.0,
      "step": 48280
    },
    {
      "entropy": 5.703875160217285,
      "epoch": 4.219610242069387,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000333079760656542,
      "loss": 5.1945,
      "mean_token_accuracy": 0.20968976765871047,
      "num_tokens": 97827635.0,
      "step": 48285
    },
    {
      "entropy": 5.804172658920288,
      "epoch": 4.220047190422092,
      "grad_norm": 1.125,
      "learning_rate": 0.00033304965506416126,
      "loss": 5.2867,
      "mean_token_accuracy": 0.20290416181087495,
      "num_tokens": 97836363.0,
      "step": 48290
    },
    {
      "entropy": 5.71501579284668,
      "epoch": 4.220484138774797,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003330195483583577,
      "loss": 5.1008,
      "mean_token_accuracy": 0.21916084885597228,
      "num_tokens": 97846877.0,
      "step": 48295
    },
    {
      "entropy": 5.715916442871094,
      "epoch": 4.220921087127501,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003329894405397088,
      "loss": 5.2882,
      "mean_token_accuracy": 0.20706462115049362,
      "num_tokens": 97856502.0,
      "step": 48300
    },
    {
      "entropy": 5.725703239440918,
      "epoch": 4.221358035480206,
      "grad_norm": 1.078125,
      "learning_rate": 0.000332959331608792,
      "loss": 5.2666,
      "mean_token_accuracy": 0.20209394842386247,
      "num_tokens": 97866213.0,
      "step": 48305
    },
    {
      "entropy": 5.848153829574585,
      "epoch": 4.221794983832911,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00033292922156618487,
      "loss": 5.3467,
      "mean_token_accuracy": 0.199546779692173,
      "num_tokens": 97877074.0,
      "step": 48310
    },
    {
      "entropy": 5.730813789367676,
      "epoch": 4.222231932185616,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000332899110412465,
      "loss": 5.1233,
      "mean_token_accuracy": 0.22141058444976808,
      "num_tokens": 97887090.0,
      "step": 48315
    },
    {
      "entropy": 5.750249767303467,
      "epoch": 4.22266888053832,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003328689981482097,
      "loss": 5.2719,
      "mean_token_accuracy": 0.19896050691604614,
      "num_tokens": 97897390.0,
      "step": 48320
    },
    {
      "entropy": 5.732787656784057,
      "epoch": 4.2231058288910255,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033283888477399676,
      "loss": 5.1057,
      "mean_token_accuracy": 0.21686035692691802,
      "num_tokens": 97907126.0,
      "step": 48325
    },
    {
      "entropy": 5.752351379394531,
      "epoch": 4.22354277724373,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033280877029040376,
      "loss": 5.245,
      "mean_token_accuracy": 0.2144569545984268,
      "num_tokens": 97916680.0,
      "step": 48330
    },
    {
      "entropy": 5.7666661739349365,
      "epoch": 4.223979725596434,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003327786546980082,
      "loss": 5.233,
      "mean_token_accuracy": 0.20817360877990723,
      "num_tokens": 97926833.0,
      "step": 48335
    },
    {
      "entropy": 5.716977453231811,
      "epoch": 4.224416673949139,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033274853799738775,
      "loss": 5.1892,
      "mean_token_accuracy": 0.20755800008773803,
      "num_tokens": 97936040.0,
      "step": 48340
    },
    {
      "entropy": 5.648671722412109,
      "epoch": 4.224853622301844,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003327184201891201,
      "loss": 5.1735,
      "mean_token_accuracy": 0.21111499518156052,
      "num_tokens": 97946799.0,
      "step": 48345
    },
    {
      "entropy": 5.723124361038208,
      "epoch": 4.225290570654549,
      "grad_norm": 1.0625,
      "learning_rate": 0.000332688301273783,
      "loss": 5.1894,
      "mean_token_accuracy": 0.21398983597755433,
      "num_tokens": 97956070.0,
      "step": 48350
    },
    {
      "entropy": 5.77937684059143,
      "epoch": 4.225727519007253,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033265818125195393,
      "loss": 5.3509,
      "mean_token_accuracy": 0.19639052748680114,
      "num_tokens": 97966088.0,
      "step": 48355
    },
    {
      "entropy": 5.813472557067871,
      "epoch": 4.226164467359958,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033262806012421075,
      "loss": 5.28,
      "mean_token_accuracy": 0.206031833589077,
      "num_tokens": 97977394.0,
      "step": 48360
    },
    {
      "entropy": 5.771489095687866,
      "epoch": 4.226601415712663,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003325979378911312,
      "loss": 5.2253,
      "mean_token_accuracy": 0.20780494064092636,
      "num_tokens": 97987117.0,
      "step": 48365
    },
    {
      "entropy": 5.731772232055664,
      "epoch": 4.2270383640653675,
      "grad_norm": 1.015625,
      "learning_rate": 0.000332567814553293,
      "loss": 5.2657,
      "mean_token_accuracy": 0.20748609751462938,
      "num_tokens": 97999127.0,
      "step": 48370
    },
    {
      "entropy": 5.757293272018432,
      "epoch": 4.227475312418072,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000332537690111274,
      "loss": 5.2897,
      "mean_token_accuracy": 0.20533322393894196,
      "num_tokens": 98008260.0,
      "step": 48375
    },
    {
      "entropy": 5.729165029525757,
      "epoch": 4.227912260770777,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003325075645656519,
      "loss": 5.1797,
      "mean_token_accuracy": 0.21539859622716903,
      "num_tokens": 98017494.0,
      "step": 48380
    },
    {
      "entropy": 5.640903282165527,
      "epoch": 4.228349209123482,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033247743791700466,
      "loss": 5.1318,
      "mean_token_accuracy": 0.22574237287044524,
      "num_tokens": 98028873.0,
      "step": 48385
    },
    {
      "entropy": 5.7424506664276125,
      "epoch": 4.228786157476186,
      "grad_norm": 1.125,
      "learning_rate": 0.00033244731016591,
      "loss": 5.226,
      "mean_token_accuracy": 0.21013033241033555,
      "num_tokens": 98039178.0,
      "step": 48390
    },
    {
      "entropy": 5.724298334121704,
      "epoch": 4.229223105828891,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003324171813129457,
      "loss": 5.2494,
      "mean_token_accuracy": 0.21122945547103883,
      "num_tokens": 98048040.0,
      "step": 48395
    },
    {
      "entropy": 5.697615289688111,
      "epoch": 4.229660054181596,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033238705135868994,
      "loss": 5.1914,
      "mean_token_accuracy": 0.2147458612918854,
      "num_tokens": 98057648.0,
      "step": 48400
    },
    {
      "entropy": 5.736773300170898,
      "epoch": 4.2300970025343005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033235692030372045,
      "loss": 5.2039,
      "mean_token_accuracy": 0.20660207867622377,
      "num_tokens": 98067509.0,
      "step": 48405
    },
    {
      "entropy": 5.789730215072632,
      "epoch": 4.230533950887005,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003323267881486151,
      "loss": 5.3401,
      "mean_token_accuracy": 0.209649059176445,
      "num_tokens": 98077946.0,
      "step": 48410
    },
    {
      "entropy": 5.770322608947754,
      "epoch": 4.2309708992397095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000332296654893952,
      "loss": 5.2774,
      "mean_token_accuracy": 0.20813980102539062,
      "num_tokens": 98088046.0,
      "step": 48415
    },
    {
      "entropy": 5.678016710281372,
      "epoch": 4.231407847592415,
      "grad_norm": 1.0625,
      "learning_rate": 0.000332266520540309,
      "loss": 5.1581,
      "mean_token_accuracy": 0.21580901443958284,
      "num_tokens": 98097501.0,
      "step": 48420
    },
    {
      "entropy": 5.716540002822876,
      "epoch": 4.231844795945119,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003322363850882641,
      "loss": 5.2157,
      "mean_token_accuracy": 0.2120915323495865,
      "num_tokens": 98109193.0,
      "step": 48425
    },
    {
      "entropy": 5.654248142242432,
      "epoch": 4.232281744297824,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003322062485383953,
      "loss": 5.1041,
      "mean_token_accuracy": 0.21664490252733232,
      "num_tokens": 98118700.0,
      "step": 48430
    },
    {
      "entropy": 5.789349317550659,
      "epoch": 4.232718692650529,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033217611089128086,
      "loss": 5.2349,
      "mean_token_accuracy": 0.20485896617174149,
      "num_tokens": 98128913.0,
      "step": 48435
    },
    {
      "entropy": 5.784092044830322,
      "epoch": 4.233155641003234,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033214597214749846,
      "loss": 5.2883,
      "mean_token_accuracy": 0.1997642457485199,
      "num_tokens": 98139714.0,
      "step": 48440
    },
    {
      "entropy": 5.720229387283325,
      "epoch": 4.233592589355938,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033211583230762637,
      "loss": 5.1888,
      "mean_token_accuracy": 0.20826911330223083,
      "num_tokens": 98149289.0,
      "step": 48445
    },
    {
      "entropy": 5.6308990001678465,
      "epoch": 4.2340295377086425,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033208569137224274,
      "loss": 5.0814,
      "mean_token_accuracy": 0.22191057205200196,
      "num_tokens": 98158370.0,
      "step": 48450
    },
    {
      "entropy": 5.747625350952148,
      "epoch": 4.234466486061348,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003320555493419257,
      "loss": 5.2575,
      "mean_token_accuracy": 0.20703041553497314,
      "num_tokens": 98167798.0,
      "step": 48455
    },
    {
      "entropy": 5.755849170684814,
      "epoch": 4.234903434414052,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003320254062172533,
      "loss": 5.2639,
      "mean_token_accuracy": 0.2054793655872345,
      "num_tokens": 98178206.0,
      "step": 48460
    },
    {
      "entropy": 5.778152370452881,
      "epoch": 4.235340382766757,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033199526199880375,
      "loss": 5.2688,
      "mean_token_accuracy": 0.21161899119615554,
      "num_tokens": 98188529.0,
      "step": 48465
    },
    {
      "entropy": 5.729399824142456,
      "epoch": 4.235777331119461,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033196511668715515,
      "loss": 5.2363,
      "mean_token_accuracy": 0.21231795251369476,
      "num_tokens": 98199578.0,
      "step": 48470
    },
    {
      "entropy": 5.733621883392334,
      "epoch": 4.236214279472167,
      "grad_norm": 1.125,
      "learning_rate": 0.00033193497028288576,
      "loss": 5.1619,
      "mean_token_accuracy": 0.2103922560811043,
      "num_tokens": 98209439.0,
      "step": 48475
    },
    {
      "entropy": 5.778590297698974,
      "epoch": 4.236651227824871,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033190482278657384,
      "loss": 5.2368,
      "mean_token_accuracy": 0.21144135445356368,
      "num_tokens": 98218877.0,
      "step": 48480
    },
    {
      "entropy": 5.756720733642578,
      "epoch": 4.237088176177576,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003318746741987976,
      "loss": 5.2148,
      "mean_token_accuracy": 0.20444027185440064,
      "num_tokens": 98228719.0,
      "step": 48485
    },
    {
      "entropy": 5.697553920745849,
      "epoch": 4.237525124530281,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003318445245201353,
      "loss": 5.1461,
      "mean_token_accuracy": 0.2092994049191475,
      "num_tokens": 98238299.0,
      "step": 48490
    },
    {
      "entropy": 5.752583837509155,
      "epoch": 4.237962072882985,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033181437375116526,
      "loss": 5.2013,
      "mean_token_accuracy": 0.20533783882856368,
      "num_tokens": 98248193.0,
      "step": 48495
    },
    {
      "entropy": 5.8114735126495365,
      "epoch": 4.23839902123569,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003317842218924658,
      "loss": 5.3387,
      "mean_token_accuracy": 0.2013396814465523,
      "num_tokens": 98258495.0,
      "step": 48500
    },
    {
      "entropy": 5.786694431304932,
      "epoch": 4.238835969588394,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003317540689446151,
      "loss": 5.2408,
      "mean_token_accuracy": 0.20739866197109222,
      "num_tokens": 98268265.0,
      "step": 48505
    },
    {
      "entropy": 5.678132104873657,
      "epoch": 4.2392729179411,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033172391490819166,
      "loss": 5.187,
      "mean_token_accuracy": 0.21352438181638717,
      "num_tokens": 98278324.0,
      "step": 48510
    },
    {
      "entropy": 5.759732484817505,
      "epoch": 4.239709866293804,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003316937597837738,
      "loss": 5.2869,
      "mean_token_accuracy": 0.20281327664852142,
      "num_tokens": 98288576.0,
      "step": 48515
    },
    {
      "entropy": 5.74978494644165,
      "epoch": 4.240146814646509,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033166360357193993,
      "loss": 5.2234,
      "mean_token_accuracy": 0.21167847663164138,
      "num_tokens": 98298435.0,
      "step": 48520
    },
    {
      "entropy": 5.714175796508789,
      "epoch": 4.240583762999213,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003316334462732685,
      "loss": 5.1673,
      "mean_token_accuracy": 0.21534232497215272,
      "num_tokens": 98308099.0,
      "step": 48525
    },
    {
      "entropy": 5.720772647857666,
      "epoch": 4.2410207113519185,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003316032878883378,
      "loss": 5.1948,
      "mean_token_accuracy": 0.21269138157367706,
      "num_tokens": 98318252.0,
      "step": 48530
    },
    {
      "entropy": 5.686562395095825,
      "epoch": 4.241457659704623,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003315731284177265,
      "loss": 5.1801,
      "mean_token_accuracy": 0.21487782448530196,
      "num_tokens": 98328851.0,
      "step": 48535
    },
    {
      "entropy": 5.721674489974975,
      "epoch": 4.241894608057327,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033154296786201286,
      "loss": 5.2082,
      "mean_token_accuracy": 0.20652694553136824,
      "num_tokens": 98340099.0,
      "step": 48540
    },
    {
      "entropy": 5.765113019943238,
      "epoch": 4.242331556410033,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003315128062217755,
      "loss": 5.2591,
      "mean_token_accuracy": 0.20589060932397843,
      "num_tokens": 98350919.0,
      "step": 48545
    },
    {
      "entropy": 5.842363977432251,
      "epoch": 4.242768504762737,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003314826434975929,
      "loss": 5.2756,
      "mean_token_accuracy": 0.2023622438311577,
      "num_tokens": 98360808.0,
      "step": 48550
    },
    {
      "entropy": 5.856243467330932,
      "epoch": 4.243205453115442,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033145247969004353,
      "loss": 5.3433,
      "mean_token_accuracy": 0.20142216235399246,
      "num_tokens": 98370620.0,
      "step": 48555
    },
    {
      "entropy": 5.787868022918701,
      "epoch": 4.243642401468146,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033142231479970616,
      "loss": 5.3572,
      "mean_token_accuracy": 0.19767156094312668,
      "num_tokens": 98379964.0,
      "step": 48560
    },
    {
      "entropy": 5.763450717926025,
      "epoch": 4.2440793498208516,
      "grad_norm": 1.125,
      "learning_rate": 0.0003313921488271591,
      "loss": 5.2011,
      "mean_token_accuracy": 0.21561572551727295,
      "num_tokens": 98389744.0,
      "step": 48565
    },
    {
      "entropy": 5.713202953338623,
      "epoch": 4.244516298173556,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003313619817729812,
      "loss": 5.2408,
      "mean_token_accuracy": 0.20775982290506362,
      "num_tokens": 98399804.0,
      "step": 48570
    },
    {
      "entropy": 5.776917886734009,
      "epoch": 4.2449532465262605,
      "grad_norm": 1.125,
      "learning_rate": 0.00033133181363775085,
      "loss": 5.2762,
      "mean_token_accuracy": 0.19986832290887832,
      "num_tokens": 98410064.0,
      "step": 48575
    },
    {
      "entropy": 5.732472038269043,
      "epoch": 4.245390194878965,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003313016444220468,
      "loss": 5.1603,
      "mean_token_accuracy": 0.21154029965400695,
      "num_tokens": 98421502.0,
      "step": 48580
    },
    {
      "entropy": 5.7205116748809814,
      "epoch": 4.24582714323167,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033127147412644774,
      "loss": 5.207,
      "mean_token_accuracy": 0.2184230476617813,
      "num_tokens": 98431495.0,
      "step": 48585
    },
    {
      "entropy": 5.7590282440185545,
      "epoch": 4.246264091584375,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033124130275153236,
      "loss": 5.3031,
      "mean_token_accuracy": 0.20436637997627258,
      "num_tokens": 98442008.0,
      "step": 48590
    },
    {
      "entropy": 5.788094806671142,
      "epoch": 4.246701039937079,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033121113029787924,
      "loss": 5.2861,
      "mean_token_accuracy": 0.19884631633758545,
      "num_tokens": 98452134.0,
      "step": 48595
    },
    {
      "entropy": 5.771834945678711,
      "epoch": 4.247137988289784,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033118095676606724,
      "loss": 5.2901,
      "mean_token_accuracy": 0.2017943412065506,
      "num_tokens": 98462713.0,
      "step": 48600
    },
    {
      "entropy": 5.743972253799439,
      "epoch": 4.247574936642489,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003311507821566751,
      "loss": 5.2158,
      "mean_token_accuracy": 0.21217633336782454,
      "num_tokens": 98473017.0,
      "step": 48605
    },
    {
      "entropy": 5.732926654815674,
      "epoch": 4.2480118849951936,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033112060647028145,
      "loss": 5.1695,
      "mean_token_accuracy": 0.2155730977654457,
      "num_tokens": 98482436.0,
      "step": 48610
    },
    {
      "entropy": 5.790978240966797,
      "epoch": 4.248448833347898,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033109042970746517,
      "loss": 5.2997,
      "mean_token_accuracy": 0.19851531982421874,
      "num_tokens": 98493289.0,
      "step": 48615
    },
    {
      "entropy": 5.805899238586425,
      "epoch": 4.248885781700603,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033106025186880505,
      "loss": 5.2683,
      "mean_token_accuracy": 0.20762613117694856,
      "num_tokens": 98503847.0,
      "step": 48620
    },
    {
      "entropy": 5.777504110336304,
      "epoch": 4.249322730053308,
      "grad_norm": 1.0,
      "learning_rate": 0.00033103007295488,
      "loss": 5.2544,
      "mean_token_accuracy": 0.2069365054368973,
      "num_tokens": 98514242.0,
      "step": 48625
    },
    {
      "entropy": 5.693485736846924,
      "epoch": 4.249759678406012,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003309998929662687,
      "loss": 5.1247,
      "mean_token_accuracy": 0.21983164995908738,
      "num_tokens": 98523323.0,
      "step": 48630
    },
    {
      "entropy": 5.668731594085694,
      "epoch": 4.250196626758717,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003309697119035502,
      "loss": 5.1593,
      "mean_token_accuracy": 0.2174496665596962,
      "num_tokens": 98532870.0,
      "step": 48635
    },
    {
      "entropy": 5.768248605728149,
      "epoch": 4.250633575111422,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033093952976730325,
      "loss": 5.2571,
      "mean_token_accuracy": 0.2069289728999138,
      "num_tokens": 98542795.0,
      "step": 48640
    },
    {
      "entropy": 5.727387809753418,
      "epoch": 4.251070523464127,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003309093465581068,
      "loss": 5.1527,
      "mean_token_accuracy": 0.2173156589269638,
      "num_tokens": 98552627.0,
      "step": 48645
    },
    {
      "entropy": 5.719924545288086,
      "epoch": 4.251507471816831,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003308791622765398,
      "loss": 5.2125,
      "mean_token_accuracy": 0.20895386338233948,
      "num_tokens": 98562562.0,
      "step": 48650
    },
    {
      "entropy": 5.722162532806396,
      "epoch": 4.2519444201695356,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003308489769231811,
      "loss": 5.209,
      "mean_token_accuracy": 0.2137931689620018,
      "num_tokens": 98572516.0,
      "step": 48655
    },
    {
      "entropy": 5.76389627456665,
      "epoch": 4.252381368522241,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003308187904986099,
      "loss": 5.2888,
      "mean_token_accuracy": 0.20532971173524855,
      "num_tokens": 98583725.0,
      "step": 48660
    },
    {
      "entropy": 5.786729431152343,
      "epoch": 4.252818316874945,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033078860300340496,
      "loss": 5.2595,
      "mean_token_accuracy": 0.2083605408668518,
      "num_tokens": 98595644.0,
      "step": 48665
    },
    {
      "entropy": 5.7550684928894045,
      "epoch": 4.25325526522765,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033075841443814543,
      "loss": 5.1756,
      "mean_token_accuracy": 0.21603538393974303,
      "num_tokens": 98605206.0,
      "step": 48670
    },
    {
      "entropy": 5.712480592727661,
      "epoch": 4.253692213580355,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033072822480341023,
      "loss": 5.2298,
      "mean_token_accuracy": 0.2015776589512825,
      "num_tokens": 98616129.0,
      "step": 48675
    },
    {
      "entropy": 5.735281276702881,
      "epoch": 4.25412916193306,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003306980340997786,
      "loss": 5.2924,
      "mean_token_accuracy": 0.20500825494527816,
      "num_tokens": 98626710.0,
      "step": 48680
    },
    {
      "entropy": 5.78398756980896,
      "epoch": 4.254566110285764,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003306678423278294,
      "loss": 5.1112,
      "mean_token_accuracy": 0.22230679541826248,
      "num_tokens": 98636458.0,
      "step": 48685
    },
    {
      "entropy": 5.800870180130005,
      "epoch": 4.255003058638469,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033063764948814187,
      "loss": 5.265,
      "mean_token_accuracy": 0.20680535584688187,
      "num_tokens": 98647033.0,
      "step": 48690
    },
    {
      "entropy": 5.72044939994812,
      "epoch": 4.255440006991174,
      "grad_norm": 1.140625,
      "learning_rate": 0.000330607455581295,
      "loss": 5.2539,
      "mean_token_accuracy": 0.20475162267684938,
      "num_tokens": 98657939.0,
      "step": 48695
    },
    {
      "entropy": 5.791774272918701,
      "epoch": 4.255876955343878,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033057726060786794,
      "loss": 5.2686,
      "mean_token_accuracy": 0.20466097742319106,
      "num_tokens": 98667453.0,
      "step": 48700
    },
    {
      "entropy": 5.769132852554321,
      "epoch": 4.256313903696583,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00033054706456844005,
      "loss": 5.2373,
      "mean_token_accuracy": 0.20556141585111617,
      "num_tokens": 98678951.0,
      "step": 48705
    },
    {
      "entropy": 5.731075239181519,
      "epoch": 4.256750852049287,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033051686746359026,
      "loss": 5.1556,
      "mean_token_accuracy": 0.21545912325382233,
      "num_tokens": 98689427.0,
      "step": 48710
    },
    {
      "entropy": 5.752020215988159,
      "epoch": 4.257187800401993,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003304866692938979,
      "loss": 5.2182,
      "mean_token_accuracy": 0.20512035489082336,
      "num_tokens": 98699671.0,
      "step": 48715
    },
    {
      "entropy": 5.807413768768311,
      "epoch": 4.257624748754697,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033045647005994216,
      "loss": 5.2375,
      "mean_token_accuracy": 0.2048241153359413,
      "num_tokens": 98709357.0,
      "step": 48720
    },
    {
      "entropy": 5.780787706375122,
      "epoch": 4.258061697107402,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003304262697623022,
      "loss": 5.294,
      "mean_token_accuracy": 0.21002990156412124,
      "num_tokens": 98720144.0,
      "step": 48725
    },
    {
      "entropy": 5.813510179519653,
      "epoch": 4.258498645460106,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033039606840155737,
      "loss": 5.2709,
      "mean_token_accuracy": 0.20236214697360994,
      "num_tokens": 98731125.0,
      "step": 48730
    },
    {
      "entropy": 5.750083684921265,
      "epoch": 4.2589355938128115,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033036586597828696,
      "loss": 5.222,
      "mean_token_accuracy": 0.2129952698945999,
      "num_tokens": 98741745.0,
      "step": 48735
    },
    {
      "entropy": 5.7210334777832035,
      "epoch": 4.259372542165516,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033033566249307024,
      "loss": 5.2624,
      "mean_token_accuracy": 0.20850825309753418,
      "num_tokens": 98750970.0,
      "step": 48740
    },
    {
      "entropy": 5.73559799194336,
      "epoch": 4.25980949051822,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003303054579464864,
      "loss": 5.2561,
      "mean_token_accuracy": 0.2046366050839424,
      "num_tokens": 98761473.0,
      "step": 48745
    },
    {
      "entropy": 5.834521102905273,
      "epoch": 4.260246438870926,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033027525233911504,
      "loss": 5.2851,
      "mean_token_accuracy": 0.20358264148235322,
      "num_tokens": 98771265.0,
      "step": 48750
    },
    {
      "entropy": 5.76550440788269,
      "epoch": 4.26068338722363,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003302450456715353,
      "loss": 5.2279,
      "mean_token_accuracy": 0.2096568301320076,
      "num_tokens": 98781297.0,
      "step": 48755
    },
    {
      "entropy": 5.669555139541626,
      "epoch": 4.261120335576335,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033021483794432667,
      "loss": 5.104,
      "mean_token_accuracy": 0.22128800600767135,
      "num_tokens": 98790804.0,
      "step": 48760
    },
    {
      "entropy": 5.7662629127502445,
      "epoch": 4.261557283929039,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003301846291580685,
      "loss": 5.2729,
      "mean_token_accuracy": 0.20275106132030488,
      "num_tokens": 98800788.0,
      "step": 48765
    },
    {
      "entropy": 5.768613958358765,
      "epoch": 4.261994232281745,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003301544193133402,
      "loss": 5.3078,
      "mean_token_accuracy": 0.21057666540145875,
      "num_tokens": 98811627.0,
      "step": 48770
    },
    {
      "entropy": 5.811990976333618,
      "epoch": 4.262431180634449,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033012420841072135,
      "loss": 5.2863,
      "mean_token_accuracy": 0.2010498195886612,
      "num_tokens": 98822206.0,
      "step": 48775
    },
    {
      "entropy": 5.774279260635376,
      "epoch": 4.2628681289871535,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033009399645079113,
      "loss": 5.2638,
      "mean_token_accuracy": 0.20535696297883987,
      "num_tokens": 98832394.0,
      "step": 48780
    },
    {
      "entropy": 5.749160575866699,
      "epoch": 4.263305077339858,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003300637834341292,
      "loss": 5.2162,
      "mean_token_accuracy": 0.21034561693668366,
      "num_tokens": 98842357.0,
      "step": 48785
    },
    {
      "entropy": 5.782548570632935,
      "epoch": 4.263742025692563,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033003356936131523,
      "loss": 5.1569,
      "mean_token_accuracy": 0.20908086746931076,
      "num_tokens": 98852376.0,
      "step": 48790
    },
    {
      "entropy": 5.746486186981201,
      "epoch": 4.264178974045268,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003300033542329284,
      "loss": 5.2516,
      "mean_token_accuracy": 0.20528248995542525,
      "num_tokens": 98862473.0,
      "step": 48795
    },
    {
      "entropy": 5.765820789337158,
      "epoch": 4.264615922397972,
      "grad_norm": 1.125,
      "learning_rate": 0.0003299731380495484,
      "loss": 5.2815,
      "mean_token_accuracy": 0.20305787026882172,
      "num_tokens": 98872915.0,
      "step": 48800
    },
    {
      "entropy": 5.7811949253082275,
      "epoch": 4.265052870750678,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000329942920811755,
      "loss": 5.292,
      "mean_token_accuracy": 0.19892615973949432,
      "num_tokens": 98883863.0,
      "step": 48805
    },
    {
      "entropy": 5.778256940841675,
      "epoch": 4.265489819103382,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003299127025201274,
      "loss": 5.2254,
      "mean_token_accuracy": 0.20551557689905167,
      "num_tokens": 98893507.0,
      "step": 48810
    },
    {
      "entropy": 5.798488283157349,
      "epoch": 4.265926767456087,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00032988248317524546,
      "loss": 5.2571,
      "mean_token_accuracy": 0.2080301448702812,
      "num_tokens": 98904109.0,
      "step": 48815
    },
    {
      "entropy": 5.699231100082398,
      "epoch": 4.266363715808791,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003298522627776887,
      "loss": 5.2265,
      "mean_token_accuracy": 0.20342027097940446,
      "num_tokens": 98914017.0,
      "step": 48820
    },
    {
      "entropy": 5.7488706588745115,
      "epoch": 4.266800664161496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003298220413280368,
      "loss": 5.1726,
      "mean_token_accuracy": 0.2068385139107704,
      "num_tokens": 98924096.0,
      "step": 48825
    },
    {
      "entropy": 5.792304849624633,
      "epoch": 4.267237612514201,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003297918188268695,
      "loss": 5.2589,
      "mean_token_accuracy": 0.20417648255825044,
      "num_tokens": 98934359.0,
      "step": 48830
    },
    {
      "entropy": 5.772960042953491,
      "epoch": 4.267674560866905,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032976159527476636,
      "loss": 5.2486,
      "mean_token_accuracy": 0.2126539096236229,
      "num_tokens": 98944557.0,
      "step": 48835
    },
    {
      "entropy": 5.836574554443359,
      "epoch": 4.26811150921961,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003297313706723071,
      "loss": 5.2513,
      "mean_token_accuracy": 0.20460422486066818,
      "num_tokens": 98954066.0,
      "step": 48840
    },
    {
      "entropy": 5.687513637542724,
      "epoch": 4.268548457572315,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003297011450200715,
      "loss": 5.1213,
      "mean_token_accuracy": 0.21791635900735856,
      "num_tokens": 98963789.0,
      "step": 48845
    },
    {
      "entropy": 5.6985218048095705,
      "epoch": 4.26898540592502,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003296709183186393,
      "loss": 5.1688,
      "mean_token_accuracy": 0.21462533622980118,
      "num_tokens": 98973441.0,
      "step": 48850
    },
    {
      "entropy": 5.755676221847534,
      "epoch": 4.269422354277724,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003296406905685902,
      "loss": 5.202,
      "mean_token_accuracy": 0.2079242706298828,
      "num_tokens": 98983362.0,
      "step": 48855
    },
    {
      "entropy": 5.754165267944336,
      "epoch": 4.2698593026304295,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032961046177050403,
      "loss": 5.1869,
      "mean_token_accuracy": 0.20850545316934585,
      "num_tokens": 98993174.0,
      "step": 48860
    },
    {
      "entropy": 5.746628189086914,
      "epoch": 4.270296250983134,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003295802319249606,
      "loss": 5.2299,
      "mean_token_accuracy": 0.2076789543032646,
      "num_tokens": 99002092.0,
      "step": 48865
    },
    {
      "entropy": 5.729915189743042,
      "epoch": 4.270733199335838,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003295500010325397,
      "loss": 5.2601,
      "mean_token_accuracy": 0.20760640352964402,
      "num_tokens": 99012818.0,
      "step": 48870
    },
    {
      "entropy": 5.752954149246216,
      "epoch": 4.271170147688543,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003295197690938212,
      "loss": 5.1887,
      "mean_token_accuracy": 0.20771627575159074,
      "num_tokens": 99022988.0,
      "step": 48875
    },
    {
      "entropy": 5.749655628204346,
      "epoch": 4.271607096041248,
      "grad_norm": 1.0625,
      "learning_rate": 0.000329489536109385,
      "loss": 5.2641,
      "mean_token_accuracy": 0.20235336273908616,
      "num_tokens": 99032884.0,
      "step": 48880
    },
    {
      "entropy": 5.7827942848205565,
      "epoch": 4.272044044393953,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003294593020798109,
      "loss": 5.2422,
      "mean_token_accuracy": 0.2043704181909561,
      "num_tokens": 99042274.0,
      "step": 48885
    },
    {
      "entropy": 5.693536472320557,
      "epoch": 4.272480992746657,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00032942906700567884,
      "loss": 5.168,
      "mean_token_accuracy": 0.21897029876708984,
      "num_tokens": 99052385.0,
      "step": 48890
    },
    {
      "entropy": 5.675080251693726,
      "epoch": 4.272917941099362,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032939883088756877,
      "loss": 5.1288,
      "mean_token_accuracy": 0.22363777607679367,
      "num_tokens": 99062589.0,
      "step": 48895
    },
    {
      "entropy": 5.789471530914307,
      "epoch": 4.273354889452067,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032936859372606063,
      "loss": 5.2369,
      "mean_token_accuracy": 0.2053367018699646,
      "num_tokens": 99072794.0,
      "step": 48900
    },
    {
      "entropy": 5.790174436569214,
      "epoch": 4.2737918378047715,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032933835552173443,
      "loss": 5.2623,
      "mean_token_accuracy": 0.205574694275856,
      "num_tokens": 99083759.0,
      "step": 48905
    },
    {
      "entropy": 5.839432430267334,
      "epoch": 4.274228786157476,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032930811627517,
      "loss": 5.3184,
      "mean_token_accuracy": 0.20282484889030455,
      "num_tokens": 99094056.0,
      "step": 48910
    },
    {
      "entropy": 5.78317928314209,
      "epoch": 4.274665734510181,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003292778759869475,
      "loss": 5.231,
      "mean_token_accuracy": 0.20785182416439058,
      "num_tokens": 99104287.0,
      "step": 48915
    },
    {
      "entropy": 5.737963771820068,
      "epoch": 4.275102682862886,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032924763465764676,
      "loss": 5.1806,
      "mean_token_accuracy": 0.21709947139024735,
      "num_tokens": 99114322.0,
      "step": 48920
    },
    {
      "entropy": 5.745499897003174,
      "epoch": 4.27553963121559,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003292173922878481,
      "loss": 5.1839,
      "mean_token_accuracy": 0.21364469528198243,
      "num_tokens": 99124145.0,
      "step": 48925
    },
    {
      "entropy": 5.627245950698852,
      "epoch": 4.275976579568295,
      "grad_norm": 1.703125,
      "learning_rate": 0.0003291871488781315,
      "loss": 5.0833,
      "mean_token_accuracy": 0.22428788095712662,
      "num_tokens": 99135388.0,
      "step": 48930
    },
    {
      "entropy": 5.664234638214111,
      "epoch": 4.276413527921,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003291569044290769,
      "loss": 5.1873,
      "mean_token_accuracy": 0.21546584516763687,
      "num_tokens": 99145615.0,
      "step": 48935
    },
    {
      "entropy": 5.72129979133606,
      "epoch": 4.2768504762737045,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003291266589412645,
      "loss": 5.2822,
      "mean_token_accuracy": 0.1999499186873436,
      "num_tokens": 99155805.0,
      "step": 48940
    },
    {
      "entropy": 5.8337749481201175,
      "epoch": 4.277287424626409,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003290964124152744,
      "loss": 5.2761,
      "mean_token_accuracy": 0.20566605031490326,
      "num_tokens": 99165124.0,
      "step": 48945
    },
    {
      "entropy": 5.766807270050049,
      "epoch": 4.2777243729791135,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032906616485168686,
      "loss": 5.2695,
      "mean_token_accuracy": 0.20290515273809434,
      "num_tokens": 99175724.0,
      "step": 48950
    },
    {
      "entropy": 5.758070564270019,
      "epoch": 4.278161321331819,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003290359162510819,
      "loss": 5.2294,
      "mean_token_accuracy": 0.2061726465821266,
      "num_tokens": 99184972.0,
      "step": 48955
    },
    {
      "entropy": 5.7587628841400145,
      "epoch": 4.278598269684523,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032900566661403974,
      "loss": 5.2312,
      "mean_token_accuracy": 0.2059638425707817,
      "num_tokens": 99194646.0,
      "step": 48960
    },
    {
      "entropy": 5.673467206954956,
      "epoch": 4.279035218037228,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032897541594114054,
      "loss": 5.1799,
      "mean_token_accuracy": 0.21622241884469987,
      "num_tokens": 99203778.0,
      "step": 48965
    },
    {
      "entropy": 5.765642929077148,
      "epoch": 4.279472166389933,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032894516423296466,
      "loss": 5.2081,
      "mean_token_accuracy": 0.21364642828702926,
      "num_tokens": 99213365.0,
      "step": 48970
    },
    {
      "entropy": 5.8164348125457765,
      "epoch": 4.279909114742638,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032891491149009225,
      "loss": 5.2406,
      "mean_token_accuracy": 0.20590434968471527,
      "num_tokens": 99223967.0,
      "step": 48975
    },
    {
      "entropy": 5.749003314971924,
      "epoch": 4.280346063095342,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003288846577131035,
      "loss": 5.2287,
      "mean_token_accuracy": 0.20761240720748902,
      "num_tokens": 99234115.0,
      "step": 48980
    },
    {
      "entropy": 5.839884185791016,
      "epoch": 4.2807830114480465,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003288544029025789,
      "loss": 5.2576,
      "mean_token_accuracy": 0.20186434984207152,
      "num_tokens": 99244405.0,
      "step": 48985
    },
    {
      "entropy": 5.795174074172974,
      "epoch": 4.281219959800752,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032882414705909854,
      "loss": 5.2816,
      "mean_token_accuracy": 0.20186886489391326,
      "num_tokens": 99256042.0,
      "step": 48990
    },
    {
      "entropy": 5.780945062637329,
      "epoch": 4.281656908153456,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003287938901832429,
      "loss": 5.2012,
      "mean_token_accuracy": 0.21568830758333207,
      "num_tokens": 99266319.0,
      "step": 48995
    },
    {
      "entropy": 5.80551061630249,
      "epoch": 4.282093856506161,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003287636322755923,
      "loss": 5.179,
      "mean_token_accuracy": 0.20678154230117798,
      "num_tokens": 99277062.0,
      "step": 49000
    },
    {
      "entropy": 5.756753826141358,
      "epoch": 4.282530804858865,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032873337333672693,
      "loss": 5.2153,
      "mean_token_accuracy": 0.20773729532957078,
      "num_tokens": 99287130.0,
      "step": 49005
    },
    {
      "entropy": 5.659932708740234,
      "epoch": 4.282967753211571,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003287031133672274,
      "loss": 5.1294,
      "mean_token_accuracy": 0.2218789279460907,
      "num_tokens": 99297721.0,
      "step": 49010
    },
    {
      "entropy": 5.772473764419556,
      "epoch": 4.283404701564275,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032867285236767396,
      "loss": 5.2099,
      "mean_token_accuracy": 0.21172335892915725,
      "num_tokens": 99307743.0,
      "step": 49015
    },
    {
      "entropy": 5.723956727981568,
      "epoch": 4.28384164991698,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003286425903386472,
      "loss": 5.205,
      "mean_token_accuracy": 0.2120936781167984,
      "num_tokens": 99317520.0,
      "step": 49020
    },
    {
      "entropy": 5.751600885391236,
      "epoch": 4.284278598269685,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003286123272807273,
      "loss": 5.2422,
      "mean_token_accuracy": 0.2062476620078087,
      "num_tokens": 99327768.0,
      "step": 49025
    },
    {
      "entropy": 5.764154720306396,
      "epoch": 4.284715546622389,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032858206319449494,
      "loss": 5.266,
      "mean_token_accuracy": 0.20054085552692413,
      "num_tokens": 99337742.0,
      "step": 49030
    },
    {
      "entropy": 5.690527534484863,
      "epoch": 4.285152494975094,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032855179808053054,
      "loss": 5.2079,
      "mean_token_accuracy": 0.2097321182489395,
      "num_tokens": 99347069.0,
      "step": 49035
    },
    {
      "entropy": 5.7290260791778564,
      "epoch": 4.285589443327798,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003285215319394146,
      "loss": 5.247,
      "mean_token_accuracy": 0.20650645345449448,
      "num_tokens": 99357801.0,
      "step": 49040
    },
    {
      "entropy": 5.829774665832519,
      "epoch": 4.286026391680504,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003284912647717276,
      "loss": 5.2966,
      "mean_token_accuracy": 0.20049884021282197,
      "num_tokens": 99368451.0,
      "step": 49045
    },
    {
      "entropy": 5.830839490890503,
      "epoch": 4.286463340033208,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003284609965780501,
      "loss": 5.2638,
      "mean_token_accuracy": 0.20517192482948304,
      "num_tokens": 99378924.0,
      "step": 49050
    },
    {
      "entropy": 5.757049417495727,
      "epoch": 4.286900288385913,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003284307273589627,
      "loss": 5.2024,
      "mean_token_accuracy": 0.21165191978216172,
      "num_tokens": 99387802.0,
      "step": 49055
    },
    {
      "entropy": 5.716234350204468,
      "epoch": 4.287337236738617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032840045711504597,
      "loss": 5.1322,
      "mean_token_accuracy": 0.21268866062164307,
      "num_tokens": 99398172.0,
      "step": 49060
    },
    {
      "entropy": 5.69637017250061,
      "epoch": 4.2877741850913225,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032837018584688045,
      "loss": 5.2173,
      "mean_token_accuracy": 0.21173731684684755,
      "num_tokens": 99408563.0,
      "step": 49065
    },
    {
      "entropy": 5.759154510498047,
      "epoch": 4.288211133444027,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003283399135550469,
      "loss": 5.2393,
      "mean_token_accuracy": 0.20743643045425414,
      "num_tokens": 99418542.0,
      "step": 49070
    },
    {
      "entropy": 5.676146030426025,
      "epoch": 4.288648081796731,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003283096402401257,
      "loss": 5.2416,
      "mean_token_accuracy": 0.2131197154521942,
      "num_tokens": 99428145.0,
      "step": 49075
    },
    {
      "entropy": 5.727808427810669,
      "epoch": 4.289085030149437,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032827936590269774,
      "loss": 5.2228,
      "mean_token_accuracy": 0.20563806295394899,
      "num_tokens": 99439460.0,
      "step": 49080
    },
    {
      "entropy": 5.811910009384155,
      "epoch": 4.289521978502141,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003282490905433437,
      "loss": 5.2661,
      "mean_token_accuracy": 0.20407632887363433,
      "num_tokens": 99448781.0,
      "step": 49085
    },
    {
      "entropy": 5.725295639038086,
      "epoch": 4.289958926854846,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032821881416264417,
      "loss": 5.1684,
      "mean_token_accuracy": 0.2110238030552864,
      "num_tokens": 99459945.0,
      "step": 49090
    },
    {
      "entropy": 5.743163347244263,
      "epoch": 4.29039587520755,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003281885367611798,
      "loss": 5.1919,
      "mean_token_accuracy": 0.21184259951114653,
      "num_tokens": 99469510.0,
      "step": 49095
    },
    {
      "entropy": 5.733627653121948,
      "epoch": 4.2908328235602555,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003281582583395315,
      "loss": 5.2644,
      "mean_token_accuracy": 0.20790978372097016,
      "num_tokens": 99480409.0,
      "step": 49100
    },
    {
      "entropy": 5.783175611495972,
      "epoch": 4.29126977191296,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032812797889827987,
      "loss": 5.2793,
      "mean_token_accuracy": 0.20428525656461716,
      "num_tokens": 99490044.0,
      "step": 49105
    },
    {
      "entropy": 5.846653890609741,
      "epoch": 4.2917067202656645,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032809769843800585,
      "loss": 5.2533,
      "mean_token_accuracy": 0.20599287152290344,
      "num_tokens": 99500257.0,
      "step": 49110
    },
    {
      "entropy": 5.760273218154907,
      "epoch": 4.292143668618369,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032806741695929015,
      "loss": 5.2086,
      "mean_token_accuracy": 0.20693501979112625,
      "num_tokens": 99509864.0,
      "step": 49115
    },
    {
      "entropy": 5.745591020584106,
      "epoch": 4.292580616971074,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032803713446271363,
      "loss": 5.1943,
      "mean_token_accuracy": 0.21501611769199372,
      "num_tokens": 99520277.0,
      "step": 49120
    },
    {
      "entropy": 5.8470062732696535,
      "epoch": 4.293017565323779,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00032800685094885694,
      "loss": 5.3512,
      "mean_token_accuracy": 0.19141287952661515,
      "num_tokens": 99531890.0,
      "step": 49125
    },
    {
      "entropy": 5.694793844223023,
      "epoch": 4.293454513676483,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032797656641830116,
      "loss": 5.0422,
      "mean_token_accuracy": 0.21977217197418214,
      "num_tokens": 99541786.0,
      "step": 49130
    },
    {
      "entropy": 5.766321802139283,
      "epoch": 4.293891462029189,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003279462808716271,
      "loss": 5.1841,
      "mean_token_accuracy": 0.21121790260076523,
      "num_tokens": 99551665.0,
      "step": 49135
    },
    {
      "entropy": 5.712840938568116,
      "epoch": 4.294328410381893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003279159943094156,
      "loss": 5.2328,
      "mean_token_accuracy": 0.2039630264043808,
      "num_tokens": 99561350.0,
      "step": 49140
    },
    {
      "entropy": 5.744746923446655,
      "epoch": 4.2947653587345975,
      "grad_norm": 1.125,
      "learning_rate": 0.0003278857067322476,
      "loss": 5.2177,
      "mean_token_accuracy": 0.21277221888303757,
      "num_tokens": 99571473.0,
      "step": 49145
    },
    {
      "entropy": 5.819475317001343,
      "epoch": 4.295202307087302,
      "grad_norm": 1.046875,
      "learning_rate": 0.000327855418140704,
      "loss": 5.243,
      "mean_token_accuracy": 0.20742502957582473,
      "num_tokens": 99580113.0,
      "step": 49150
    },
    {
      "entropy": 5.773524236679077,
      "epoch": 4.295639255440007,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003278251285353658,
      "loss": 5.2325,
      "mean_token_accuracy": 0.20674293488264084,
      "num_tokens": 99591660.0,
      "step": 49155
    },
    {
      "entropy": 5.722611999511718,
      "epoch": 4.296076203792712,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032779483791681396,
      "loss": 5.1779,
      "mean_token_accuracy": 0.21427285224199294,
      "num_tokens": 99601470.0,
      "step": 49160
    },
    {
      "entropy": 5.724073076248169,
      "epoch": 4.296513152145416,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032776454628562945,
      "loss": 5.2434,
      "mean_token_accuracy": 0.21248802840709685,
      "num_tokens": 99612327.0,
      "step": 49165
    },
    {
      "entropy": 5.705012798309326,
      "epoch": 4.296950100498121,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003277342536423933,
      "loss": 5.2191,
      "mean_token_accuracy": 0.20945999026298523,
      "num_tokens": 99621879.0,
      "step": 49170
    },
    {
      "entropy": 5.821924924850464,
      "epoch": 4.297387048850826,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003277039599876866,
      "loss": 5.2363,
      "mean_token_accuracy": 0.20202989727258683,
      "num_tokens": 99632052.0,
      "step": 49175
    },
    {
      "entropy": 5.879569578170776,
      "epoch": 4.297823997203531,
      "grad_norm": 1.0,
      "learning_rate": 0.00032767366532209023,
      "loss": 5.3776,
      "mean_token_accuracy": 0.19769757837057114,
      "num_tokens": 99643128.0,
      "step": 49180
    },
    {
      "entropy": 5.6893545627594,
      "epoch": 4.298260945556235,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032764336964618553,
      "loss": 5.1778,
      "mean_token_accuracy": 0.21894370764493942,
      "num_tokens": 99654043.0,
      "step": 49185
    },
    {
      "entropy": 5.742464828491211,
      "epoch": 4.29869789390894,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032761307296055327,
      "loss": 5.2459,
      "mean_token_accuracy": 0.2191907212138176,
      "num_tokens": 99664889.0,
      "step": 49190
    },
    {
      "entropy": 5.820853853225708,
      "epoch": 4.299134842261645,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032758277526577474,
      "loss": 5.2954,
      "mean_token_accuracy": 0.1979976549744606,
      "num_tokens": 99675025.0,
      "step": 49195
    },
    {
      "entropy": 5.798661470413208,
      "epoch": 4.299571790614349,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032755247656243095,
      "loss": 5.3519,
      "mean_token_accuracy": 0.19922257512807845,
      "num_tokens": 99685947.0,
      "step": 49200
    },
    {
      "entropy": 5.738269567489624,
      "epoch": 4.300008738967054,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003275221768511033,
      "loss": 5.2582,
      "mean_token_accuracy": 0.2011851340532303,
      "num_tokens": 99695787.0,
      "step": 49205
    },
    {
      "entropy": 5.724356746673584,
      "epoch": 4.300445687319759,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003274918761323726,
      "loss": 5.1736,
      "mean_token_accuracy": 0.21275639832019805,
      "num_tokens": 99705489.0,
      "step": 49210
    },
    {
      "entropy": 5.854618549346924,
      "epoch": 4.300882635672464,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032746157440682026,
      "loss": 5.3005,
      "mean_token_accuracy": 0.1970829337835312,
      "num_tokens": 99715580.0,
      "step": 49215
    },
    {
      "entropy": 5.773229789733887,
      "epoch": 4.301319584025168,
      "grad_norm": 1.125,
      "learning_rate": 0.00032743127167502746,
      "loss": 5.2733,
      "mean_token_accuracy": 0.2030945226550102,
      "num_tokens": 99725532.0,
      "step": 49220
    },
    {
      "entropy": 5.687661075592041,
      "epoch": 4.301756532377873,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032740096793757535,
      "loss": 5.1442,
      "mean_token_accuracy": 0.2142956018447876,
      "num_tokens": 99736673.0,
      "step": 49225
    },
    {
      "entropy": 5.781227922439575,
      "epoch": 4.302193480730578,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032737066319504527,
      "loss": 5.2431,
      "mean_token_accuracy": 0.2100657269358635,
      "num_tokens": 99747394.0,
      "step": 49230
    },
    {
      "entropy": 5.801958179473877,
      "epoch": 4.302630429083282,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003273403574480184,
      "loss": 5.2341,
      "mean_token_accuracy": 0.21151113063097,
      "num_tokens": 99757671.0,
      "step": 49235
    },
    {
      "entropy": 5.765813589096069,
      "epoch": 4.303067377435987,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003273100506970761,
      "loss": 5.2257,
      "mean_token_accuracy": 0.20738303363323213,
      "num_tokens": 99767937.0,
      "step": 49240
    },
    {
      "entropy": 5.79308443069458,
      "epoch": 4.303504325788692,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032727974294279955,
      "loss": 5.2974,
      "mean_token_accuracy": 0.205491741001606,
      "num_tokens": 99778300.0,
      "step": 49245
    },
    {
      "entropy": 5.765468263626099,
      "epoch": 4.303941274141397,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003272494341857701,
      "loss": 5.2031,
      "mean_token_accuracy": 0.2069451168179512,
      "num_tokens": 99787740.0,
      "step": 49250
    },
    {
      "entropy": 5.759879493713379,
      "epoch": 4.304378222494101,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003272191244265693,
      "loss": 5.2676,
      "mean_token_accuracy": 0.20504757463932038,
      "num_tokens": 99797436.0,
      "step": 49255
    },
    {
      "entropy": 5.71657133102417,
      "epoch": 4.304815170846806,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032718881366577827,
      "loss": 5.2246,
      "mean_token_accuracy": 0.21680000871419908,
      "num_tokens": 99807293.0,
      "step": 49260
    },
    {
      "entropy": 5.723176193237305,
      "epoch": 4.305252119199511,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003271585019039783,
      "loss": 5.2378,
      "mean_token_accuracy": 0.21042994707822799,
      "num_tokens": 99816369.0,
      "step": 49265
    },
    {
      "entropy": 5.756451559066773,
      "epoch": 4.3056890675522155,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000327128189141751,
      "loss": 5.1759,
      "mean_token_accuracy": 0.21255509108304976,
      "num_tokens": 99825858.0,
      "step": 49270
    },
    {
      "entropy": 5.773692989349366,
      "epoch": 4.30612601590492,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003270978753796778,
      "loss": 5.3381,
      "mean_token_accuracy": 0.19682026356458665,
      "num_tokens": 99836664.0,
      "step": 49275
    },
    {
      "entropy": 5.83559284210205,
      "epoch": 4.306562964257624,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032706756061834006,
      "loss": 5.301,
      "mean_token_accuracy": 0.2048583969473839,
      "num_tokens": 99846497.0,
      "step": 49280
    },
    {
      "entropy": 5.844741725921631,
      "epoch": 4.30699991261033,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003270372448583191,
      "loss": 5.3153,
      "mean_token_accuracy": 0.20322612524032593,
      "num_tokens": 99856039.0,
      "step": 49285
    },
    {
      "entropy": 5.766893577575684,
      "epoch": 4.307436860963034,
      "grad_norm": 1.125,
      "learning_rate": 0.00032700692810019654,
      "loss": 5.2436,
      "mean_token_accuracy": 0.2121206432580948,
      "num_tokens": 99866541.0,
      "step": 49290
    },
    {
      "entropy": 5.7296568870544435,
      "epoch": 4.307873809315739,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032697661034455396,
      "loss": 5.2021,
      "mean_token_accuracy": 0.20835048258304595,
      "num_tokens": 99877550.0,
      "step": 49295
    },
    {
      "entropy": 5.719585943222046,
      "epoch": 4.308310757668443,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003269462915919726,
      "loss": 5.145,
      "mean_token_accuracy": 0.21641955971717836,
      "num_tokens": 99887653.0,
      "step": 49300
    },
    {
      "entropy": 5.85492205619812,
      "epoch": 4.3087477060211485,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003269159718430343,
      "loss": 5.2383,
      "mean_token_accuracy": 0.20733308494091035,
      "num_tokens": 99896900.0,
      "step": 49305
    },
    {
      "entropy": 5.794846677780152,
      "epoch": 4.309184654373853,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003268856510983204,
      "loss": 5.263,
      "mean_token_accuracy": 0.20813482254743576,
      "num_tokens": 99907019.0,
      "step": 49310
    },
    {
      "entropy": 5.708095741271973,
      "epoch": 4.3096216027265575,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003268553293584125,
      "loss": 5.2254,
      "mean_token_accuracy": 0.20928152054548263,
      "num_tokens": 99917212.0,
      "step": 49315
    },
    {
      "entropy": 5.810342693328858,
      "epoch": 4.310058551079263,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032682500662389223,
      "loss": 5.3222,
      "mean_token_accuracy": 0.1976830244064331,
      "num_tokens": 99927239.0,
      "step": 49320
    },
    {
      "entropy": 5.7316399097442625,
      "epoch": 4.310495499431967,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003267946828953411,
      "loss": 5.1928,
      "mean_token_accuracy": 0.21478842049837113,
      "num_tokens": 99937490.0,
      "step": 49325
    },
    {
      "entropy": 5.808409690856934,
      "epoch": 4.310932447784672,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003267643581733409,
      "loss": 5.3198,
      "mean_token_accuracy": 0.2051791340112686,
      "num_tokens": 99947083.0,
      "step": 49330
    },
    {
      "entropy": 5.766139030456543,
      "epoch": 4.311369396137376,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003267340324584732,
      "loss": 5.2311,
      "mean_token_accuracy": 0.21344813704490662,
      "num_tokens": 99957842.0,
      "step": 49335
    },
    {
      "entropy": 5.7631980895996096,
      "epoch": 4.311806344490082,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003267037057513196,
      "loss": 5.2025,
      "mean_token_accuracy": 0.21817411184310914,
      "num_tokens": 99968384.0,
      "step": 49340
    },
    {
      "entropy": 5.753182077407837,
      "epoch": 4.312243292842786,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003266733780524618,
      "loss": 5.2242,
      "mean_token_accuracy": 0.20820482969284057,
      "num_tokens": 99978749.0,
      "step": 49345
    },
    {
      "entropy": 5.723936653137207,
      "epoch": 4.3126802411954905,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003266430493624816,
      "loss": 5.2192,
      "mean_token_accuracy": 0.21239617764949797,
      "num_tokens": 99988335.0,
      "step": 49350
    },
    {
      "entropy": 5.744138765335083,
      "epoch": 4.313117189548195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032661271968196067,
      "loss": 5.2081,
      "mean_token_accuracy": 0.21025354862213136,
      "num_tokens": 99999657.0,
      "step": 49355
    },
    {
      "entropy": 5.7764829158782955,
      "epoch": 4.3135541379009,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032658238901148075,
      "loss": 5.1634,
      "mean_token_accuracy": 0.2175203323364258,
      "num_tokens": 100009604.0,
      "step": 49360
    },
    {
      "entropy": 5.780533027648926,
      "epoch": 4.313991086253605,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003265520573516235,
      "loss": 5.1934,
      "mean_token_accuracy": 0.21180597096681594,
      "num_tokens": 100019372.0,
      "step": 49365
    },
    {
      "entropy": 5.712290239334107,
      "epoch": 4.314428034606309,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003265217247029708,
      "loss": 5.2102,
      "mean_token_accuracy": 0.21067940592765808,
      "num_tokens": 100029097.0,
      "step": 49370
    },
    {
      "entropy": 5.726408910751343,
      "epoch": 4.314864982959014,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003264913910661044,
      "loss": 5.1949,
      "mean_token_accuracy": 0.20894595086574555,
      "num_tokens": 100039506.0,
      "step": 49375
    },
    {
      "entropy": 5.720028495788574,
      "epoch": 4.315301931311719,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032646105644160617,
      "loss": 5.1133,
      "mean_token_accuracy": 0.2174106016755104,
      "num_tokens": 100049721.0,
      "step": 49380
    },
    {
      "entropy": 5.804169225692749,
      "epoch": 4.315738879664424,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003264307208300579,
      "loss": 5.3153,
      "mean_token_accuracy": 0.19654495865106583,
      "num_tokens": 100059329.0,
      "step": 49385
    },
    {
      "entropy": 5.701730060577392,
      "epoch": 4.316175828017128,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003264003842320415,
      "loss": 5.1752,
      "mean_token_accuracy": 0.21018291115760804,
      "num_tokens": 100069237.0,
      "step": 49390
    },
    {
      "entropy": 5.776158666610717,
      "epoch": 4.316612776369833,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032637004664813873,
      "loss": 5.3191,
      "mean_token_accuracy": 0.1951804980635643,
      "num_tokens": 100079475.0,
      "step": 49395
    },
    {
      "entropy": 5.778986310958862,
      "epoch": 4.317049724722538,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003263397080789316,
      "loss": 5.1965,
      "mean_token_accuracy": 0.20818117409944534,
      "num_tokens": 100089135.0,
      "step": 49400
    },
    {
      "entropy": 5.756512928009033,
      "epoch": 4.317486673075242,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032630936852500196,
      "loss": 5.1682,
      "mean_token_accuracy": 0.21676040440797806,
      "num_tokens": 100098465.0,
      "step": 49405
    },
    {
      "entropy": 5.737497186660766,
      "epoch": 4.317923621427947,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003262790279869317,
      "loss": 5.2312,
      "mean_token_accuracy": 0.2073950082063675,
      "num_tokens": 100110057.0,
      "step": 49410
    },
    {
      "entropy": 5.768460512161255,
      "epoch": 4.318360569780652,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032624868646530285,
      "loss": 5.1858,
      "mean_token_accuracy": 0.21024761348962784,
      "num_tokens": 100119127.0,
      "step": 49415
    },
    {
      "entropy": 5.7390251636505125,
      "epoch": 4.318797518133357,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003262183439606974,
      "loss": 5.2023,
      "mean_token_accuracy": 0.2117741197347641,
      "num_tokens": 100129528.0,
      "step": 49420
    },
    {
      "entropy": 5.775366687774659,
      "epoch": 4.319234466486061,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003261880004736972,
      "loss": 5.2294,
      "mean_token_accuracy": 0.204293891787529,
      "num_tokens": 100139566.0,
      "step": 49425
    },
    {
      "entropy": 5.781241416931152,
      "epoch": 4.319671414838766,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032615765600488435,
      "loss": 5.2863,
      "mean_token_accuracy": 0.21051884293556214,
      "num_tokens": 100150173.0,
      "step": 49430
    },
    {
      "entropy": 5.70802526473999,
      "epoch": 4.320108363191471,
      "grad_norm": 1.25,
      "learning_rate": 0.0003261273105548408,
      "loss": 5.1538,
      "mean_token_accuracy": 0.21303484439849854,
      "num_tokens": 100160923.0,
      "step": 49435
    },
    {
      "entropy": 5.766775512695313,
      "epoch": 4.320545311544175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003260969641241487,
      "loss": 5.2227,
      "mean_token_accuracy": 0.20458059757947922,
      "num_tokens": 100170682.0,
      "step": 49440
    },
    {
      "entropy": 5.6691216945648195,
      "epoch": 4.32098225989688,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032606661671339,
      "loss": 5.13,
      "mean_token_accuracy": 0.2172448605298996,
      "num_tokens": 100180901.0,
      "step": 49445
    },
    {
      "entropy": 5.711229467391968,
      "epoch": 4.321419208249585,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032603626832314694,
      "loss": 5.221,
      "mean_token_accuracy": 0.209789702296257,
      "num_tokens": 100190842.0,
      "step": 49450
    },
    {
      "entropy": 5.754915237426758,
      "epoch": 4.32185615660229,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003260059189540014,
      "loss": 5.1312,
      "mean_token_accuracy": 0.2107924610376358,
      "num_tokens": 100200324.0,
      "step": 49455
    },
    {
      "entropy": 5.691422748565674,
      "epoch": 4.322293104954994,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003259755686065356,
      "loss": 5.111,
      "mean_token_accuracy": 0.22463093549013138,
      "num_tokens": 100209977.0,
      "step": 49460
    },
    {
      "entropy": 5.668156957626342,
      "epoch": 4.322730053307699,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032594521728133177,
      "loss": 5.1749,
      "mean_token_accuracy": 0.20932372957468032,
      "num_tokens": 100219773.0,
      "step": 49465
    },
    {
      "entropy": 5.764641189575196,
      "epoch": 4.323167001660404,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000325914864978972,
      "loss": 5.2122,
      "mean_token_accuracy": 0.21088757812976838,
      "num_tokens": 100229089.0,
      "step": 49470
    },
    {
      "entropy": 5.855195379257202,
      "epoch": 4.3236039500131085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003258845117000384,
      "loss": 5.3523,
      "mean_token_accuracy": 0.20163746327161788,
      "num_tokens": 100239017.0,
      "step": 49475
    },
    {
      "entropy": 5.745127105712891,
      "epoch": 4.324040898365813,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003258541574451131,
      "loss": 5.1936,
      "mean_token_accuracy": 0.2086445853114128,
      "num_tokens": 100250056.0,
      "step": 49480
    },
    {
      "entropy": 5.707823801040649,
      "epoch": 4.324477846718517,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032582380221477853,
      "loss": 5.1986,
      "mean_token_accuracy": 0.2086143434047699,
      "num_tokens": 100260752.0,
      "step": 49485
    },
    {
      "entropy": 5.778440523147583,
      "epoch": 4.324914795071223,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032579344600961667,
      "loss": 5.2248,
      "mean_token_accuracy": 0.20828149765729903,
      "num_tokens": 100270379.0,
      "step": 49490
    },
    {
      "entropy": 5.763307762145996,
      "epoch": 4.325351743423927,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032576308883021013,
      "loss": 5.2341,
      "mean_token_accuracy": 0.2029222086071968,
      "num_tokens": 100280692.0,
      "step": 49495
    },
    {
      "entropy": 5.810465717315674,
      "epoch": 4.325788691776632,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032573273067714077,
      "loss": 5.2209,
      "mean_token_accuracy": 0.20580660104751586,
      "num_tokens": 100290726.0,
      "step": 49500
    },
    {
      "entropy": 5.8139355182647705,
      "epoch": 4.326225640129337,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032570237155099107,
      "loss": 5.2682,
      "mean_token_accuracy": 0.2117358297109604,
      "num_tokens": 100301080.0,
      "step": 49505
    },
    {
      "entropy": 5.6778570175170895,
      "epoch": 4.3266625884820415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003256720114523433,
      "loss": 5.1342,
      "mean_token_accuracy": 0.22400881350040436,
      "num_tokens": 100311412.0,
      "step": 49510
    },
    {
      "entropy": 5.703539657592773,
      "epoch": 4.327099536834746,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003256416503817799,
      "loss": 5.2234,
      "mean_token_accuracy": 0.20354543179273604,
      "num_tokens": 100321504.0,
      "step": 49515
    },
    {
      "entropy": 5.737063837051392,
      "epoch": 4.3275364851874505,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032561128833988307,
      "loss": 5.2145,
      "mean_token_accuracy": 0.20711242854595185,
      "num_tokens": 100332096.0,
      "step": 49520
    },
    {
      "entropy": 5.684759521484375,
      "epoch": 4.327973433540156,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032558092532723515,
      "loss": 5.2089,
      "mean_token_accuracy": 0.20826739370822905,
      "num_tokens": 100342544.0,
      "step": 49525
    },
    {
      "entropy": 5.748354482650757,
      "epoch": 4.32841038189286,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003255505613444186,
      "loss": 5.1795,
      "mean_token_accuracy": 0.20875155925750732,
      "num_tokens": 100351469.0,
      "step": 49530
    },
    {
      "entropy": 5.745532751083374,
      "epoch": 4.328847330245565,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003255201963920159,
      "loss": 5.2309,
      "mean_token_accuracy": 0.20597974210977554,
      "num_tokens": 100361378.0,
      "step": 49535
    },
    {
      "entropy": 5.78288049697876,
      "epoch": 4.329284278598269,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003254898304706092,
      "loss": 5.2874,
      "mean_token_accuracy": 0.20267123728990555,
      "num_tokens": 100371851.0,
      "step": 49540
    },
    {
      "entropy": 5.745867919921875,
      "epoch": 4.329721226950975,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003254594635807812,
      "loss": 5.2088,
      "mean_token_accuracy": 0.21193340420722961,
      "num_tokens": 100382422.0,
      "step": 49545
    },
    {
      "entropy": 5.7509276390075685,
      "epoch": 4.330158175303679,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032542909572311424,
      "loss": 5.3065,
      "mean_token_accuracy": 0.2081896498799324,
      "num_tokens": 100393181.0,
      "step": 49550
    },
    {
      "entropy": 5.7870440006256105,
      "epoch": 4.3305951236563835,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032539872689819075,
      "loss": 5.3119,
      "mean_token_accuracy": 0.20197782963514327,
      "num_tokens": 100403464.0,
      "step": 49555
    },
    {
      "entropy": 5.819881868362427,
      "epoch": 4.331032072009089,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032536835710659335,
      "loss": 5.257,
      "mean_token_accuracy": 0.20595579892396926,
      "num_tokens": 100413963.0,
      "step": 49560
    },
    {
      "entropy": 5.76946120262146,
      "epoch": 4.331469020361793,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003253379863489044,
      "loss": 5.1444,
      "mean_token_accuracy": 0.21729957014322282,
      "num_tokens": 100423364.0,
      "step": 49565
    },
    {
      "entropy": 5.695345878601074,
      "epoch": 4.331905968714498,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032530761462570644,
      "loss": 5.2028,
      "mean_token_accuracy": 0.21598340421915055,
      "num_tokens": 100432444.0,
      "step": 49570
    },
    {
      "entropy": 5.723438787460327,
      "epoch": 4.332342917067202,
      "grad_norm": 1.125,
      "learning_rate": 0.00032527724193758223,
      "loss": 5.2289,
      "mean_token_accuracy": 0.21149376183748245,
      "num_tokens": 100442840.0,
      "step": 49575
    },
    {
      "entropy": 5.823379850387573,
      "epoch": 4.332779865419908,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032524686828511407,
      "loss": 5.2283,
      "mean_token_accuracy": 0.20614594221115112,
      "num_tokens": 100452958.0,
      "step": 49580
    },
    {
      "entropy": 5.732685661315918,
      "epoch": 4.333216813772612,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032521649366888466,
      "loss": 5.1601,
      "mean_token_accuracy": 0.210765540599823,
      "num_tokens": 100463532.0,
      "step": 49585
    },
    {
      "entropy": 5.775433826446533,
      "epoch": 4.333653762125317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032518611808947655,
      "loss": 5.2497,
      "mean_token_accuracy": 0.2073057010769844,
      "num_tokens": 100474161.0,
      "step": 49590
    },
    {
      "entropy": 5.764243316650391,
      "epoch": 4.334090710478021,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032515574154747245,
      "loss": 5.3176,
      "mean_token_accuracy": 0.20241241306066513,
      "num_tokens": 100484078.0,
      "step": 49595
    },
    {
      "entropy": 5.786607360839843,
      "epoch": 4.334527658830726,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032512536404345494,
      "loss": 5.2665,
      "mean_token_accuracy": 0.2032202512025833,
      "num_tokens": 100493722.0,
      "step": 49600
    },
    {
      "entropy": 5.760560655593872,
      "epoch": 4.334964607183431,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003250949855780067,
      "loss": 5.2094,
      "mean_token_accuracy": 0.21600980907678605,
      "num_tokens": 100504671.0,
      "step": 49605
    },
    {
      "entropy": 5.7938941478729244,
      "epoch": 4.335401555536135,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032506460615171033,
      "loss": 5.2074,
      "mean_token_accuracy": 0.21423865854740143,
      "num_tokens": 100514452.0,
      "step": 49610
    },
    {
      "entropy": 5.6833271980285645,
      "epoch": 4.335838503888841,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003250342257651487,
      "loss": 5.1249,
      "mean_token_accuracy": 0.21578070521354675,
      "num_tokens": 100524339.0,
      "step": 49615
    },
    {
      "entropy": 5.710317468643188,
      "epoch": 4.336275452241545,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003250038444189043,
      "loss": 5.1564,
      "mean_token_accuracy": 0.22229590713977815,
      "num_tokens": 100534724.0,
      "step": 49620
    },
    {
      "entropy": 5.752397441864014,
      "epoch": 4.33671240059425,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032497346211356,
      "loss": 5.2067,
      "mean_token_accuracy": 0.21035714447498322,
      "num_tokens": 100545309.0,
      "step": 49625
    },
    {
      "entropy": 5.7482297897338865,
      "epoch": 4.337149348946954,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003249430788496984,
      "loss": 5.228,
      "mean_token_accuracy": 0.20807354301214218,
      "num_tokens": 100555315.0,
      "step": 49630
    },
    {
      "entropy": 5.721460151672363,
      "epoch": 4.3375862972996595,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032491269462790257,
      "loss": 5.2353,
      "mean_token_accuracy": 0.20223595649003984,
      "num_tokens": 100565624.0,
      "step": 49635
    },
    {
      "entropy": 5.765813398361206,
      "epoch": 4.338023245652364,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032488230944875495,
      "loss": 5.2202,
      "mean_token_accuracy": 0.21161216199398042,
      "num_tokens": 100576430.0,
      "step": 49640
    },
    {
      "entropy": 5.780316543579102,
      "epoch": 4.338460194005068,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003248519233128386,
      "loss": 5.1658,
      "mean_token_accuracy": 0.216297909617424,
      "num_tokens": 100586537.0,
      "step": 49645
    },
    {
      "entropy": 5.761633682250976,
      "epoch": 4.338897142357773,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003248215362207362,
      "loss": 5.2106,
      "mean_token_accuracy": 0.20801353752613067,
      "num_tokens": 100596694.0,
      "step": 49650
    },
    {
      "entropy": 5.8222935676574705,
      "epoch": 4.339334090710478,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003247911481730306,
      "loss": 5.3416,
      "mean_token_accuracy": 0.20391945093870162,
      "num_tokens": 100606711.0,
      "step": 49655
    },
    {
      "entropy": 5.74165301322937,
      "epoch": 4.339771039063183,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032476075917030475,
      "loss": 5.1826,
      "mean_token_accuracy": 0.20918028950691223,
      "num_tokens": 100616778.0,
      "step": 49660
    },
    {
      "entropy": 5.731914758682251,
      "epoch": 4.340207987415887,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032473036921314136,
      "loss": 5.2176,
      "mean_token_accuracy": 0.21181367933750153,
      "num_tokens": 100627009.0,
      "step": 49665
    },
    {
      "entropy": 5.760243511199951,
      "epoch": 4.340644935768593,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003246999783021235,
      "loss": 5.227,
      "mean_token_accuracy": 0.2096664324402809,
      "num_tokens": 100636735.0,
      "step": 49670
    },
    {
      "entropy": 5.780793476104736,
      "epoch": 4.341081884121297,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032466958643783404,
      "loss": 5.1994,
      "mean_token_accuracy": 0.2123490408062935,
      "num_tokens": 100646984.0,
      "step": 49675
    },
    {
      "entropy": 5.75465030670166,
      "epoch": 4.3415188324740015,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032463919362085583,
      "loss": 5.2575,
      "mean_token_accuracy": 0.20510323494672775,
      "num_tokens": 100656854.0,
      "step": 49680
    },
    {
      "entropy": 5.766635084152222,
      "epoch": 4.341955780826706,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003246087998517719,
      "loss": 5.2509,
      "mean_token_accuracy": 0.2030278518795967,
      "num_tokens": 100666660.0,
      "step": 49685
    },
    {
      "entropy": 5.769235563278198,
      "epoch": 4.342392729179411,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003245784051311651,
      "loss": 5.2425,
      "mean_token_accuracy": 0.20728524774312973,
      "num_tokens": 100677451.0,
      "step": 49690
    },
    {
      "entropy": 5.7798041820526125,
      "epoch": 4.342829677532116,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032454800945961866,
      "loss": 5.2319,
      "mean_token_accuracy": 0.203826405107975,
      "num_tokens": 100687889.0,
      "step": 49695
    },
    {
      "entropy": 5.710021686553955,
      "epoch": 4.34326662588482,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003245176128377154,
      "loss": 5.1842,
      "mean_token_accuracy": 0.21276461631059645,
      "num_tokens": 100697395.0,
      "step": 49700
    },
    {
      "entropy": 5.6687087059021,
      "epoch": 4.343703574237525,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003244872152660384,
      "loss": 5.1214,
      "mean_token_accuracy": 0.21244161128997802,
      "num_tokens": 100707712.0,
      "step": 49705
    },
    {
      "entropy": 5.758753108978271,
      "epoch": 4.34414052259023,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032445681674517056,
      "loss": 5.2259,
      "mean_token_accuracy": 0.2083490177989006,
      "num_tokens": 100717979.0,
      "step": 49710
    },
    {
      "entropy": 5.68388032913208,
      "epoch": 4.344577470942935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032442641727569517,
      "loss": 5.1124,
      "mean_token_accuracy": 0.22178292721509935,
      "num_tokens": 100727046.0,
      "step": 49715
    },
    {
      "entropy": 5.707447624206543,
      "epoch": 4.345014419295639,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003243960168581951,
      "loss": 5.1939,
      "mean_token_accuracy": 0.2105154201388359,
      "num_tokens": 100736645.0,
      "step": 49720
    },
    {
      "entropy": 5.667775821685791,
      "epoch": 4.345451367648344,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032436561549325366,
      "loss": 5.1412,
      "mean_token_accuracy": 0.21652900725603103,
      "num_tokens": 100746523.0,
      "step": 49725
    },
    {
      "entropy": 5.777493190765381,
      "epoch": 4.345888316001049,
      "grad_norm": 1.125,
      "learning_rate": 0.0003243352131814538,
      "loss": 5.272,
      "mean_token_accuracy": 0.20323044508695604,
      "num_tokens": 100756122.0,
      "step": 49730
    },
    {
      "entropy": 5.732828044891358,
      "epoch": 4.346325264353753,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032430480992337867,
      "loss": 5.1999,
      "mean_token_accuracy": 0.2117069035768509,
      "num_tokens": 100766665.0,
      "step": 49735
    },
    {
      "entropy": 5.760940456390381,
      "epoch": 4.346762212706458,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003242744057196115,
      "loss": 5.211,
      "mean_token_accuracy": 0.20552070587873458,
      "num_tokens": 100775759.0,
      "step": 49740
    },
    {
      "entropy": 5.682236528396606,
      "epoch": 4.347199161059163,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00032424400057073536,
      "loss": 5.2274,
      "mean_token_accuracy": 0.21534419506788255,
      "num_tokens": 100785988.0,
      "step": 49745
    },
    {
      "entropy": 5.700650787353515,
      "epoch": 4.347636109411868,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003242135944773335,
      "loss": 5.1161,
      "mean_token_accuracy": 0.21519820243120194,
      "num_tokens": 100794352.0,
      "step": 49750
    },
    {
      "entropy": 5.664819383621216,
      "epoch": 4.348073057764572,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032418318743998917,
      "loss": 5.1309,
      "mean_token_accuracy": 0.22021485716104508,
      "num_tokens": 100804146.0,
      "step": 49755
    },
    {
      "entropy": 5.7168800830841064,
      "epoch": 4.348510006117277,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003241527794592855,
      "loss": 5.2037,
      "mean_token_accuracy": 0.20544549375772475,
      "num_tokens": 100815422.0,
      "step": 49760
    },
    {
      "entropy": 5.819243049621582,
      "epoch": 4.348946954469982,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032412237053580574,
      "loss": 5.3258,
      "mean_token_accuracy": 0.19671052247285842,
      "num_tokens": 100826104.0,
      "step": 49765
    },
    {
      "entropy": 5.741803598403931,
      "epoch": 4.349383902822686,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003240919606701331,
      "loss": 5.1357,
      "mean_token_accuracy": 0.2161615625023842,
      "num_tokens": 100835175.0,
      "step": 49770
    },
    {
      "entropy": 5.717784738540649,
      "epoch": 4.349820851175391,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00032406154986285116,
      "loss": 5.2049,
      "mean_token_accuracy": 0.2192624941468239,
      "num_tokens": 100845724.0,
      "step": 49775
    },
    {
      "entropy": 5.716913557052612,
      "epoch": 4.350257799528096,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032403113811454277,
      "loss": 5.2171,
      "mean_token_accuracy": 0.2115801677107811,
      "num_tokens": 100855902.0,
      "step": 49780
    },
    {
      "entropy": 5.7316694259643555,
      "epoch": 4.350694747880801,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003240007254257915,
      "loss": 5.1985,
      "mean_token_accuracy": 0.214680053293705,
      "num_tokens": 100866368.0,
      "step": 49785
    },
    {
      "entropy": 5.783724355697632,
      "epoch": 4.351131696233505,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032397031179718066,
      "loss": 5.1918,
      "mean_token_accuracy": 0.21174630969762803,
      "num_tokens": 100875631.0,
      "step": 49790
    },
    {
      "entropy": 5.74842586517334,
      "epoch": 4.35156864458621,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032393989722929355,
      "loss": 5.1722,
      "mean_token_accuracy": 0.21906437128782272,
      "num_tokens": 100885237.0,
      "step": 49795
    },
    {
      "entropy": 5.6688018321990965,
      "epoch": 4.352005592938915,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003239094817227137,
      "loss": 5.2311,
      "mean_token_accuracy": 0.21108331084251403,
      "num_tokens": 100895097.0,
      "step": 49800
    },
    {
      "entropy": 5.700389385223389,
      "epoch": 4.3524425412916194,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032387906527802426,
      "loss": 5.1636,
      "mean_token_accuracy": 0.21156719326972961,
      "num_tokens": 100904275.0,
      "step": 49805
    },
    {
      "entropy": 5.73117299079895,
      "epoch": 4.352879489644324,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003238486478958087,
      "loss": 5.1465,
      "mean_token_accuracy": 0.2158484309911728,
      "num_tokens": 100913950.0,
      "step": 49810
    },
    {
      "entropy": 5.651483631134033,
      "epoch": 4.353316437997028,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003238182295766505,
      "loss": 5.169,
      "mean_token_accuracy": 0.21695850044488907,
      "num_tokens": 100924446.0,
      "step": 49815
    },
    {
      "entropy": 5.727306699752807,
      "epoch": 4.353753386349734,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032378781032113313,
      "loss": 5.136,
      "mean_token_accuracy": 0.21769683957099914,
      "num_tokens": 100933717.0,
      "step": 49820
    },
    {
      "entropy": 5.739152956008911,
      "epoch": 4.354190334702438,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032375739012984004,
      "loss": 5.1722,
      "mean_token_accuracy": 0.2144845798611641,
      "num_tokens": 100943800.0,
      "step": 49825
    },
    {
      "entropy": 5.738648414611816,
      "epoch": 4.354627283055143,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032372696900335454,
      "loss": 5.2256,
      "mean_token_accuracy": 0.20081020146608353,
      "num_tokens": 100954733.0,
      "step": 49830
    },
    {
      "entropy": 5.759275341033936,
      "epoch": 4.355064231407848,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032369654694226025,
      "loss": 5.2576,
      "mean_token_accuracy": 0.2006338894367218,
      "num_tokens": 100964536.0,
      "step": 49835
    },
    {
      "entropy": 5.747474670410156,
      "epoch": 4.3555011797605525,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032366612394714065,
      "loss": 5.2913,
      "mean_token_accuracy": 0.20375409573316575,
      "num_tokens": 100975650.0,
      "step": 49840
    },
    {
      "entropy": 5.786600923538208,
      "epoch": 4.355938128113257,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032363570001857947,
      "loss": 5.2307,
      "mean_token_accuracy": 0.204900860786438,
      "num_tokens": 100986439.0,
      "step": 49845
    },
    {
      "entropy": 5.74268274307251,
      "epoch": 4.3563750764659614,
      "grad_norm": 1.125,
      "learning_rate": 0.00032360527515715994,
      "loss": 5.2386,
      "mean_token_accuracy": 0.20303076654672622,
      "num_tokens": 100996593.0,
      "step": 49850
    },
    {
      "entropy": 5.7360546588897705,
      "epoch": 4.356812024818667,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032357484936346577,
      "loss": 5.2253,
      "mean_token_accuracy": 0.21145318299531937,
      "num_tokens": 101007614.0,
      "step": 49855
    },
    {
      "entropy": 5.759788227081299,
      "epoch": 4.357248973171371,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032354442263808056,
      "loss": 5.2683,
      "mean_token_accuracy": 0.20549824386835097,
      "num_tokens": 101018234.0,
      "step": 49860
    },
    {
      "entropy": 5.715687990188599,
      "epoch": 4.357685921524076,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003235139949815878,
      "loss": 5.1769,
      "mean_token_accuracy": 0.21793080270290374,
      "num_tokens": 101028001.0,
      "step": 49865
    },
    {
      "entropy": 5.630558824539184,
      "epoch": 4.35812286987678,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003234835663945713,
      "loss": 5.1447,
      "mean_token_accuracy": 0.2156265690922737,
      "num_tokens": 101037857.0,
      "step": 49870
    },
    {
      "entropy": 5.784075212478638,
      "epoch": 4.358559818229486,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003234531368776145,
      "loss": 5.1751,
      "mean_token_accuracy": 0.20915285050868987,
      "num_tokens": 101047224.0,
      "step": 49875
    },
    {
      "entropy": 5.77028169631958,
      "epoch": 4.35899676658219,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032342270643130123,
      "loss": 5.1936,
      "mean_token_accuracy": 0.21123697012662887,
      "num_tokens": 101057124.0,
      "step": 49880
    },
    {
      "entropy": 5.729512643814087,
      "epoch": 4.3594337149348945,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032339227505621504,
      "loss": 5.2327,
      "mean_token_accuracy": 0.2028108596801758,
      "num_tokens": 101066712.0,
      "step": 49885
    },
    {
      "entropy": 5.775117254257202,
      "epoch": 4.3598706632876,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032336184275293974,
      "loss": 5.2623,
      "mean_token_accuracy": 0.2026071384549141,
      "num_tokens": 101076787.0,
      "step": 49890
    },
    {
      "entropy": 5.775557374954223,
      "epoch": 4.360307611640304,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032333140952205884,
      "loss": 5.1676,
      "mean_token_accuracy": 0.21687602400779724,
      "num_tokens": 101087525.0,
      "step": 49895
    },
    {
      "entropy": 5.765124797821045,
      "epoch": 4.360744559993009,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003233009753641562,
      "loss": 5.2076,
      "mean_token_accuracy": 0.2062956854701042,
      "num_tokens": 101097114.0,
      "step": 49900
    },
    {
      "entropy": 5.698634481430053,
      "epoch": 4.361181508345713,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003232705402798156,
      "loss": 5.2045,
      "mean_token_accuracy": 0.2146714672446251,
      "num_tokens": 101106732.0,
      "step": 49905
    },
    {
      "entropy": 5.734081983566284,
      "epoch": 4.361618456698419,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003232401042696207,
      "loss": 5.2342,
      "mean_token_accuracy": 0.2083747610449791,
      "num_tokens": 101116639.0,
      "step": 49910
    },
    {
      "entropy": 5.733810091018677,
      "epoch": 4.362055405051123,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032320966733415536,
      "loss": 5.2692,
      "mean_token_accuracy": 0.21287218928337098,
      "num_tokens": 101126716.0,
      "step": 49915
    },
    {
      "entropy": 5.758394336700439,
      "epoch": 4.362492353403828,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003231792294740033,
      "loss": 5.1515,
      "mean_token_accuracy": 0.21354936808347702,
      "num_tokens": 101137362.0,
      "step": 49920
    },
    {
      "entropy": 5.748951816558838,
      "epoch": 4.362929301756532,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032314879068974837,
      "loss": 5.254,
      "mean_token_accuracy": 0.21018072962760925,
      "num_tokens": 101147437.0,
      "step": 49925
    },
    {
      "entropy": 5.777267742156982,
      "epoch": 4.363366250109237,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003231183509819744,
      "loss": 5.3024,
      "mean_token_accuracy": 0.2028139278292656,
      "num_tokens": 101157664.0,
      "step": 49930
    },
    {
      "entropy": 5.761725568771363,
      "epoch": 4.363803198461942,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003230879103512653,
      "loss": 5.2426,
      "mean_token_accuracy": 0.20776804238557817,
      "num_tokens": 101168996.0,
      "step": 49935
    },
    {
      "entropy": 5.736214351654053,
      "epoch": 4.364240146814646,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032305746879820487,
      "loss": 5.2525,
      "mean_token_accuracy": 0.20237679779529572,
      "num_tokens": 101179089.0,
      "step": 49940
    },
    {
      "entropy": 5.809031820297241,
      "epoch": 4.364677095167351,
      "grad_norm": 1.046875,
      "learning_rate": 0.000323027026323377,
      "loss": 5.2701,
      "mean_token_accuracy": 0.21100465208292007,
      "num_tokens": 101189874.0,
      "step": 49945
    },
    {
      "entropy": 5.765937328338623,
      "epoch": 4.365114043520056,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032299658292736554,
      "loss": 5.2222,
      "mean_token_accuracy": 0.204480841755867,
      "num_tokens": 101199715.0,
      "step": 49950
    },
    {
      "entropy": 5.742356300354004,
      "epoch": 4.365550991872761,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032296613861075454,
      "loss": 5.2316,
      "mean_token_accuracy": 0.2069638639688492,
      "num_tokens": 101209996.0,
      "step": 49955
    },
    {
      "entropy": 5.739048337936401,
      "epoch": 4.365987940225465,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003229356933741278,
      "loss": 5.2517,
      "mean_token_accuracy": 0.21071085780858995,
      "num_tokens": 101220403.0,
      "step": 49960
    },
    {
      "entropy": 5.7416259288787845,
      "epoch": 4.3664248885781705,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032290524721806943,
      "loss": 5.1968,
      "mean_token_accuracy": 0.20978244692087172,
      "num_tokens": 101230490.0,
      "step": 49965
    },
    {
      "entropy": 5.757237815856934,
      "epoch": 4.366861836930875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003228748001431633,
      "loss": 5.2132,
      "mean_token_accuracy": 0.21025692522525788,
      "num_tokens": 101240817.0,
      "step": 49970
    },
    {
      "entropy": 5.7699141025543215,
      "epoch": 4.367298785283579,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003228443521499934,
      "loss": 5.3035,
      "mean_token_accuracy": 0.2056013375520706,
      "num_tokens": 101251893.0,
      "step": 49975
    },
    {
      "entropy": 5.795415782928467,
      "epoch": 4.367735733636284,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032281390323914366,
      "loss": 5.2077,
      "mean_token_accuracy": 0.20181504637002945,
      "num_tokens": 101262241.0,
      "step": 49980
    },
    {
      "entropy": 5.678329849243164,
      "epoch": 4.368172681988989,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003227834534111983,
      "loss": 5.2406,
      "mean_token_accuracy": 0.20888499468564986,
      "num_tokens": 101272285.0,
      "step": 49985
    },
    {
      "entropy": 5.769741868972778,
      "epoch": 4.368609630341694,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003227530026667413,
      "loss": 5.3126,
      "mean_token_accuracy": 0.20331162363290786,
      "num_tokens": 101283026.0,
      "step": 49990
    },
    {
      "entropy": 5.791450023651123,
      "epoch": 4.369046578694398,
      "grad_norm": 1.125,
      "learning_rate": 0.0003227225510063566,
      "loss": 5.2604,
      "mean_token_accuracy": 0.21330829858779907,
      "num_tokens": 101293376.0,
      "step": 49995
    },
    {
      "entropy": 5.80920147895813,
      "epoch": 4.369483527047103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003226920984306284,
      "loss": 5.2288,
      "mean_token_accuracy": 0.21215232014656066,
      "num_tokens": 101305232.0,
      "step": 50000
    },
    {
      "entropy": 5.799699163436889,
      "epoch": 4.369920475399808,
      "grad_norm": 1.125,
      "learning_rate": 0.00032266164494014075,
      "loss": 5.2661,
      "mean_token_accuracy": 0.20805153399705886,
      "num_tokens": 101314942.0,
      "step": 50005
    },
    {
      "entropy": 5.7464405536651615,
      "epoch": 4.3703574237525125,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032263119053547784,
      "loss": 5.1798,
      "mean_token_accuracy": 0.21059973984956742,
      "num_tokens": 101323655.0,
      "step": 50010
    },
    {
      "entropy": 5.698749208450318,
      "epoch": 4.370794372105217,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003226007352172237,
      "loss": 5.1048,
      "mean_token_accuracy": 0.21171242296695708,
      "num_tokens": 101333594.0,
      "step": 50015
    },
    {
      "entropy": 5.691592979431152,
      "epoch": 4.371231320457922,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003225702789859625,
      "loss": 5.2206,
      "mean_token_accuracy": 0.20966424345970153,
      "num_tokens": 101344580.0,
      "step": 50020
    },
    {
      "entropy": 5.693030500411988,
      "epoch": 4.371668268810627,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032253982184227836,
      "loss": 5.1704,
      "mean_token_accuracy": 0.21265850216150284,
      "num_tokens": 101354574.0,
      "step": 50025
    },
    {
      "entropy": 5.725378370285034,
      "epoch": 4.372105217163331,
      "grad_norm": 1.125,
      "learning_rate": 0.0003225093637867556,
      "loss": 5.1546,
      "mean_token_accuracy": 0.21449111253023148,
      "num_tokens": 101365150.0,
      "step": 50030
    },
    {
      "entropy": 5.782832384109497,
      "epoch": 4.372542165516036,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003224789048199784,
      "loss": 5.2234,
      "mean_token_accuracy": 0.21234821379184723,
      "num_tokens": 101375726.0,
      "step": 50035
    },
    {
      "entropy": 5.8167143821716305,
      "epoch": 4.372979113868741,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003224484449425308,
      "loss": 5.3021,
      "mean_token_accuracy": 0.20311434268951417,
      "num_tokens": 101385457.0,
      "step": 50040
    },
    {
      "entropy": 5.804802274703979,
      "epoch": 4.3734160622214455,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003224179841549972,
      "loss": 5.275,
      "mean_token_accuracy": 0.2054639235138893,
      "num_tokens": 101395513.0,
      "step": 50045
    },
    {
      "entropy": 5.732926607131958,
      "epoch": 4.37385301057415,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032238752245796186,
      "loss": 5.1758,
      "mean_token_accuracy": 0.21032065004110337,
      "num_tokens": 101405886.0,
      "step": 50050
    },
    {
      "entropy": 5.699498081207276,
      "epoch": 4.3742899589268545,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032235705985200904,
      "loss": 5.2257,
      "mean_token_accuracy": 0.21411301642656327,
      "num_tokens": 101416130.0,
      "step": 50055
    },
    {
      "entropy": 5.70190281867981,
      "epoch": 4.37472690727956,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003223265963377229,
      "loss": 5.1841,
      "mean_token_accuracy": 0.21968593448400497,
      "num_tokens": 101426817.0,
      "step": 50060
    },
    {
      "entropy": 5.785248279571533,
      "epoch": 4.375163855632264,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003222961319156879,
      "loss": 5.2358,
      "mean_token_accuracy": 0.2111579954624176,
      "num_tokens": 101436873.0,
      "step": 50065
    },
    {
      "entropy": 5.819790983200074,
      "epoch": 4.375600803984969,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003222656665864884,
      "loss": 5.247,
      "mean_token_accuracy": 0.2140297219157219,
      "num_tokens": 101447048.0,
      "step": 50070
    },
    {
      "entropy": 5.820613813400269,
      "epoch": 4.376037752337673,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032223520035070844,
      "loss": 5.2709,
      "mean_token_accuracy": 0.20518235862255096,
      "num_tokens": 101457536.0,
      "step": 50075
    },
    {
      "entropy": 5.756344890594482,
      "epoch": 4.376474700690379,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003222047332089328,
      "loss": 5.2323,
      "mean_token_accuracy": 0.2103796347975731,
      "num_tokens": 101467833.0,
      "step": 50080
    },
    {
      "entropy": 5.768861484527588,
      "epoch": 4.376911649043083,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032217426516174553,
      "loss": 5.2784,
      "mean_token_accuracy": 0.20698014050722122,
      "num_tokens": 101477960.0,
      "step": 50085
    },
    {
      "entropy": 5.780249977111817,
      "epoch": 4.3773485973957875,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032214379620973117,
      "loss": 5.1996,
      "mean_token_accuracy": 0.21112946122884751,
      "num_tokens": 101487192.0,
      "step": 50090
    },
    {
      "entropy": 5.72087345123291,
      "epoch": 4.377785545748493,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032211332635347404,
      "loss": 5.1647,
      "mean_token_accuracy": 0.21357913166284562,
      "num_tokens": 101496835.0,
      "step": 50095
    },
    {
      "entropy": 5.6931713104248045,
      "epoch": 4.378222494101197,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032208285559355863,
      "loss": 5.1764,
      "mean_token_accuracy": 0.2105598509311676,
      "num_tokens": 101506584.0,
      "step": 50100
    },
    {
      "entropy": 5.787537336349487,
      "epoch": 4.378659442453902,
      "grad_norm": 0.984375,
      "learning_rate": 0.00032205238393056944,
      "loss": 5.2923,
      "mean_token_accuracy": 0.20649295002222062,
      "num_tokens": 101517428.0,
      "step": 50105
    },
    {
      "entropy": 5.743819284439087,
      "epoch": 4.379096390806606,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032202191136509084,
      "loss": 5.2116,
      "mean_token_accuracy": 0.2078921005129814,
      "num_tokens": 101527594.0,
      "step": 50110
    },
    {
      "entropy": 5.797280120849609,
      "epoch": 4.379533339159312,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003219914378977073,
      "loss": 5.2394,
      "mean_token_accuracy": 0.2076169118285179,
      "num_tokens": 101538381.0,
      "step": 50115
    },
    {
      "entropy": 5.707452344894409,
      "epoch": 4.379970287512016,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032196096352900337,
      "loss": 5.1916,
      "mean_token_accuracy": 0.21686748415231705,
      "num_tokens": 101548110.0,
      "step": 50120
    },
    {
      "entropy": 5.734598684310913,
      "epoch": 4.380407235864721,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032193048825956353,
      "loss": 5.1752,
      "mean_token_accuracy": 0.2175902709364891,
      "num_tokens": 101556966.0,
      "step": 50125
    },
    {
      "entropy": 5.7063699722290036,
      "epoch": 4.380844184217425,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032190001208997254,
      "loss": 5.1011,
      "mean_token_accuracy": 0.21939035654067993,
      "num_tokens": 101567310.0,
      "step": 50130
    },
    {
      "entropy": 5.7636312484741214,
      "epoch": 4.38128113257013,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032186953502081456,
      "loss": 5.2324,
      "mean_token_accuracy": 0.20452145040035247,
      "num_tokens": 101577546.0,
      "step": 50135
    },
    {
      "entropy": 5.715137529373169,
      "epoch": 4.381718080922835,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032183905705267436,
      "loss": 5.1916,
      "mean_token_accuracy": 0.21138621717691422,
      "num_tokens": 101587856.0,
      "step": 50140
    },
    {
      "entropy": 5.736137199401855,
      "epoch": 4.382155029275539,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003218085781861365,
      "loss": 5.2347,
      "mean_token_accuracy": 0.20346134901046753,
      "num_tokens": 101597587.0,
      "step": 50145
    },
    {
      "entropy": 5.752693843841553,
      "epoch": 4.382591977628245,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003217780984217856,
      "loss": 5.245,
      "mean_token_accuracy": 0.2166620522737503,
      "num_tokens": 101608239.0,
      "step": 50150
    },
    {
      "entropy": 5.753950119018555,
      "epoch": 4.383028925980949,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003217476177602062,
      "loss": 5.1929,
      "mean_token_accuracy": 0.21476748138666152,
      "num_tokens": 101618400.0,
      "step": 50155
    },
    {
      "entropy": 5.701413488388061,
      "epoch": 4.383465874333654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032171713620198313,
      "loss": 5.1304,
      "mean_token_accuracy": 0.21834678053855897,
      "num_tokens": 101628434.0,
      "step": 50160
    },
    {
      "entropy": 5.693453550338745,
      "epoch": 4.383902822686358,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003216866537477008,
      "loss": 5.1472,
      "mean_token_accuracy": 0.21880542784929274,
      "num_tokens": 101638209.0,
      "step": 50165
    },
    {
      "entropy": 5.829857349395752,
      "epoch": 4.3843397710390635,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000321656170397944,
      "loss": 5.3541,
      "mean_token_accuracy": 0.19859227240085603,
      "num_tokens": 101649162.0,
      "step": 50170
    },
    {
      "entropy": 5.803869771957397,
      "epoch": 4.384776719391768,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003216256861532974,
      "loss": 5.2304,
      "mean_token_accuracy": 0.20665827095508577,
      "num_tokens": 101658890.0,
      "step": 50175
    },
    {
      "entropy": 5.734537553787232,
      "epoch": 4.385213667744472,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00032159520101434567,
      "loss": 5.1748,
      "mean_token_accuracy": 0.21417591571807862,
      "num_tokens": 101670160.0,
      "step": 50180
    },
    {
      "entropy": 5.686343193054199,
      "epoch": 4.385650616097177,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003215647149816737,
      "loss": 5.1964,
      "mean_token_accuracy": 0.20795079916715623,
      "num_tokens": 101680025.0,
      "step": 50185
    },
    {
      "entropy": 5.765822839736939,
      "epoch": 4.386087564449882,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000321534228055866,
      "loss": 5.2611,
      "mean_token_accuracy": 0.20562591254711152,
      "num_tokens": 101690662.0,
      "step": 50190
    },
    {
      "entropy": 5.751938343048096,
      "epoch": 4.386524512802587,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032150374023750747,
      "loss": 5.1858,
      "mean_token_accuracy": 0.20801415741443635,
      "num_tokens": 101701605.0,
      "step": 50195
    },
    {
      "entropy": 5.777870607376099,
      "epoch": 4.386961461155291,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003214732515271828,
      "loss": 5.1835,
      "mean_token_accuracy": 0.2088839590549469,
      "num_tokens": 101711442.0,
      "step": 50200
    },
    {
      "entropy": 5.796543073654175,
      "epoch": 4.3873984095079965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032144276192547683,
      "loss": 5.2028,
      "mean_token_accuracy": 0.2158300384879112,
      "num_tokens": 101721718.0,
      "step": 50205
    },
    {
      "entropy": 5.709700012207032,
      "epoch": 4.387835357860701,
      "grad_norm": 1.125,
      "learning_rate": 0.00032141227143297433,
      "loss": 5.2106,
      "mean_token_accuracy": 0.20810741186141968,
      "num_tokens": 101731603.0,
      "step": 50210
    },
    {
      "entropy": 5.735689687728882,
      "epoch": 4.3882723062134055,
      "grad_norm": 1.125,
      "learning_rate": 0.0003213817800502601,
      "loss": 5.2081,
      "mean_token_accuracy": 0.20719740837812423,
      "num_tokens": 101742369.0,
      "step": 50215
    },
    {
      "entropy": 5.738127565383911,
      "epoch": 4.38870925456611,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003213512877779191,
      "loss": 5.2207,
      "mean_token_accuracy": 0.21089177429676056,
      "num_tokens": 101752893.0,
      "step": 50220
    },
    {
      "entropy": 5.702201509475708,
      "epoch": 4.389146202918815,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003213207946165361,
      "loss": 5.1472,
      "mean_token_accuracy": 0.20842067003250123,
      "num_tokens": 101763090.0,
      "step": 50225
    },
    {
      "entropy": 5.74901065826416,
      "epoch": 4.38958315127152,
      "grad_norm": 1.078125,
      "learning_rate": 0.000321290300566696,
      "loss": 5.1706,
      "mean_token_accuracy": 0.2146821841597557,
      "num_tokens": 101772052.0,
      "step": 50230
    },
    {
      "entropy": 5.744333124160766,
      "epoch": 4.390020099624224,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003212598056289836,
      "loss": 5.1563,
      "mean_token_accuracy": 0.21463514417409896,
      "num_tokens": 101782758.0,
      "step": 50235
    },
    {
      "entropy": 5.737517547607422,
      "epoch": 4.390457047976929,
      "grad_norm": 1.078125,
      "learning_rate": 0.000321229309803984,
      "loss": 5.1588,
      "mean_token_accuracy": 0.2180195540189743,
      "num_tokens": 101791681.0,
      "step": 50240
    },
    {
      "entropy": 5.739202404022217,
      "epoch": 4.390893996329634,
      "grad_norm": 0.9375,
      "learning_rate": 0.00032119881309228195,
      "loss": 5.2502,
      "mean_token_accuracy": 0.20922525227069855,
      "num_tokens": 101803446.0,
      "step": 50245
    },
    {
      "entropy": 5.7621642589569095,
      "epoch": 4.3913309446823385,
      "grad_norm": 1.125,
      "learning_rate": 0.00032116831549446246,
      "loss": 5.2479,
      "mean_token_accuracy": 0.2068438559770584,
      "num_tokens": 101813907.0,
      "step": 50250
    },
    {
      "entropy": 5.669008445739746,
      "epoch": 4.391767893035043,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003211378170111105,
      "loss": 5.1534,
      "mean_token_accuracy": 0.21962470561265945,
      "num_tokens": 101823715.0,
      "step": 50255
    },
    {
      "entropy": 5.733395433425903,
      "epoch": 4.392204841387748,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032110731764281103,
      "loss": 5.2327,
      "mean_token_accuracy": 0.21020950078964235,
      "num_tokens": 101834208.0,
      "step": 50260
    },
    {
      "entropy": 5.729732322692871,
      "epoch": 4.392641789740453,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032107681739014906,
      "loss": 5.1201,
      "mean_token_accuracy": 0.2197648376226425,
      "num_tokens": 101845354.0,
      "step": 50265
    },
    {
      "entropy": 5.81764178276062,
      "epoch": 4.393078738093157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003210463162537096,
      "loss": 5.2552,
      "mean_token_accuracy": 0.20510135740041732,
      "num_tokens": 101855818.0,
      "step": 50270
    },
    {
      "entropy": 5.741311883926391,
      "epoch": 4.393515686445862,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032101581423407764,
      "loss": 5.226,
      "mean_token_accuracy": 0.2142959102988243,
      "num_tokens": 101866012.0,
      "step": 50275
    },
    {
      "entropy": 5.7487053871154785,
      "epoch": 4.393952634798567,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032098531133183827,
      "loss": 5.1902,
      "mean_token_accuracy": 0.2046777918934822,
      "num_tokens": 101875834.0,
      "step": 50280
    },
    {
      "entropy": 5.75966968536377,
      "epoch": 4.394389583151272,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003209548075475765,
      "loss": 5.2301,
      "mean_token_accuracy": 0.21239711195230485,
      "num_tokens": 101885261.0,
      "step": 50285
    },
    {
      "entropy": 5.716061449050903,
      "epoch": 4.394826531503976,
      "grad_norm": 1.125,
      "learning_rate": 0.00032092430288187745,
      "loss": 5.2067,
      "mean_token_accuracy": 0.21003870964050292,
      "num_tokens": 101896290.0,
      "step": 50290
    },
    {
      "entropy": 5.757887458801269,
      "epoch": 4.3952634798566805,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032089379733532623,
      "loss": 5.2266,
      "mean_token_accuracy": 0.21029955595731736,
      "num_tokens": 101906830.0,
      "step": 50295
    },
    {
      "entropy": 5.78879680633545,
      "epoch": 4.395700428209386,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003208632909085079,
      "loss": 5.1989,
      "mean_token_accuracy": 0.21287311166524886,
      "num_tokens": 101917146.0,
      "step": 50300
    },
    {
      "entropy": 5.761564588546753,
      "epoch": 4.39613737656209,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003208327836020076,
      "loss": 5.2348,
      "mean_token_accuracy": 0.20990998148918152,
      "num_tokens": 101927326.0,
      "step": 50305
    },
    {
      "entropy": 5.800827693939209,
      "epoch": 4.396574324914795,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003208022754164106,
      "loss": 5.2978,
      "mean_token_accuracy": 0.20417685210704803,
      "num_tokens": 101938333.0,
      "step": 50310
    },
    {
      "entropy": 5.755375528335572,
      "epoch": 4.3970112732675,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003207717663523019,
      "loss": 5.1675,
      "mean_token_accuracy": 0.21544222682714462,
      "num_tokens": 101948847.0,
      "step": 50315
    },
    {
      "entropy": 5.762104129791259,
      "epoch": 4.397448221620205,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032074125641026676,
      "loss": 5.2692,
      "mean_token_accuracy": 0.2046595498919487,
      "num_tokens": 101959887.0,
      "step": 50320
    },
    {
      "entropy": 5.791827821731568,
      "epoch": 4.397885169972909,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003207107455908902,
      "loss": 5.2045,
      "mean_token_accuracy": 0.2082943394780159,
      "num_tokens": 101969336.0,
      "step": 50325
    },
    {
      "entropy": 5.777326250076294,
      "epoch": 4.398322118325614,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003206802338947578,
      "loss": 5.2811,
      "mean_token_accuracy": 0.20461440831422806,
      "num_tokens": 101979884.0,
      "step": 50330
    },
    {
      "entropy": 5.776006555557251,
      "epoch": 4.398759066678319,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003206497213224544,
      "loss": 5.2704,
      "mean_token_accuracy": 0.20897177308797837,
      "num_tokens": 101989803.0,
      "step": 50335
    },
    {
      "entropy": 5.768719673156738,
      "epoch": 4.399196015031023,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032061920787456556,
      "loss": 5.214,
      "mean_token_accuracy": 0.20931021124124527,
      "num_tokens": 101999519.0,
      "step": 50340
    },
    {
      "entropy": 5.749287748336792,
      "epoch": 4.399632963383728,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032058869355167634,
      "loss": 5.2229,
      "mean_token_accuracy": 0.21144922971725463,
      "num_tokens": 102009017.0,
      "step": 50345
    },
    {
      "entropy": 5.766206121444702,
      "epoch": 4.400069911736432,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003205581783543721,
      "loss": 5.2341,
      "mean_token_accuracy": 0.2049313098192215,
      "num_tokens": 102019357.0,
      "step": 50350
    },
    {
      "entropy": 5.762083244323731,
      "epoch": 4.400506860089138,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032052766228323815,
      "loss": 5.1818,
      "mean_token_accuracy": 0.21203929632902146,
      "num_tokens": 102029994.0,
      "step": 50355
    },
    {
      "entropy": 5.6862774848937985,
      "epoch": 4.400943808441842,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032049714533885975,
      "loss": 5.1394,
      "mean_token_accuracy": 0.2114195078611374,
      "num_tokens": 102039314.0,
      "step": 50360
    },
    {
      "entropy": 5.736450338363648,
      "epoch": 4.401380756794547,
      "grad_norm": 1.25,
      "learning_rate": 0.0003204666275218223,
      "loss": 5.2509,
      "mean_token_accuracy": 0.20148188769817352,
      "num_tokens": 102049449.0,
      "step": 50365
    },
    {
      "entropy": 5.787611436843872,
      "epoch": 4.401817705147252,
      "grad_norm": 1.125,
      "learning_rate": 0.0003204361088327111,
      "loss": 5.2814,
      "mean_token_accuracy": 0.20339420586824417,
      "num_tokens": 102059928.0,
      "step": 50370
    },
    {
      "entropy": 5.798432111740112,
      "epoch": 4.4022546534999565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032040558927211147,
      "loss": 5.2322,
      "mean_token_accuracy": 0.20991914868354797,
      "num_tokens": 102070065.0,
      "step": 50375
    },
    {
      "entropy": 5.809516382217407,
      "epoch": 4.402691601852661,
      "grad_norm": 1.125,
      "learning_rate": 0.0003203750688406089,
      "loss": 5.2488,
      "mean_token_accuracy": 0.20735037922859192,
      "num_tokens": 102080990.0,
      "step": 50380
    },
    {
      "entropy": 5.697838068008423,
      "epoch": 4.403128550205365,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032034454753878874,
      "loss": 5.1474,
      "mean_token_accuracy": 0.22073110491037368,
      "num_tokens": 102090237.0,
      "step": 50385
    },
    {
      "entropy": 5.700302982330323,
      "epoch": 4.403565498558071,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032031402536723634,
      "loss": 5.2264,
      "mean_token_accuracy": 0.21180673092603683,
      "num_tokens": 102100685.0,
      "step": 50390
    },
    {
      "entropy": 5.75528883934021,
      "epoch": 4.404002446910775,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003202835023265372,
      "loss": 5.1996,
      "mean_token_accuracy": 0.2078606441617012,
      "num_tokens": 102110218.0,
      "step": 50395
    },
    {
      "entropy": 5.821447134017944,
      "epoch": 4.40443939526348,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003202529784172768,
      "loss": 5.274,
      "mean_token_accuracy": 0.20267583578824996,
      "num_tokens": 102121009.0,
      "step": 50400
    },
    {
      "entropy": 5.78174057006836,
      "epoch": 4.404876343616184,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032022245364004057,
      "loss": 5.185,
      "mean_token_accuracy": 0.20998328626155854,
      "num_tokens": 102130840.0,
      "step": 50405
    },
    {
      "entropy": 5.71011061668396,
      "epoch": 4.4053132919688895,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032019192799541396,
      "loss": 5.2268,
      "mean_token_accuracy": 0.21057790219783784,
      "num_tokens": 102141289.0,
      "step": 50410
    },
    {
      "entropy": 5.684201955795288,
      "epoch": 4.405750240321594,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032016140148398247,
      "loss": 5.1606,
      "mean_token_accuracy": 0.21737851351499557,
      "num_tokens": 102150945.0,
      "step": 50415
    },
    {
      "entropy": 5.7628796100616455,
      "epoch": 4.4061871886742985,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003201308741063317,
      "loss": 5.2787,
      "mean_token_accuracy": 0.20283884555101395,
      "num_tokens": 102161251.0,
      "step": 50420
    },
    {
      "entropy": 5.742268085479736,
      "epoch": 4.406624137027004,
      "grad_norm": 1.15625,
      "learning_rate": 0.000320100345863047,
      "loss": 5.2333,
      "mean_token_accuracy": 0.2053096979856491,
      "num_tokens": 102170848.0,
      "step": 50425
    },
    {
      "entropy": 5.7979577541351315,
      "epoch": 4.407061085379708,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003200698167547141,
      "loss": 5.2699,
      "mean_token_accuracy": 0.20893613249063492,
      "num_tokens": 102181533.0,
      "step": 50430
    },
    {
      "entropy": 5.7442341327667235,
      "epoch": 4.407498033732413,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032003928678191857,
      "loss": 5.1807,
      "mean_token_accuracy": 0.21073989868164061,
      "num_tokens": 102191868.0,
      "step": 50435
    },
    {
      "entropy": 5.7136609077453615,
      "epoch": 4.407934982085117,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032000875594524584,
      "loss": 5.1795,
      "mean_token_accuracy": 0.2130434036254883,
      "num_tokens": 102201164.0,
      "step": 50440
    },
    {
      "entropy": 5.725419807434082,
      "epoch": 4.408371930437823,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031997822424528164,
      "loss": 5.2079,
      "mean_token_accuracy": 0.21580520272254944,
      "num_tokens": 102211309.0,
      "step": 50445
    },
    {
      "entropy": 5.824331951141358,
      "epoch": 4.408808878790527,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003199476916826115,
      "loss": 5.2844,
      "mean_token_accuracy": 0.19647733718156815,
      "num_tokens": 102221721.0,
      "step": 50450
    },
    {
      "entropy": 5.728116035461426,
      "epoch": 4.4092458271432315,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031991715825782104,
      "loss": 5.2306,
      "mean_token_accuracy": 0.20462077409029006,
      "num_tokens": 102231880.0,
      "step": 50455
    },
    {
      "entropy": 5.7582988262176515,
      "epoch": 4.409682775495936,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031988662397149597,
      "loss": 5.1967,
      "mean_token_accuracy": 0.21251836866140367,
      "num_tokens": 102243027.0,
      "step": 50460
    },
    {
      "entropy": 5.731796932220459,
      "epoch": 4.410119723848641,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003198560888242219,
      "loss": 5.1845,
      "mean_token_accuracy": 0.21624772548675536,
      "num_tokens": 102253117.0,
      "step": 50465
    },
    {
      "entropy": 5.701040744781494,
      "epoch": 4.410556672201346,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003198255528165846,
      "loss": 5.1285,
      "mean_token_accuracy": 0.21496167331933974,
      "num_tokens": 102262483.0,
      "step": 50470
    },
    {
      "entropy": 5.675022172927856,
      "epoch": 4.41099362055405,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003197950159491697,
      "loss": 5.1133,
      "mean_token_accuracy": 0.22045219093561172,
      "num_tokens": 102271213.0,
      "step": 50475
    },
    {
      "entropy": 5.775885009765625,
      "epoch": 4.411430568906756,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031976447822256295,
      "loss": 5.2724,
      "mean_token_accuracy": 0.20296328365802765,
      "num_tokens": 102280830.0,
      "step": 50480
    },
    {
      "entropy": 5.755123567581177,
      "epoch": 4.41186751725946,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031973393963735,
      "loss": 5.2352,
      "mean_token_accuracy": 0.21019665747880936,
      "num_tokens": 102290869.0,
      "step": 50485
    },
    {
      "entropy": 5.677643918991089,
      "epoch": 4.412304465612165,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003197034001941166,
      "loss": 5.1982,
      "mean_token_accuracy": 0.21656055748462677,
      "num_tokens": 102301037.0,
      "step": 50490
    },
    {
      "entropy": 5.760178518295288,
      "epoch": 4.412741413964869,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031967285989344867,
      "loss": 5.2314,
      "mean_token_accuracy": 0.2145061120390892,
      "num_tokens": 102311485.0,
      "step": 50495
    },
    {
      "entropy": 5.790382623672485,
      "epoch": 4.413178362317574,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003196423187359319,
      "loss": 5.2155,
      "mean_token_accuracy": 0.2129572957754135,
      "num_tokens": 102320685.0,
      "step": 50500
    },
    {
      "entropy": 5.708336734771729,
      "epoch": 4.413615310670279,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031961177672215196,
      "loss": 5.1418,
      "mean_token_accuracy": 0.22678590416908265,
      "num_tokens": 102330946.0,
      "step": 50505
    },
    {
      "entropy": 5.769126319885254,
      "epoch": 4.414052259022983,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031958123385269485,
      "loss": 5.224,
      "mean_token_accuracy": 0.2100039929151535,
      "num_tokens": 102340638.0,
      "step": 50510
    },
    {
      "entropy": 5.769824695587158,
      "epoch": 4.414489207375688,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031955069012814624,
      "loss": 5.2195,
      "mean_token_accuracy": 0.21042299121618271,
      "num_tokens": 102350691.0,
      "step": 50515
    },
    {
      "entropy": 5.803566551208496,
      "epoch": 4.414926155728393,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003195201455490921,
      "loss": 5.2815,
      "mean_token_accuracy": 0.20191489160060883,
      "num_tokens": 102361572.0,
      "step": 50520
    },
    {
      "entropy": 5.7826958179473875,
      "epoch": 4.415363104081098,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031948960011611815,
      "loss": 5.2464,
      "mean_token_accuracy": 0.20477522760629654,
      "num_tokens": 102371599.0,
      "step": 50525
    },
    {
      "entropy": 5.69868106842041,
      "epoch": 4.415800052433802,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003194590538298105,
      "loss": 5.2038,
      "mean_token_accuracy": 0.2113535076379776,
      "num_tokens": 102381987.0,
      "step": 50530
    },
    {
      "entropy": 5.7724686622619625,
      "epoch": 4.4162370007865075,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003194285066907548,
      "loss": 5.2635,
      "mean_token_accuracy": 0.2059922471642494,
      "num_tokens": 102392677.0,
      "step": 50535
    },
    {
      "entropy": 5.743250322341919,
      "epoch": 4.416673949139212,
      "grad_norm": 1.125,
      "learning_rate": 0.0003193979586995371,
      "loss": 5.1843,
      "mean_token_accuracy": 0.21778950542211534,
      "num_tokens": 102402553.0,
      "step": 50540
    },
    {
      "entropy": 5.746118259429932,
      "epoch": 4.417110897491916,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031936740985674325,
      "loss": 5.1675,
      "mean_token_accuracy": 0.21470805406570434,
      "num_tokens": 102412291.0,
      "step": 50545
    },
    {
      "entropy": 5.713998603820801,
      "epoch": 4.417547845844621,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003193368601629593,
      "loss": 5.2028,
      "mean_token_accuracy": 0.2120678648352623,
      "num_tokens": 102421749.0,
      "step": 50550
    },
    {
      "entropy": 5.703938245773315,
      "epoch": 4.417984794197326,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031930630961877105,
      "loss": 5.2335,
      "mean_token_accuracy": 0.2091955617070198,
      "num_tokens": 102432170.0,
      "step": 50555
    },
    {
      "entropy": 5.7769636631011965,
      "epoch": 4.418421742550031,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003192757582247646,
      "loss": 5.224,
      "mean_token_accuracy": 0.20576864778995513,
      "num_tokens": 102442454.0,
      "step": 50560
    },
    {
      "entropy": 5.697019100189209,
      "epoch": 4.418858690902735,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031924520598152595,
      "loss": 5.1565,
      "mean_token_accuracy": 0.21980048716068268,
      "num_tokens": 102454007.0,
      "step": 50565
    },
    {
      "entropy": 5.763720369338989,
      "epoch": 4.41929563925544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031921465288964096,
      "loss": 5.2661,
      "mean_token_accuracy": 0.20995951741933822,
      "num_tokens": 102464601.0,
      "step": 50570
    },
    {
      "entropy": 5.758732795715332,
      "epoch": 4.419732587608145,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00031918409894969585,
      "loss": 5.3029,
      "mean_token_accuracy": 0.20416154712438583,
      "num_tokens": 102475552.0,
      "step": 50575
    },
    {
      "entropy": 5.779656076431275,
      "epoch": 4.4201695359608495,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003191535441622766,
      "loss": 5.2488,
      "mean_token_accuracy": 0.2065800666809082,
      "num_tokens": 102485934.0,
      "step": 50580
    },
    {
      "entropy": 5.76207275390625,
      "epoch": 4.420606484313554,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003191229885279692,
      "loss": 5.2552,
      "mean_token_accuracy": 0.20519187152385712,
      "num_tokens": 102494987.0,
      "step": 50585
    },
    {
      "entropy": 5.783018636703491,
      "epoch": 4.421043432666258,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031909243204735974,
      "loss": 5.2195,
      "mean_token_accuracy": 0.19965447038412093,
      "num_tokens": 102506556.0,
      "step": 50590
    },
    {
      "entropy": 5.779876041412353,
      "epoch": 4.421480381018964,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003190618747210343,
      "loss": 5.2332,
      "mean_token_accuracy": 0.20960990488529205,
      "num_tokens": 102516483.0,
      "step": 50595
    },
    {
      "entropy": 5.68052568435669,
      "epoch": 4.421917329371668,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003190313165495791,
      "loss": 5.1819,
      "mean_token_accuracy": 0.21663731336593628,
      "num_tokens": 102526502.0,
      "step": 50600
    },
    {
      "entropy": 5.789612483978272,
      "epoch": 4.422354277724373,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003190007575335801,
      "loss": 5.2413,
      "mean_token_accuracy": 0.2041676625609398,
      "num_tokens": 102536376.0,
      "step": 50605
    },
    {
      "entropy": 5.711915588378906,
      "epoch": 4.422791226077078,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003189701976736236,
      "loss": 5.147,
      "mean_token_accuracy": 0.21221895813941954,
      "num_tokens": 102546221.0,
      "step": 50610
    },
    {
      "entropy": 5.745306158065796,
      "epoch": 4.4232281744297826,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031893963697029554,
      "loss": 5.2029,
      "mean_token_accuracy": 0.21728623360395433,
      "num_tokens": 102556412.0,
      "step": 50615
    },
    {
      "entropy": 5.733313751220703,
      "epoch": 4.423665122782487,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031890907542418234,
      "loss": 5.2093,
      "mean_token_accuracy": 0.21097391843795776,
      "num_tokens": 102566535.0,
      "step": 50620
    },
    {
      "entropy": 5.822951555252075,
      "epoch": 4.4241020711351915,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031887851303587,
      "loss": 5.3331,
      "mean_token_accuracy": 0.2003961205482483,
      "num_tokens": 102576902.0,
      "step": 50625
    },
    {
      "entropy": 5.829863357543945,
      "epoch": 4.424539019487897,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003188479498059448,
      "loss": 5.2448,
      "mean_token_accuracy": 0.20459116399288177,
      "num_tokens": 102587604.0,
      "step": 50630
    },
    {
      "entropy": 5.736758995056152,
      "epoch": 4.424975967840601,
      "grad_norm": 1.0625,
      "learning_rate": 0.000318817385734993,
      "loss": 5.1907,
      "mean_token_accuracy": 0.21026345044374467,
      "num_tokens": 102597978.0,
      "step": 50635
    },
    {
      "entropy": 5.8124488353729244,
      "epoch": 4.425412916193306,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003187868208236007,
      "loss": 5.2678,
      "mean_token_accuracy": 0.20555341839790345,
      "num_tokens": 102608156.0,
      "step": 50640
    },
    {
      "entropy": 5.745951795578003,
      "epoch": 4.42584986454601,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031875625507235436,
      "loss": 5.2446,
      "mean_token_accuracy": 0.21182074695825576,
      "num_tokens": 102617726.0,
      "step": 50645
    },
    {
      "entropy": 5.76117434501648,
      "epoch": 4.426286812898716,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031872568848183996,
      "loss": 5.2315,
      "mean_token_accuracy": 0.21142585426568986,
      "num_tokens": 102628444.0,
      "step": 50650
    },
    {
      "entropy": 5.755072784423828,
      "epoch": 4.42672376125142,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031869512105264395,
      "loss": 5.2069,
      "mean_token_accuracy": 0.20965819358825682,
      "num_tokens": 102637751.0,
      "step": 50655
    },
    {
      "entropy": 5.756903982162475,
      "epoch": 4.4271607096041246,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003186645527853527,
      "loss": 5.2832,
      "mean_token_accuracy": 0.2014614686369896,
      "num_tokens": 102648276.0,
      "step": 50660
    },
    {
      "entropy": 5.730854892730713,
      "epoch": 4.42759765795683,
      "grad_norm": 0.96875,
      "learning_rate": 0.00031863398368055246,
      "loss": 5.1178,
      "mean_token_accuracy": 0.2197858840227127,
      "num_tokens": 102659113.0,
      "step": 50665
    },
    {
      "entropy": 5.713173151016235,
      "epoch": 4.428034606309534,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003186034137388296,
      "loss": 5.2285,
      "mean_token_accuracy": 0.20751500874757767,
      "num_tokens": 102669502.0,
      "step": 50670
    },
    {
      "entropy": 5.749173784255982,
      "epoch": 4.428471554662239,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003185728429607703,
      "loss": 5.1269,
      "mean_token_accuracy": 0.21779083758592604,
      "num_tokens": 102679804.0,
      "step": 50675
    },
    {
      "entropy": 5.740651178359985,
      "epoch": 4.428908503014943,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031854227134696107,
      "loss": 5.1349,
      "mean_token_accuracy": 0.21736924201250077,
      "num_tokens": 102689466.0,
      "step": 50680
    },
    {
      "entropy": 5.696164178848266,
      "epoch": 4.429345451367649,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031851169889798826,
      "loss": 5.1567,
      "mean_token_accuracy": 0.22009893655776977,
      "num_tokens": 102699957.0,
      "step": 50685
    },
    {
      "entropy": 5.735061311721802,
      "epoch": 4.429782399720353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031848112561443825,
      "loss": 5.1465,
      "mean_token_accuracy": 0.21483104526996613,
      "num_tokens": 102710575.0,
      "step": 50690
    },
    {
      "entropy": 5.705440092086792,
      "epoch": 4.430219348073058,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031845055149689754,
      "loss": 5.1892,
      "mean_token_accuracy": 0.21252232044935226,
      "num_tokens": 102720575.0,
      "step": 50695
    },
    {
      "entropy": 5.754639625549316,
      "epoch": 4.430656296425762,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031841997654595246,
      "loss": 5.2487,
      "mean_token_accuracy": 0.20444642007350922,
      "num_tokens": 102730202.0,
      "step": 50700
    },
    {
      "entropy": 5.837565755844116,
      "epoch": 4.431093244778467,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003183894007621894,
      "loss": 5.3543,
      "mean_token_accuracy": 0.20100207477808,
      "num_tokens": 102741305.0,
      "step": 50705
    },
    {
      "entropy": 5.823951768875122,
      "epoch": 4.431530193131172,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000318358824146195,
      "loss": 5.1952,
      "mean_token_accuracy": 0.20930145233869552,
      "num_tokens": 102752453.0,
      "step": 50710
    },
    {
      "entropy": 5.7503691673278805,
      "epoch": 4.431967141483876,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003183282466985555,
      "loss": 5.2276,
      "mean_token_accuracy": 0.2096305325627327,
      "num_tokens": 102763095.0,
      "step": 50715
    },
    {
      "entropy": 5.748608732223511,
      "epoch": 4.432404089836581,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031829766841985765,
      "loss": 5.2305,
      "mean_token_accuracy": 0.20464946329593658,
      "num_tokens": 102774239.0,
      "step": 50720
    },
    {
      "entropy": 5.741477203369141,
      "epoch": 4.432841038189286,
      "grad_norm": 1.0,
      "learning_rate": 0.00031826708931068776,
      "loss": 5.1468,
      "mean_token_accuracy": 0.21234796792268754,
      "num_tokens": 102783577.0,
      "step": 50725
    },
    {
      "entropy": 5.708409929275513,
      "epoch": 4.433277986541991,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031823650937163245,
      "loss": 5.0895,
      "mean_token_accuracy": 0.21623615622520448,
      "num_tokens": 102793647.0,
      "step": 50730
    },
    {
      "entropy": 5.673356246948242,
      "epoch": 4.433714934894695,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031820592860327826,
      "loss": 5.1155,
      "mean_token_accuracy": 0.220275616645813,
      "num_tokens": 102804732.0,
      "step": 50735
    },
    {
      "entropy": 5.7444370746612545,
      "epoch": 4.4341518832474005,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031817534700621176,
      "loss": 5.2029,
      "mean_token_accuracy": 0.21761397868394852,
      "num_tokens": 102813783.0,
      "step": 50740
    },
    {
      "entropy": 5.6779749393463135,
      "epoch": 4.434588831600105,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003181447645810194,
      "loss": 5.1375,
      "mean_token_accuracy": 0.22400065511465073,
      "num_tokens": 102824856.0,
      "step": 50745
    },
    {
      "entropy": 5.779857873916626,
      "epoch": 4.435025779952809,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031811418132828776,
      "loss": 5.2472,
      "mean_token_accuracy": 0.20168827772140502,
      "num_tokens": 102836321.0,
      "step": 50750
    },
    {
      "entropy": 5.783146858215332,
      "epoch": 4.435462728305514,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031808359724860364,
      "loss": 5.2173,
      "mean_token_accuracy": 0.20702200531959533,
      "num_tokens": 102846446.0,
      "step": 50755
    },
    {
      "entropy": 5.73444299697876,
      "epoch": 4.435899676658219,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031805301234255363,
      "loss": 5.2291,
      "mean_token_accuracy": 0.20853864401578903,
      "num_tokens": 102856702.0,
      "step": 50760
    },
    {
      "entropy": 5.739704179763794,
      "epoch": 4.436336625010924,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003180224266107242,
      "loss": 5.152,
      "mean_token_accuracy": 0.214667709171772,
      "num_tokens": 102866271.0,
      "step": 50765
    },
    {
      "entropy": 5.757342576980591,
      "epoch": 4.436773573363628,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031799184005370205,
      "loss": 5.2056,
      "mean_token_accuracy": 0.21085194647312164,
      "num_tokens": 102876063.0,
      "step": 50770
    },
    {
      "entropy": 5.671745681762696,
      "epoch": 4.437210521716333,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003179612526720739,
      "loss": 5.1172,
      "mean_token_accuracy": 0.2165062114596367,
      "num_tokens": 102885237.0,
      "step": 50775
    },
    {
      "entropy": 5.74724850654602,
      "epoch": 4.437647470069038,
      "grad_norm": 1.125,
      "learning_rate": 0.00031793066446642645,
      "loss": 5.191,
      "mean_token_accuracy": 0.2168621227145195,
      "num_tokens": 102895515.0,
      "step": 50780
    },
    {
      "entropy": 5.718772935867309,
      "epoch": 4.4380844184217425,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031790007543734634,
      "loss": 5.2471,
      "mean_token_accuracy": 0.20346767604351043,
      "num_tokens": 102905651.0,
      "step": 50785
    },
    {
      "entropy": 5.794508838653565,
      "epoch": 4.438521366774447,
      "grad_norm": 1.125,
      "learning_rate": 0.0003178694855854203,
      "loss": 5.2824,
      "mean_token_accuracy": 0.2046641007065773,
      "num_tokens": 102916195.0,
      "step": 50790
    },
    {
      "entropy": 5.7509942054748535,
      "epoch": 4.438958315127152,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003178388949112351,
      "loss": 5.1395,
      "mean_token_accuracy": 0.219295896589756,
      "num_tokens": 102925952.0,
      "step": 50795
    },
    {
      "entropy": 5.8519855499267575,
      "epoch": 4.439395263479857,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003178083034153774,
      "loss": 5.3576,
      "mean_token_accuracy": 0.19090425074100495,
      "num_tokens": 102936313.0,
      "step": 50800
    },
    {
      "entropy": 5.753152990341187,
      "epoch": 4.439832211832561,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031777771109843403,
      "loss": 5.2505,
      "mean_token_accuracy": 0.2052447259426117,
      "num_tokens": 102947031.0,
      "step": 50805
    },
    {
      "entropy": 5.826598691940307,
      "epoch": 4.440269160185266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003177471179609918,
      "loss": 5.3166,
      "mean_token_accuracy": 0.2110891118645668,
      "num_tokens": 102958063.0,
      "step": 50810
    },
    {
      "entropy": 5.8159102439880375,
      "epoch": 4.440706108537971,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00031771652400363736,
      "loss": 5.2609,
      "mean_token_accuracy": 0.20675889849662782,
      "num_tokens": 102968234.0,
      "step": 50815
    },
    {
      "entropy": 5.80595211982727,
      "epoch": 4.441143056890676,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003176859292269576,
      "loss": 5.2298,
      "mean_token_accuracy": 0.2069159597158432,
      "num_tokens": 102977220.0,
      "step": 50820
    },
    {
      "entropy": 5.7515191555023195,
      "epoch": 4.44158000524338,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031765533363153935,
      "loss": 5.2533,
      "mean_token_accuracy": 0.20773472338914872,
      "num_tokens": 102986190.0,
      "step": 50825
    },
    {
      "entropy": 5.653506946563721,
      "epoch": 4.4420169535960845,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031762473721796947,
      "loss": 5.1972,
      "mean_token_accuracy": 0.21623332053422928,
      "num_tokens": 102996299.0,
      "step": 50830
    },
    {
      "entropy": 5.798867464065552,
      "epoch": 4.44245390194879,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003175941399868349,
      "loss": 5.2173,
      "mean_token_accuracy": 0.20430122464895248,
      "num_tokens": 103005045.0,
      "step": 50835
    },
    {
      "entropy": 5.755445861816407,
      "epoch": 4.442890850301494,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031756354193872234,
      "loss": 5.2033,
      "mean_token_accuracy": 0.20960768610239028,
      "num_tokens": 103014299.0,
      "step": 50840
    },
    {
      "entropy": 5.789973020553589,
      "epoch": 4.443327798654199,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003175329430742187,
      "loss": 5.2079,
      "mean_token_accuracy": 0.21184249818325043,
      "num_tokens": 103023114.0,
      "step": 50845
    },
    {
      "entropy": 5.814491271972656,
      "epoch": 4.443764747006904,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031750234339391085,
      "loss": 5.2666,
      "mean_token_accuracy": 0.2083463415503502,
      "num_tokens": 103033405.0,
      "step": 50850
    },
    {
      "entropy": 5.70550217628479,
      "epoch": 4.444201695359609,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003174717428983859,
      "loss": 5.1098,
      "mean_token_accuracy": 0.21933805644512178,
      "num_tokens": 103042935.0,
      "step": 50855
    },
    {
      "entropy": 5.707387447357178,
      "epoch": 4.444638643712313,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031744114158823074,
      "loss": 5.1616,
      "mean_token_accuracy": 0.21928768306970597,
      "num_tokens": 103052104.0,
      "step": 50860
    },
    {
      "entropy": 5.658131408691406,
      "epoch": 4.445075592065018,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003174105394640321,
      "loss": 5.1559,
      "mean_token_accuracy": 0.2171228513121605,
      "num_tokens": 103061953.0,
      "step": 50865
    },
    {
      "entropy": 5.733641958236694,
      "epoch": 4.445512540417723,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003173799365263771,
      "loss": 5.1719,
      "mean_token_accuracy": 0.22078555524349214,
      "num_tokens": 103072168.0,
      "step": 50870
    },
    {
      "entropy": 5.740268802642822,
      "epoch": 4.445949488770427,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031734933277585275,
      "loss": 5.1955,
      "mean_token_accuracy": 0.2153225213289261,
      "num_tokens": 103083029.0,
      "step": 50875
    },
    {
      "entropy": 5.8232824325561525,
      "epoch": 4.446386437123132,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031731872821304604,
      "loss": 5.3219,
      "mean_token_accuracy": 0.20483147203922272,
      "num_tokens": 103094631.0,
      "step": 50880
    },
    {
      "entropy": 5.773932313919067,
      "epoch": 4.446823385475836,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031728812283854396,
      "loss": 5.2614,
      "mean_token_accuracy": 0.21058017611503602,
      "num_tokens": 103104906.0,
      "step": 50885
    },
    {
      "entropy": 5.72383828163147,
      "epoch": 4.447260333828542,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003172575166529335,
      "loss": 5.2327,
      "mean_token_accuracy": 0.21549891233444213,
      "num_tokens": 103115591.0,
      "step": 50890
    },
    {
      "entropy": 5.75708794593811,
      "epoch": 4.447697282181246,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031722690965680164,
      "loss": 5.1533,
      "mean_token_accuracy": 0.21338302344083787,
      "num_tokens": 103125808.0,
      "step": 50895
    },
    {
      "entropy": 5.768477487564087,
      "epoch": 4.448134230533951,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031719630185073555,
      "loss": 5.2107,
      "mean_token_accuracy": 0.2149913415312767,
      "num_tokens": 103136291.0,
      "step": 50900
    },
    {
      "entropy": 5.805545711517334,
      "epoch": 4.448571178886656,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003171656932353224,
      "loss": 5.2699,
      "mean_token_accuracy": 0.20426324754953384,
      "num_tokens": 103145949.0,
      "step": 50905
    },
    {
      "entropy": 5.760156202316284,
      "epoch": 4.4490081272393605,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031713508381114907,
      "loss": 5.1864,
      "mean_token_accuracy": 0.2120803639292717,
      "num_tokens": 103155375.0,
      "step": 50910
    },
    {
      "entropy": 5.678969287872315,
      "epoch": 4.449445075592065,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031710447357880277,
      "loss": 5.1262,
      "mean_token_accuracy": 0.2167724460363388,
      "num_tokens": 103164617.0,
      "step": 50915
    },
    {
      "entropy": 5.75317611694336,
      "epoch": 4.449882023944769,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003170738625388706,
      "loss": 5.2359,
      "mean_token_accuracy": 0.20900696218013765,
      "num_tokens": 103174688.0,
      "step": 50920
    },
    {
      "entropy": 5.798268842697143,
      "epoch": 4.450318972297475,
      "grad_norm": 1.125,
      "learning_rate": 0.0003170432506919397,
      "loss": 5.2934,
      "mean_token_accuracy": 0.198031884431839,
      "num_tokens": 103185186.0,
      "step": 50925
    },
    {
      "entropy": 5.77655725479126,
      "epoch": 4.450755920650179,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003170126380385972,
      "loss": 5.2128,
      "mean_token_accuracy": 0.21251717656850816,
      "num_tokens": 103194818.0,
      "step": 50930
    },
    {
      "entropy": 5.6838541507720945,
      "epoch": 4.451192869002884,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031698202457943034,
      "loss": 5.1671,
      "mean_token_accuracy": 0.2162016898393631,
      "num_tokens": 103205299.0,
      "step": 50935
    },
    {
      "entropy": 5.788860702514649,
      "epoch": 4.451629817355588,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031695141031502615,
      "loss": 5.3141,
      "mean_token_accuracy": 0.2031777337193489,
      "num_tokens": 103216157.0,
      "step": 50940
    },
    {
      "entropy": 5.7745545387268065,
      "epoch": 4.4520667657082935,
      "grad_norm": 1.140625,
      "learning_rate": 0.000316920795245972,
      "loss": 5.1697,
      "mean_token_accuracy": 0.21280278265476227,
      "num_tokens": 103227009.0,
      "step": 50945
    },
    {
      "entropy": 5.801250457763672,
      "epoch": 4.452503714060998,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031689017937285507,
      "loss": 5.2053,
      "mean_token_accuracy": 0.20407239347696304,
      "num_tokens": 103236848.0,
      "step": 50950
    },
    {
      "entropy": 5.739700555801392,
      "epoch": 4.4529406624137025,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031685956269626253,
      "loss": 5.2712,
      "mean_token_accuracy": 0.2068374216556549,
      "num_tokens": 103247664.0,
      "step": 50955
    },
    {
      "entropy": 5.750329446792603,
      "epoch": 4.453377610766408,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003168289452167816,
      "loss": 5.2208,
      "mean_token_accuracy": 0.2088846430182457,
      "num_tokens": 103258605.0,
      "step": 50960
    },
    {
      "entropy": 5.769655084609985,
      "epoch": 4.453814559119112,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003167983269349996,
      "loss": 5.2103,
      "mean_token_accuracy": 0.2120337724685669,
      "num_tokens": 103267903.0,
      "step": 50965
    },
    {
      "entropy": 5.739063739776611,
      "epoch": 4.454251507471817,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003167677078515038,
      "loss": 5.1591,
      "mean_token_accuracy": 0.22083741128444673,
      "num_tokens": 103278189.0,
      "step": 50970
    },
    {
      "entropy": 5.697249126434326,
      "epoch": 4.454688455824521,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003167370879668815,
      "loss": 5.1703,
      "mean_token_accuracy": 0.21378941535949708,
      "num_tokens": 103289250.0,
      "step": 50975
    },
    {
      "entropy": 5.826480484008789,
      "epoch": 4.455125404177227,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031670646728172,
      "loss": 5.2528,
      "mean_token_accuracy": 0.2067952498793602,
      "num_tokens": 103298572.0,
      "step": 50980
    },
    {
      "entropy": 5.709837961196899,
      "epoch": 4.455562352529931,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003166758457966065,
      "loss": 5.1236,
      "mean_token_accuracy": 0.21886558085680008,
      "num_tokens": 103310560.0,
      "step": 50985
    },
    {
      "entropy": 5.749885416030883,
      "epoch": 4.4559993008826355,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003166452235121285,
      "loss": 5.2373,
      "mean_token_accuracy": 0.20662304610013962,
      "num_tokens": 103319854.0,
      "step": 50990
    },
    {
      "entropy": 5.773276519775391,
      "epoch": 4.45643624923534,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031661460042887323,
      "loss": 5.2327,
      "mean_token_accuracy": 0.21189794391393663,
      "num_tokens": 103330286.0,
      "step": 50995
    },
    {
      "entropy": 5.770065450668335,
      "epoch": 4.456873197588045,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031658397654742826,
      "loss": 5.273,
      "mean_token_accuracy": 0.2109268456697464,
      "num_tokens": 103340253.0,
      "step": 51000
    },
    {
      "epoch": 4.456873197588045,
      "eval_entropy": 5.480615053007869,
      "eval_loss": 5.301793098449707,
      "eval_mean_token_accuracy": 0.21348277688097278,
      "eval_num_tokens": 103340253.0,
      "eval_runtime": 28.3567,
      "eval_samples_per_second": 1305.298,
      "eval_steps_per_second": 163.171,
      "step": 51000
    },
    {
      "entropy": 5.739485406875611,
      "epoch": 4.45731014594075,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031655335186838075,
      "loss": 5.2245,
      "mean_token_accuracy": 0.20056513994932174,
      "num_tokens": 103349785.0,
      "step": 51005
    },
    {
      "entropy": 5.748442554473877,
      "epoch": 4.457747094293454,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031652272639231817,
      "loss": 5.2499,
      "mean_token_accuracy": 0.20552408546209336,
      "num_tokens": 103360378.0,
      "step": 51010
    },
    {
      "entropy": 5.867218017578125,
      "epoch": 4.45818404264616,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00031649210011982793,
      "loss": 5.2423,
      "mean_token_accuracy": 0.20882003009319305,
      "num_tokens": 103372101.0,
      "step": 51015
    },
    {
      "entropy": 5.817474889755249,
      "epoch": 4.458620990998864,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003164614730514975,
      "loss": 5.2354,
      "mean_token_accuracy": 0.20691160559654237,
      "num_tokens": 103381988.0,
      "step": 51020
    },
    {
      "entropy": 5.691746759414673,
      "epoch": 4.459057939351569,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003164308451879143,
      "loss": 5.2603,
      "mean_token_accuracy": 0.2066364571452141,
      "num_tokens": 103392410.0,
      "step": 51025
    },
    {
      "entropy": 5.6253876209259035,
      "epoch": 4.459494887704273,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003164002165296657,
      "loss": 5.0767,
      "mean_token_accuracy": 0.22134896665811538,
      "num_tokens": 103402502.0,
      "step": 51030
    },
    {
      "entropy": 5.704686164855957,
      "epoch": 4.459931836056978,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003163695870773393,
      "loss": 5.2029,
      "mean_token_accuracy": 0.21332536786794662,
      "num_tokens": 103412918.0,
      "step": 51035
    },
    {
      "entropy": 5.79235429763794,
      "epoch": 4.460368784409683,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031633895683152253,
      "loss": 5.2459,
      "mean_token_accuracy": 0.2078450158238411,
      "num_tokens": 103423160.0,
      "step": 51040
    },
    {
      "entropy": 5.738766717910766,
      "epoch": 4.460805732762387,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000316308325792803,
      "loss": 5.1903,
      "mean_token_accuracy": 0.20923637300729753,
      "num_tokens": 103433617.0,
      "step": 51045
    },
    {
      "entropy": 5.783624505996704,
      "epoch": 4.461242681115092,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031627769396176805,
      "loss": 5.2597,
      "mean_token_accuracy": 0.20631895810365677,
      "num_tokens": 103444080.0,
      "step": 51050
    },
    {
      "entropy": 5.732080554962158,
      "epoch": 4.461679629467797,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031624706133900535,
      "loss": 5.1931,
      "mean_token_accuracy": 0.21046814173460007,
      "num_tokens": 103454562.0,
      "step": 51055
    },
    {
      "entropy": 5.766240215301513,
      "epoch": 4.462116577820502,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003162164279251023,
      "loss": 5.2719,
      "mean_token_accuracy": 0.20710860788822175,
      "num_tokens": 103465407.0,
      "step": 51060
    },
    {
      "entropy": 5.722429037094116,
      "epoch": 4.462553526173206,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031618579372064663,
      "loss": 5.1792,
      "mean_token_accuracy": 0.21363016366958618,
      "num_tokens": 103475777.0,
      "step": 51065
    },
    {
      "entropy": 5.714059448242187,
      "epoch": 4.4629904745259115,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003161551587262259,
      "loss": 5.1719,
      "mean_token_accuracy": 0.21138861477375032,
      "num_tokens": 103486380.0,
      "step": 51070
    },
    {
      "entropy": 5.746279287338257,
      "epoch": 4.463427422878616,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003161245229424276,
      "loss": 5.1845,
      "mean_token_accuracy": 0.2136567860841751,
      "num_tokens": 103495127.0,
      "step": 51075
    },
    {
      "entropy": 5.767485570907593,
      "epoch": 4.46386437123132,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003160938863698394,
      "loss": 5.2793,
      "mean_token_accuracy": 0.20132714956998826,
      "num_tokens": 103504781.0,
      "step": 51080
    },
    {
      "entropy": 5.6789875507354735,
      "epoch": 4.464301319584025,
      "grad_norm": 1.109375,
      "learning_rate": 0.000316063249009049,
      "loss": 5.1709,
      "mean_token_accuracy": 0.21668038815259932,
      "num_tokens": 103514969.0,
      "step": 51085
    },
    {
      "entropy": 5.806318998336792,
      "epoch": 4.46473826793673,
      "grad_norm": 1.125,
      "learning_rate": 0.0003160326108606438,
      "loss": 5.2308,
      "mean_token_accuracy": 0.20806910693645478,
      "num_tokens": 103525097.0,
      "step": 51090
    },
    {
      "entropy": 5.778359842300415,
      "epoch": 4.465175216289435,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031600197192521186,
      "loss": 5.2035,
      "mean_token_accuracy": 0.21645378619432448,
      "num_tokens": 103535201.0,
      "step": 51095
    },
    {
      "entropy": 5.690120983123779,
      "epoch": 4.465612164642139,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003159713322033405,
      "loss": 5.1698,
      "mean_token_accuracy": 0.21752961575984955,
      "num_tokens": 103546257.0,
      "step": 51100
    },
    {
      "entropy": 5.724799966812133,
      "epoch": 4.466049112994844,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003159406916956174,
      "loss": 5.2023,
      "mean_token_accuracy": 0.20572661757469177,
      "num_tokens": 103556825.0,
      "step": 51105
    },
    {
      "entropy": 5.726980781555175,
      "epoch": 4.466486061347549,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003159100504026305,
      "loss": 5.1704,
      "mean_token_accuracy": 0.21482352018356324,
      "num_tokens": 103567819.0,
      "step": 51110
    },
    {
      "entropy": 5.802557945251465,
      "epoch": 4.4669230097002535,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003158794083249674,
      "loss": 5.2262,
      "mean_token_accuracy": 0.20515458583831786,
      "num_tokens": 103577629.0,
      "step": 51115
    },
    {
      "entropy": 5.764522743225098,
      "epoch": 4.467359958052958,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003158487654632158,
      "loss": 5.2397,
      "mean_token_accuracy": 0.20434861481189728,
      "num_tokens": 103587855.0,
      "step": 51120
    },
    {
      "entropy": 5.786198091506958,
      "epoch": 4.467796906405663,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003158181218179635,
      "loss": 5.262,
      "mean_token_accuracy": 0.20708741396665573,
      "num_tokens": 103598022.0,
      "step": 51125
    },
    {
      "entropy": 5.694246244430542,
      "epoch": 4.468233854758368,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031578747738979826,
      "loss": 5.1804,
      "mean_token_accuracy": 0.21975843459367753,
      "num_tokens": 103608212.0,
      "step": 51130
    },
    {
      "entropy": 5.686838817596436,
      "epoch": 4.468670803111072,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003157568321793077,
      "loss": 5.1353,
      "mean_token_accuracy": 0.22209790647029876,
      "num_tokens": 103617394.0,
      "step": 51135
    },
    {
      "entropy": 5.693966293334961,
      "epoch": 4.469107751463777,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031572618618707987,
      "loss": 5.1384,
      "mean_token_accuracy": 0.21921238154172898,
      "num_tokens": 103627566.0,
      "step": 51140
    },
    {
      "entropy": 5.710153865814209,
      "epoch": 4.469544699816482,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031569553941370246,
      "loss": 5.0782,
      "mean_token_accuracy": 0.22147616147994995,
      "num_tokens": 103637066.0,
      "step": 51145
    },
    {
      "entropy": 5.72841854095459,
      "epoch": 4.4699816481691865,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031566489185976324,
      "loss": 5.3159,
      "mean_token_accuracy": 0.2040296196937561,
      "num_tokens": 103649944.0,
      "step": 51150
    },
    {
      "entropy": 5.775235509872436,
      "epoch": 4.470418596521891,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031563424352585,
      "loss": 5.2058,
      "mean_token_accuracy": 0.2095445767045021,
      "num_tokens": 103660703.0,
      "step": 51155
    },
    {
      "entropy": 5.715068435668945,
      "epoch": 4.4708555448745955,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031560359441255083,
      "loss": 5.1771,
      "mean_token_accuracy": 0.2166891872882843,
      "num_tokens": 103671062.0,
      "step": 51160
    },
    {
      "entropy": 5.697284793853759,
      "epoch": 4.471292493227301,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031557294452045335,
      "loss": 5.2215,
      "mean_token_accuracy": 0.21703457236289977,
      "num_tokens": 103681470.0,
      "step": 51165
    },
    {
      "entropy": 5.751542472839356,
      "epoch": 4.471729441580005,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031554229385014553,
      "loss": 5.2587,
      "mean_token_accuracy": 0.2031521439552307,
      "num_tokens": 103691557.0,
      "step": 51170
    },
    {
      "entropy": 5.790063667297363,
      "epoch": 4.47216638993271,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003155116424022153,
      "loss": 5.2638,
      "mean_token_accuracy": 0.1967421814799309,
      "num_tokens": 103701152.0,
      "step": 51175
    },
    {
      "entropy": 5.752032327651977,
      "epoch": 4.472603338285415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031548099017725056,
      "loss": 5.2249,
      "mean_token_accuracy": 0.2128397434949875,
      "num_tokens": 103711852.0,
      "step": 51180
    },
    {
      "entropy": 5.72812933921814,
      "epoch": 4.47304028663812,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003154503371758392,
      "loss": 5.2474,
      "mean_token_accuracy": 0.2084866061806679,
      "num_tokens": 103722195.0,
      "step": 51185
    },
    {
      "entropy": 5.760050916671753,
      "epoch": 4.473477234990824,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003154196833985691,
      "loss": 5.2194,
      "mean_token_accuracy": 0.21157661080360413,
      "num_tokens": 103732575.0,
      "step": 51190
    },
    {
      "entropy": 5.735042762756348,
      "epoch": 4.4739141833435285,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031538902884602844,
      "loss": 5.1785,
      "mean_token_accuracy": 0.21136951595544815,
      "num_tokens": 103741837.0,
      "step": 51195
    },
    {
      "entropy": 5.807127857208252,
      "epoch": 4.474351131696234,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031535837351880493,
      "loss": 5.283,
      "mean_token_accuracy": 0.21032540649175643,
      "num_tokens": 103753390.0,
      "step": 51200
    },
    {
      "entropy": 5.858497762680054,
      "epoch": 4.474788080048938,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003153277174174867,
      "loss": 5.2937,
      "mean_token_accuracy": 0.20106059461832046,
      "num_tokens": 103762286.0,
      "step": 51205
    },
    {
      "entropy": 5.747781467437744,
      "epoch": 4.475225028401643,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003152970605426618,
      "loss": 5.2572,
      "mean_token_accuracy": 0.20496280938386918,
      "num_tokens": 103771962.0,
      "step": 51210
    },
    {
      "entropy": 5.766908073425293,
      "epoch": 4.475661976754347,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031526640289491803,
      "loss": 5.2071,
      "mean_token_accuracy": 0.21058274954557418,
      "num_tokens": 103782213.0,
      "step": 51215
    },
    {
      "entropy": 5.710809659957886,
      "epoch": 4.476098925107053,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003152357444748436,
      "loss": 5.2098,
      "mean_token_accuracy": 0.20743173807859422,
      "num_tokens": 103792848.0,
      "step": 51220
    },
    {
      "entropy": 5.7400623798370365,
      "epoch": 4.476535873459757,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00031520508528302656,
      "loss": 5.1666,
      "mean_token_accuracy": 0.2097577139735222,
      "num_tokens": 103803162.0,
      "step": 51225
    },
    {
      "entropy": 5.662403964996338,
      "epoch": 4.476972821812462,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003151744253200548,
      "loss": 5.0903,
      "mean_token_accuracy": 0.2212464064359665,
      "num_tokens": 103813335.0,
      "step": 51230
    },
    {
      "entropy": 5.71158275604248,
      "epoch": 4.477409770165167,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031514376458651665,
      "loss": 5.13,
      "mean_token_accuracy": 0.21851998418569565,
      "num_tokens": 103823135.0,
      "step": 51235
    },
    {
      "entropy": 5.775790786743164,
      "epoch": 4.477846718517871,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000315113103083,
      "loss": 5.2171,
      "mean_token_accuracy": 0.20861642807722092,
      "num_tokens": 103833178.0,
      "step": 51240
    },
    {
      "entropy": 5.7121538639068605,
      "epoch": 4.478283666870576,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031508244081009297,
      "loss": 5.0841,
      "mean_token_accuracy": 0.2265514075756073,
      "num_tokens": 103843742.0,
      "step": 51245
    },
    {
      "entropy": 5.754175758361816,
      "epoch": 4.47872061522328,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031505177776838377,
      "loss": 5.2992,
      "mean_token_accuracy": 0.20739645957946778,
      "num_tokens": 103853287.0,
      "step": 51250
    },
    {
      "entropy": 5.757966470718384,
      "epoch": 4.479157563575986,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031502111395846047,
      "loss": 5.1971,
      "mean_token_accuracy": 0.21192837953567506,
      "num_tokens": 103863522.0,
      "step": 51255
    },
    {
      "entropy": 5.73951530456543,
      "epoch": 4.47959451192869,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031499044938091124,
      "loss": 5.2259,
      "mean_token_accuracy": 0.21178605258464814,
      "num_tokens": 103873749.0,
      "step": 51260
    },
    {
      "entropy": 5.625480031967163,
      "epoch": 4.480031460281395,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031495978403632427,
      "loss": 5.1551,
      "mean_token_accuracy": 0.21807155311107634,
      "num_tokens": 103884055.0,
      "step": 51265
    },
    {
      "entropy": 5.77354211807251,
      "epoch": 4.480468408634099,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003149291179252877,
      "loss": 5.2335,
      "mean_token_accuracy": 0.2166660711169243,
      "num_tokens": 103893242.0,
      "step": 51270
    },
    {
      "entropy": 5.855956649780273,
      "epoch": 4.4809053569868045,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031489845104838964,
      "loss": 5.2578,
      "mean_token_accuracy": 0.2144496202468872,
      "num_tokens": 103902952.0,
      "step": 51275
    },
    {
      "entropy": 5.783125686645508,
      "epoch": 4.481342305339509,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003148677834062184,
      "loss": 5.1807,
      "mean_token_accuracy": 0.21468752026557922,
      "num_tokens": 103913659.0,
      "step": 51280
    },
    {
      "entropy": 5.761524248123169,
      "epoch": 4.481779253692213,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003148371149993623,
      "loss": 5.2174,
      "mean_token_accuracy": 0.20695524513721467,
      "num_tokens": 103923201.0,
      "step": 51285
    },
    {
      "entropy": 5.752075958251953,
      "epoch": 4.482216202044918,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00031480644582840934,
      "loss": 5.2516,
      "mean_token_accuracy": 0.20660108476877212,
      "num_tokens": 103933996.0,
      "step": 51290
    },
    {
      "entropy": 5.726119041442871,
      "epoch": 4.482653150397623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003147757758939479,
      "loss": 5.1537,
      "mean_token_accuracy": 0.21259852051734923,
      "num_tokens": 103943404.0,
      "step": 51295
    },
    {
      "entropy": 5.754104709625244,
      "epoch": 4.483090098750328,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003147451051965663,
      "loss": 5.2205,
      "mean_token_accuracy": 0.20963830649852752,
      "num_tokens": 103952669.0,
      "step": 51300
    },
    {
      "entropy": 5.733058881759644,
      "epoch": 4.483527047103032,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031471443373685276,
      "loss": 5.1314,
      "mean_token_accuracy": 0.22123482525348664,
      "num_tokens": 103963138.0,
      "step": 51305
    },
    {
      "entropy": 5.763264179229736,
      "epoch": 4.4839639954557375,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003146837615153955,
      "loss": 5.1901,
      "mean_token_accuracy": 0.201974356174469,
      "num_tokens": 103972268.0,
      "step": 51310
    },
    {
      "entropy": 5.735763072967529,
      "epoch": 4.484400943808442,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031465308853278296,
      "loss": 5.3045,
      "mean_token_accuracy": 0.2027178540825844,
      "num_tokens": 103983536.0,
      "step": 51315
    },
    {
      "entropy": 5.758023977279663,
      "epoch": 4.4848378921611465,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031462241478960335,
      "loss": 5.2315,
      "mean_token_accuracy": 0.2164178565144539,
      "num_tokens": 103993402.0,
      "step": 51320
    },
    {
      "entropy": 5.731354427337647,
      "epoch": 4.485274840513851,
      "grad_norm": 1.125,
      "learning_rate": 0.0003145917402864451,
      "loss": 5.1844,
      "mean_token_accuracy": 0.21491057127714158,
      "num_tokens": 104003834.0,
      "step": 51325
    },
    {
      "entropy": 5.770574045181275,
      "epoch": 4.485711788866556,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003145610650238965,
      "loss": 5.2264,
      "mean_token_accuracy": 0.20472753345966338,
      "num_tokens": 104014608.0,
      "step": 51330
    },
    {
      "entropy": 5.797913360595703,
      "epoch": 4.486148737219261,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000314530389002546,
      "loss": 5.2036,
      "mean_token_accuracy": 0.2054375723004341,
      "num_tokens": 104024826.0,
      "step": 51335
    },
    {
      "entropy": 5.79094820022583,
      "epoch": 4.486585685571965,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031449971222298186,
      "loss": 5.299,
      "mean_token_accuracy": 0.20436974614858627,
      "num_tokens": 104035108.0,
      "step": 51340
    },
    {
      "entropy": 5.698137521743774,
      "epoch": 4.48702263392467,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031446903468579265,
      "loss": 5.1718,
      "mean_token_accuracy": 0.21385787278413773,
      "num_tokens": 104045402.0,
      "step": 51345
    },
    {
      "entropy": 5.743751907348633,
      "epoch": 4.487459582277375,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003144383563915666,
      "loss": 5.1979,
      "mean_token_accuracy": 0.209719480574131,
      "num_tokens": 104055023.0,
      "step": 51350
    },
    {
      "entropy": 5.794718551635742,
      "epoch": 4.4878965306300795,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031440767734089214,
      "loss": 5.2815,
      "mean_token_accuracy": 0.21256355196237564,
      "num_tokens": 104065450.0,
      "step": 51355
    },
    {
      "entropy": 5.795553731918335,
      "epoch": 4.488333478982784,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003143769975343579,
      "loss": 5.2639,
      "mean_token_accuracy": 0.2041744500398636,
      "num_tokens": 104076601.0,
      "step": 51360
    },
    {
      "entropy": 5.709704637527466,
      "epoch": 4.4887704273354885,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003143463169725521,
      "loss": 5.19,
      "mean_token_accuracy": 0.21380458325147628,
      "num_tokens": 104086931.0,
      "step": 51365
    },
    {
      "entropy": 5.711650609970093,
      "epoch": 4.489207375688194,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031431563565606334,
      "loss": 5.175,
      "mean_token_accuracy": 0.21923891007900237,
      "num_tokens": 104096930.0,
      "step": 51370
    },
    {
      "entropy": 5.696463632583618,
      "epoch": 4.489644324040898,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003142849535854802,
      "loss": 5.2202,
      "mean_token_accuracy": 0.21434115022420883,
      "num_tokens": 104107645.0,
      "step": 51375
    },
    {
      "entropy": 5.733732223510742,
      "epoch": 4.490081272393603,
      "grad_norm": 1.0,
      "learning_rate": 0.00031425427076139093,
      "loss": 5.1816,
      "mean_token_accuracy": 0.21471721678972244,
      "num_tokens": 104118217.0,
      "step": 51380
    },
    {
      "entropy": 5.748183488845825,
      "epoch": 4.490518220746308,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031422358718438426,
      "loss": 5.1785,
      "mean_token_accuracy": 0.2141368344426155,
      "num_tokens": 104127933.0,
      "step": 51385
    },
    {
      "entropy": 5.716896915435791,
      "epoch": 4.490955169099013,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003141929028550486,
      "loss": 5.1287,
      "mean_token_accuracy": 0.2125071942806244,
      "num_tokens": 104138025.0,
      "step": 51390
    },
    {
      "entropy": 5.674310302734375,
      "epoch": 4.491392117451717,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031416221777397257,
      "loss": 5.1571,
      "mean_token_accuracy": 0.21815485805273055,
      "num_tokens": 104147969.0,
      "step": 51395
    },
    {
      "entropy": 5.7560937881469725,
      "epoch": 4.4918290658044215,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003141315319417447,
      "loss": 5.2698,
      "mean_token_accuracy": 0.19878187477588655,
      "num_tokens": 104158245.0,
      "step": 51400
    },
    {
      "entropy": 5.722480821609497,
      "epoch": 4.492266014157127,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003141008453589535,
      "loss": 5.2148,
      "mean_token_accuracy": 0.20704216957092286,
      "num_tokens": 104168470.0,
      "step": 51405
    },
    {
      "entropy": 5.781161975860596,
      "epoch": 4.492702962509831,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003140701580261875,
      "loss": 5.2169,
      "mean_token_accuracy": 0.21208655685186387,
      "num_tokens": 104178198.0,
      "step": 51410
    },
    {
      "entropy": 5.786324405670166,
      "epoch": 4.493139910862536,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003140394699440355,
      "loss": 5.2673,
      "mean_token_accuracy": 0.2042618215084076,
      "num_tokens": 104188801.0,
      "step": 51415
    },
    {
      "entropy": 5.807389879226685,
      "epoch": 4.49357685921524,
      "grad_norm": 1.015625,
      "learning_rate": 0.000314008781113086,
      "loss": 5.2782,
      "mean_token_accuracy": 0.2031669184565544,
      "num_tokens": 104199871.0,
      "step": 51420
    },
    {
      "entropy": 5.701415157318115,
      "epoch": 4.494013807567946,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003139780915339277,
      "loss": 5.1743,
      "mean_token_accuracy": 0.21233091950416566,
      "num_tokens": 104209865.0,
      "step": 51425
    },
    {
      "entropy": 5.7159195899963375,
      "epoch": 4.49445075592065,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003139474012071491,
      "loss": 5.2083,
      "mean_token_accuracy": 0.20875398963689804,
      "num_tokens": 104220375.0,
      "step": 51430
    },
    {
      "entropy": 5.709240245819092,
      "epoch": 4.494887704273355,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000313916710133339,
      "loss": 5.1566,
      "mean_token_accuracy": 0.21455801874399186,
      "num_tokens": 104230220.0,
      "step": 51435
    },
    {
      "entropy": 5.716312026977539,
      "epoch": 4.49532465262606,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003138860183130859,
      "loss": 5.2031,
      "mean_token_accuracy": 0.20910135954618453,
      "num_tokens": 104241071.0,
      "step": 51440
    },
    {
      "entropy": 5.747757196426392,
      "epoch": 4.495761600978764,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003138553257469787,
      "loss": 5.1885,
      "mean_token_accuracy": 0.2108457624912262,
      "num_tokens": 104251450.0,
      "step": 51445
    },
    {
      "entropy": 5.699952888488769,
      "epoch": 4.496198549331469,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031382463243560593,
      "loss": 5.1982,
      "mean_token_accuracy": 0.2108417272567749,
      "num_tokens": 104261898.0,
      "step": 51450
    },
    {
      "entropy": 5.7401293277740475,
      "epoch": 4.496635497684173,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003137939383795564,
      "loss": 5.1645,
      "mean_token_accuracy": 0.21796122342348098,
      "num_tokens": 104270694.0,
      "step": 51455
    },
    {
      "entropy": 5.7553609848022464,
      "epoch": 4.497072446036879,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031376324357941873,
      "loss": 5.2496,
      "mean_token_accuracy": 0.20300917625427245,
      "num_tokens": 104280755.0,
      "step": 51460
    },
    {
      "entropy": 5.7299493789672855,
      "epoch": 4.497509394389583,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031373254803578185,
      "loss": 5.253,
      "mean_token_accuracy": 0.20743947178125383,
      "num_tokens": 104290349.0,
      "step": 51465
    },
    {
      "entropy": 5.706475877761841,
      "epoch": 4.497946342742288,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031370185174923435,
      "loss": 5.1744,
      "mean_token_accuracy": 0.213775072991848,
      "num_tokens": 104301343.0,
      "step": 51470
    },
    {
      "entropy": 5.747866916656494,
      "epoch": 4.498383291094992,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000313671154720365,
      "loss": 5.2245,
      "mean_token_accuracy": 0.21210044473409653,
      "num_tokens": 104311678.0,
      "step": 51475
    },
    {
      "entropy": 5.783815240859985,
      "epoch": 4.4988202394476975,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003136404569497627,
      "loss": 5.2414,
      "mean_token_accuracy": 0.20798373818397523,
      "num_tokens": 104321936.0,
      "step": 51480
    },
    {
      "entropy": 5.801916170120239,
      "epoch": 4.499257187800402,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003136097584380162,
      "loss": 5.2729,
      "mean_token_accuracy": 0.200654798746109,
      "num_tokens": 104332281.0,
      "step": 51485
    },
    {
      "entropy": 5.793176460266113,
      "epoch": 4.499694136153106,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003135790591857143,
      "loss": 5.1816,
      "mean_token_accuracy": 0.2122051626443863,
      "num_tokens": 104341977.0,
      "step": 51490
    },
    {
      "entropy": 5.829136991500855,
      "epoch": 4.500131084505812,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031354835919344574,
      "loss": 5.3374,
      "mean_token_accuracy": 0.19553779512643815,
      "num_tokens": 104353707.0,
      "step": 51495
    },
    {
      "entropy": 5.826704454421997,
      "epoch": 4.500568032858516,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003135176584617996,
      "loss": 5.2963,
      "mean_token_accuracy": 0.20091818124055863,
      "num_tokens": 104363754.0,
      "step": 51500
    },
    {
      "entropy": 5.782478094100952,
      "epoch": 4.501004981211221,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003134869569913645,
      "loss": 5.2331,
      "mean_token_accuracy": 0.21277016997337342,
      "num_tokens": 104373229.0,
      "step": 51505
    },
    {
      "entropy": 5.697209262847901,
      "epoch": 4.501441929563925,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031345625478272935,
      "loss": 5.1547,
      "mean_token_accuracy": 0.2183019697666168,
      "num_tokens": 104384142.0,
      "step": 51510
    },
    {
      "entropy": 5.760156345367432,
      "epoch": 4.5018788779166306,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003134255518364832,
      "loss": 5.2033,
      "mean_token_accuracy": 0.2119467630982399,
      "num_tokens": 104394649.0,
      "step": 51515
    },
    {
      "entropy": 5.783554410934448,
      "epoch": 4.502315826269335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003133948481532148,
      "loss": 5.2077,
      "mean_token_accuracy": 0.21166670173406602,
      "num_tokens": 104405243.0,
      "step": 51520
    },
    {
      "entropy": 5.745669221878051,
      "epoch": 4.5027527746220395,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000313364143733513,
      "loss": 5.2068,
      "mean_token_accuracy": 0.2179347828030586,
      "num_tokens": 104415628.0,
      "step": 51525
    },
    {
      "entropy": 5.697147464752197,
      "epoch": 4.503189722974744,
      "grad_norm": 1.125,
      "learning_rate": 0.0003133334385779669,
      "loss": 5.1515,
      "mean_token_accuracy": 0.2172574833035469,
      "num_tokens": 104425137.0,
      "step": 51530
    },
    {
      "entropy": 5.677600717544555,
      "epoch": 4.503626671327449,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031330273268716534,
      "loss": 5.0993,
      "mean_token_accuracy": 0.22319612503051758,
      "num_tokens": 104435654.0,
      "step": 51535
    },
    {
      "entropy": 5.710723495483398,
      "epoch": 4.504063619680154,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031327202606169736,
      "loss": 5.2289,
      "mean_token_accuracy": 0.20279073119163513,
      "num_tokens": 104446050.0,
      "step": 51540
    },
    {
      "entropy": 5.724100780487061,
      "epoch": 4.504500568032858,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031324131870215176,
      "loss": 5.2098,
      "mean_token_accuracy": 0.21144257336854935,
      "num_tokens": 104456681.0,
      "step": 51545
    },
    {
      "entropy": 5.765711975097656,
      "epoch": 4.504937516385564,
      "grad_norm": 1.125,
      "learning_rate": 0.0003132106106091177,
      "loss": 5.2427,
      "mean_token_accuracy": 0.2042088270187378,
      "num_tokens": 104467540.0,
      "step": 51550
    },
    {
      "entropy": 5.7214189052581785,
      "epoch": 4.505374464738268,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003131799017831841,
      "loss": 5.2065,
      "mean_token_accuracy": 0.20954359024763108,
      "num_tokens": 104478072.0,
      "step": 51555
    },
    {
      "entropy": 5.76963620185852,
      "epoch": 4.5058114130909726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031314919222493994,
      "loss": 5.1232,
      "mean_token_accuracy": 0.22271706461906432,
      "num_tokens": 104487741.0,
      "step": 51560
    },
    {
      "entropy": 5.753550624847412,
      "epoch": 4.506248361443677,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031311848193497433,
      "loss": 5.2556,
      "mean_token_accuracy": 0.20910575687885286,
      "num_tokens": 104497481.0,
      "step": 51565
    },
    {
      "entropy": 5.720451879501343,
      "epoch": 4.506685309796382,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003130877709138762,
      "loss": 5.209,
      "mean_token_accuracy": 0.21205708086490632,
      "num_tokens": 104507351.0,
      "step": 51570
    },
    {
      "entropy": 5.824541854858398,
      "epoch": 4.507122258149087,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003130570591622348,
      "loss": 5.259,
      "mean_token_accuracy": 0.20556556880474092,
      "num_tokens": 104517625.0,
      "step": 51575
    },
    {
      "entropy": 5.780755138397216,
      "epoch": 4.507559206501791,
      "grad_norm": 1.0,
      "learning_rate": 0.0003130263466806389,
      "loss": 5.2684,
      "mean_token_accuracy": 0.2120634689927101,
      "num_tokens": 104528467.0,
      "step": 51580
    },
    {
      "entropy": 5.756224536895752,
      "epoch": 4.507996154854496,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031299563346967786,
      "loss": 5.2422,
      "mean_token_accuracy": 0.20651399940252305,
      "num_tokens": 104538606.0,
      "step": 51585
    },
    {
      "entropy": 5.79321060180664,
      "epoch": 4.508433103207201,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031296491952994063,
      "loss": 5.2674,
      "mean_token_accuracy": 0.20267854034900665,
      "num_tokens": 104548963.0,
      "step": 51590
    },
    {
      "entropy": 5.755841159820557,
      "epoch": 4.508870051559906,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003129342048620163,
      "loss": 5.219,
      "mean_token_accuracy": 0.2093428060412407,
      "num_tokens": 104558844.0,
      "step": 51595
    },
    {
      "entropy": 5.769850063323974,
      "epoch": 4.50930699991261,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031290348946649404,
      "loss": 5.2475,
      "mean_token_accuracy": 0.2073484927415848,
      "num_tokens": 104569334.0,
      "step": 51600
    },
    {
      "entropy": 5.7869648933410645,
      "epoch": 4.509743948265315,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003128727733439631,
      "loss": 5.256,
      "mean_token_accuracy": 0.2005597770214081,
      "num_tokens": 104579478.0,
      "step": 51605
    },
    {
      "entropy": 5.770999908447266,
      "epoch": 4.51018089661802,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003128420564950124,
      "loss": 5.2132,
      "mean_token_accuracy": 0.20750879496335983,
      "num_tokens": 104589963.0,
      "step": 51610
    },
    {
      "entropy": 5.716558361053467,
      "epoch": 4.510617844970724,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003128113389202313,
      "loss": 5.1567,
      "mean_token_accuracy": 0.21326869130134582,
      "num_tokens": 104599600.0,
      "step": 51615
    },
    {
      "entropy": 5.741253423690796,
      "epoch": 4.511054793323429,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003127806206202089,
      "loss": 5.1568,
      "mean_token_accuracy": 0.2172021448612213,
      "num_tokens": 104608771.0,
      "step": 51620
    },
    {
      "entropy": 5.732288455963134,
      "epoch": 4.511491741676134,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003127499015955344,
      "loss": 5.1875,
      "mean_token_accuracy": 0.2112699866294861,
      "num_tokens": 104619257.0,
      "step": 51625
    },
    {
      "entropy": 5.744142007827759,
      "epoch": 4.511928690028839,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000312719181846797,
      "loss": 5.1405,
      "mean_token_accuracy": 0.2138466104865074,
      "num_tokens": 104628493.0,
      "step": 51630
    },
    {
      "entropy": 5.727453279495239,
      "epoch": 4.512365638381543,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003126884613745859,
      "loss": 5.1664,
      "mean_token_accuracy": 0.21720887571573258,
      "num_tokens": 104637901.0,
      "step": 51635
    },
    {
      "entropy": 5.7400397777557375,
      "epoch": 4.512802586734248,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031265774017949033,
      "loss": 5.1583,
      "mean_token_accuracy": 0.21634956151247026,
      "num_tokens": 104647104.0,
      "step": 51640
    },
    {
      "entropy": 5.765585327148438,
      "epoch": 4.513239535086953,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003126270182620997,
      "loss": 5.1327,
      "mean_token_accuracy": 0.2130614310503006,
      "num_tokens": 104656125.0,
      "step": 51645
    },
    {
      "entropy": 5.718589925765992,
      "epoch": 4.513676483439657,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031259629562300297,
      "loss": 5.238,
      "mean_token_accuracy": 0.21672221273183823,
      "num_tokens": 104666216.0,
      "step": 51650
    },
    {
      "entropy": 5.788166093826294,
      "epoch": 4.514113431792362,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003125655722627898,
      "loss": 5.2523,
      "mean_token_accuracy": 0.20704320818185806,
      "num_tokens": 104675239.0,
      "step": 51655
    },
    {
      "entropy": 5.7351466655731205,
      "epoch": 4.514550380145067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031253484818204906,
      "loss": 5.2172,
      "mean_token_accuracy": 0.20996888279914855,
      "num_tokens": 104686324.0,
      "step": 51660
    },
    {
      "entropy": 5.778326416015625,
      "epoch": 4.514987328497772,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003125041233813704,
      "loss": 5.2639,
      "mean_token_accuracy": 0.2015189915895462,
      "num_tokens": 104696163.0,
      "step": 51665
    },
    {
      "entropy": 5.72471251487732,
      "epoch": 4.515424276850476,
      "grad_norm": 1.0625,
      "learning_rate": 0.000312473397861343,
      "loss": 5.1832,
      "mean_token_accuracy": 0.2090858742594719,
      "num_tokens": 104707126.0,
      "step": 51670
    },
    {
      "entropy": 5.73780460357666,
      "epoch": 4.515861225203181,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031244267162255614,
      "loss": 5.1897,
      "mean_token_accuracy": 0.21113371700048447,
      "num_tokens": 104718052.0,
      "step": 51675
    },
    {
      "entropy": 5.680583524703979,
      "epoch": 4.516298173555886,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00031241194466559923,
      "loss": 5.1559,
      "mean_token_accuracy": 0.2178647369146347,
      "num_tokens": 104727956.0,
      "step": 51680
    },
    {
      "entropy": 5.779957485198975,
      "epoch": 4.5167351219085905,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003123812169910616,
      "loss": 5.212,
      "mean_token_accuracy": 0.21160620152950288,
      "num_tokens": 104738258.0,
      "step": 51685
    },
    {
      "entropy": 5.776462078094482,
      "epoch": 4.517172070261295,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003123504885995327,
      "loss": 5.2168,
      "mean_token_accuracy": 0.20899940729141236,
      "num_tokens": 104748608.0,
      "step": 51690
    },
    {
      "entropy": 5.7504229068756105,
      "epoch": 4.517609018613999,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003123197594916018,
      "loss": 5.2056,
      "mean_token_accuracy": 0.21789100766181946,
      "num_tokens": 104758410.0,
      "step": 51695
    },
    {
      "entropy": 5.77942123413086,
      "epoch": 4.518045966966705,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003122890296678584,
      "loss": 5.2499,
      "mean_token_accuracy": 0.20868034809827804,
      "num_tokens": 104768657.0,
      "step": 51700
    },
    {
      "entropy": 5.80675368309021,
      "epoch": 4.518482915319409,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031225829912889184,
      "loss": 5.2979,
      "mean_token_accuracy": 0.20231283754110335,
      "num_tokens": 104779053.0,
      "step": 51705
    },
    {
      "entropy": 5.74366455078125,
      "epoch": 4.518919863672114,
      "grad_norm": 1.125,
      "learning_rate": 0.00031222756787529154,
      "loss": 5.1249,
      "mean_token_accuracy": 0.21598464399576187,
      "num_tokens": 104788652.0,
      "step": 51710
    },
    {
      "entropy": 5.734405899047852,
      "epoch": 4.519356812024819,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003121968359076471,
      "loss": 5.2088,
      "mean_token_accuracy": 0.21240859776735305,
      "num_tokens": 104798757.0,
      "step": 51715
    },
    {
      "entropy": 5.756525993347168,
      "epoch": 4.519793760377524,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031216610322654776,
      "loss": 5.1982,
      "mean_token_accuracy": 0.21156946569681168,
      "num_tokens": 104808174.0,
      "step": 51720
    },
    {
      "entropy": 5.7435017108917235,
      "epoch": 4.520230708730228,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003121353698325831,
      "loss": 5.1682,
      "mean_token_accuracy": 0.2181831955909729,
      "num_tokens": 104817509.0,
      "step": 51725
    },
    {
      "entropy": 5.68941559791565,
      "epoch": 4.5206676570829325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003121046357263426,
      "loss": 5.1318,
      "mean_token_accuracy": 0.22614143043756485,
      "num_tokens": 104827518.0,
      "step": 51730
    },
    {
      "entropy": 5.739877986907959,
      "epoch": 4.521104605435638,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003120739009084158,
      "loss": 5.2278,
      "mean_token_accuracy": 0.2106218621134758,
      "num_tokens": 104837868.0,
      "step": 51735
    },
    {
      "entropy": 5.761039543151855,
      "epoch": 4.521541553788342,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031204316537939206,
      "loss": 5.2545,
      "mean_token_accuracy": 0.20525452196598054,
      "num_tokens": 104847849.0,
      "step": 51740
    },
    {
      "entropy": 5.729694938659668,
      "epoch": 4.521978502141047,
      "grad_norm": 1.125,
      "learning_rate": 0.000312012429139861,
      "loss": 5.2112,
      "mean_token_accuracy": 0.21091946661472322,
      "num_tokens": 104857751.0,
      "step": 51745
    },
    {
      "entropy": 5.754829692840576,
      "epoch": 4.522415450493751,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003119816921904122,
      "loss": 5.1524,
      "mean_token_accuracy": 0.21541061550378798,
      "num_tokens": 104867092.0,
      "step": 51750
    },
    {
      "entropy": 5.819614315032959,
      "epoch": 4.522852398846457,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003119509545316352,
      "loss": 5.3067,
      "mean_token_accuracy": 0.2051269069314003,
      "num_tokens": 104876791.0,
      "step": 51755
    },
    {
      "entropy": 5.732180547714234,
      "epoch": 4.523289347199161,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031192021616411946,
      "loss": 5.234,
      "mean_token_accuracy": 0.21281500905752182,
      "num_tokens": 104886715.0,
      "step": 51760
    },
    {
      "entropy": 5.798616123199463,
      "epoch": 4.523726295551866,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031188947708845466,
      "loss": 5.2901,
      "mean_token_accuracy": 0.20708574801683427,
      "num_tokens": 104898204.0,
      "step": 51765
    },
    {
      "entropy": 5.782123279571533,
      "epoch": 4.524163243904571,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003118587373052304,
      "loss": 5.1857,
      "mean_token_accuracy": 0.21613783538341522,
      "num_tokens": 104908875.0,
      "step": 51770
    },
    {
      "entropy": 5.68035888671875,
      "epoch": 4.524600192257275,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031182799681503615,
      "loss": 5.1551,
      "mean_token_accuracy": 0.2182004451751709,
      "num_tokens": 104918487.0,
      "step": 51775
    },
    {
      "entropy": 5.75933313369751,
      "epoch": 4.52503714060998,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003117972556184617,
      "loss": 5.2407,
      "mean_token_accuracy": 0.20756561756134034,
      "num_tokens": 104928899.0,
      "step": 51780
    },
    {
      "entropy": 5.777212333679199,
      "epoch": 4.525474088962684,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003117665137160966,
      "loss": 5.219,
      "mean_token_accuracy": 0.20658042132854462,
      "num_tokens": 104939728.0,
      "step": 51785
    },
    {
      "entropy": 5.800076532363891,
      "epoch": 4.52591103731539,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003117357711085305,
      "loss": 5.2637,
      "mean_token_accuracy": 0.20745507776737213,
      "num_tokens": 104949181.0,
      "step": 51790
    },
    {
      "entropy": 5.751013565063476,
      "epoch": 4.526347985668094,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031170502779635296,
      "loss": 5.2673,
      "mean_token_accuracy": 0.2040533035993576,
      "num_tokens": 104959014.0,
      "step": 51795
    },
    {
      "entropy": 5.77334246635437,
      "epoch": 4.526784934020799,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031167428378015396,
      "loss": 5.2365,
      "mean_token_accuracy": 0.20954794436693192,
      "num_tokens": 104970063.0,
      "step": 51800
    },
    {
      "entropy": 5.716647005081176,
      "epoch": 4.527221882373503,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031164353906052285,
      "loss": 5.155,
      "mean_token_accuracy": 0.2163340002298355,
      "num_tokens": 104979452.0,
      "step": 51805
    },
    {
      "entropy": 5.731733131408691,
      "epoch": 4.5276588307262084,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003116127936380494,
      "loss": 5.2033,
      "mean_token_accuracy": 0.21135183274745942,
      "num_tokens": 104989647.0,
      "step": 51810
    },
    {
      "entropy": 5.728695201873779,
      "epoch": 4.528095779078913,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031158204751332343,
      "loss": 5.1159,
      "mean_token_accuracy": 0.2182372897863388,
      "num_tokens": 105000148.0,
      "step": 51815
    },
    {
      "entropy": 5.802879905700683,
      "epoch": 4.528532727431617,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031155130068693463,
      "loss": 5.249,
      "mean_token_accuracy": 0.20041414648294448,
      "num_tokens": 105012057.0,
      "step": 51820
    },
    {
      "entropy": 5.756858682632446,
      "epoch": 4.528969675784323,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031152055315947273,
      "loss": 5.2029,
      "mean_token_accuracy": 0.21282915621995926,
      "num_tokens": 105022074.0,
      "step": 51825
    },
    {
      "entropy": 5.798188877105713,
      "epoch": 4.529406624137027,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003114898049315275,
      "loss": 5.2624,
      "mean_token_accuracy": 0.20726867467164994,
      "num_tokens": 105032589.0,
      "step": 51830
    },
    {
      "entropy": 5.745664167404175,
      "epoch": 4.529843572489732,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00031145905600368865,
      "loss": 5.2289,
      "mean_token_accuracy": 0.21334016919136048,
      "num_tokens": 105043592.0,
      "step": 51835
    },
    {
      "entropy": 5.72345232963562,
      "epoch": 4.530280520842436,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003114283063765461,
      "loss": 5.135,
      "mean_token_accuracy": 0.21694294661283492,
      "num_tokens": 105053804.0,
      "step": 51840
    },
    {
      "entropy": 5.721292400360108,
      "epoch": 4.5307174691951415,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003113975560506895,
      "loss": 5.2223,
      "mean_token_accuracy": 0.20863345712423326,
      "num_tokens": 105063341.0,
      "step": 51845
    },
    {
      "entropy": 5.7559561252594,
      "epoch": 4.531154417547846,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003113668050267086,
      "loss": 5.2685,
      "mean_token_accuracy": 0.21092299818992616,
      "num_tokens": 105075228.0,
      "step": 51850
    },
    {
      "entropy": 5.738774919509888,
      "epoch": 4.5315913659005504,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031133605330519343,
      "loss": 5.1202,
      "mean_token_accuracy": 0.23061641603708266,
      "num_tokens": 105085653.0,
      "step": 51855
    },
    {
      "entropy": 5.823555612564087,
      "epoch": 4.532028314253255,
      "grad_norm": 1.125,
      "learning_rate": 0.0003113053008867337,
      "loss": 5.2826,
      "mean_token_accuracy": 0.20761634409427643,
      "num_tokens": 105095969.0,
      "step": 51860
    },
    {
      "entropy": 5.772940254211425,
      "epoch": 4.53246526260596,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003112745477719193,
      "loss": 5.1652,
      "mean_token_accuracy": 0.2155867800116539,
      "num_tokens": 105105260.0,
      "step": 51865
    },
    {
      "entropy": 5.750512933731079,
      "epoch": 4.532902210958665,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031124379396134003,
      "loss": 5.1826,
      "mean_token_accuracy": 0.20862141400575637,
      "num_tokens": 105114749.0,
      "step": 51870
    },
    {
      "entropy": 5.7289622783660885,
      "epoch": 4.533339159311369,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031121303945558584,
      "loss": 5.2135,
      "mean_token_accuracy": 0.20970012843608857,
      "num_tokens": 105124520.0,
      "step": 51875
    },
    {
      "entropy": 5.791741609573364,
      "epoch": 4.533776107664075,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031118228425524653,
      "loss": 5.2445,
      "mean_token_accuracy": 0.21137310117483138,
      "num_tokens": 105133886.0,
      "step": 51880
    },
    {
      "entropy": 5.696262216567993,
      "epoch": 4.534213056016779,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031115152836091206,
      "loss": 5.0738,
      "mean_token_accuracy": 0.2273305296897888,
      "num_tokens": 105143781.0,
      "step": 51885
    },
    {
      "entropy": 5.728287982940674,
      "epoch": 4.5346500043694835,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003111207717731723,
      "loss": 5.2307,
      "mean_token_accuracy": 0.2120058313012123,
      "num_tokens": 105154383.0,
      "step": 51890
    },
    {
      "entropy": 5.745407915115356,
      "epoch": 4.535086952722188,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031109001449261726,
      "loss": 5.1614,
      "mean_token_accuracy": 0.20934998989105225,
      "num_tokens": 105163450.0,
      "step": 51895
    },
    {
      "entropy": 5.729206514358521,
      "epoch": 4.5355239010748924,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031105925651983687,
      "loss": 5.293,
      "mean_token_accuracy": 0.20113370716571807,
      "num_tokens": 105174046.0,
      "step": 51900
    },
    {
      "entropy": 5.741891384124756,
      "epoch": 4.535960849427598,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000311028497855421,
      "loss": 5.225,
      "mean_token_accuracy": 0.20624199211597444,
      "num_tokens": 105183527.0,
      "step": 51905
    },
    {
      "entropy": 5.764447975158691,
      "epoch": 4.536397797780302,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003109977384999596,
      "loss": 5.1719,
      "mean_token_accuracy": 0.21144520938396455,
      "num_tokens": 105194002.0,
      "step": 51910
    },
    {
      "entropy": 5.759985399246216,
      "epoch": 4.536834746133007,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003109669784540428,
      "loss": 5.2157,
      "mean_token_accuracy": 0.20677966475486756,
      "num_tokens": 105204722.0,
      "step": 51915
    },
    {
      "entropy": 5.765243911743164,
      "epoch": 4.537271694485712,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031093621771826046,
      "loss": 5.1407,
      "mean_token_accuracy": 0.21954849660396575,
      "num_tokens": 105214066.0,
      "step": 51920
    },
    {
      "entropy": 5.767229461669922,
      "epoch": 4.537708642838417,
      "grad_norm": 1.125,
      "learning_rate": 0.00031090545629320265,
      "loss": 5.24,
      "mean_token_accuracy": 0.20705857425928115,
      "num_tokens": 105224943.0,
      "step": 51925
    },
    {
      "entropy": 5.78437876701355,
      "epoch": 4.538145591191121,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00031087469417945935,
      "loss": 5.2645,
      "mean_token_accuracy": 0.2089986249804497,
      "num_tokens": 105235102.0,
      "step": 51930
    },
    {
      "entropy": 5.688948440551758,
      "epoch": 4.538582539543826,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003108439313776207,
      "loss": 5.1134,
      "mean_token_accuracy": 0.21346740424633026,
      "num_tokens": 105245505.0,
      "step": 51935
    },
    {
      "entropy": 5.758323097229004,
      "epoch": 4.539019487896531,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031081316788827664,
      "loss": 5.21,
      "mean_token_accuracy": 0.2095666915178299,
      "num_tokens": 105256386.0,
      "step": 51940
    },
    {
      "entropy": 5.76808009147644,
      "epoch": 4.539456436249235,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003107824037120172,
      "loss": 5.2275,
      "mean_token_accuracy": 0.21629583686590195,
      "num_tokens": 105266457.0,
      "step": 51945
    },
    {
      "entropy": 5.731579208374024,
      "epoch": 4.53989338460194,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003107516388494325,
      "loss": 5.1843,
      "mean_token_accuracy": 0.21539448201656342,
      "num_tokens": 105276558.0,
      "step": 51950
    },
    {
      "entropy": 5.716495132446289,
      "epoch": 4.540330332954644,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031072087330111264,
      "loss": 5.194,
      "mean_token_accuracy": 0.20745036005973816,
      "num_tokens": 105285619.0,
      "step": 51955
    },
    {
      "entropy": 5.706608009338379,
      "epoch": 4.54076728130735,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003106901070676478,
      "loss": 5.0898,
      "mean_token_accuracy": 0.22070057541131974,
      "num_tokens": 105296269.0,
      "step": 51960
    },
    {
      "entropy": 5.7225751876831055,
      "epoch": 4.541204229660054,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031065934014962794,
      "loss": 5.2073,
      "mean_token_accuracy": 0.21446353942155838,
      "num_tokens": 105305498.0,
      "step": 51965
    },
    {
      "entropy": 5.683291482925415,
      "epoch": 4.541641178012759,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031062857254764327,
      "loss": 5.1026,
      "mean_token_accuracy": 0.2185037225484848,
      "num_tokens": 105316259.0,
      "step": 51970
    },
    {
      "entropy": 5.72829236984253,
      "epoch": 4.542078126365464,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031059780426228385,
      "loss": 5.1874,
      "mean_token_accuracy": 0.21743519008159637,
      "num_tokens": 105325802.0,
      "step": 51975
    },
    {
      "entropy": 5.772668504714966,
      "epoch": 4.542515074718168,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031056703529413994,
      "loss": 5.1514,
      "mean_token_accuracy": 0.21824404895305632,
      "num_tokens": 105335349.0,
      "step": 51980
    },
    {
      "entropy": 5.724466323852539,
      "epoch": 4.542952023070873,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003105362656438016,
      "loss": 5.2238,
      "mean_token_accuracy": 0.20830315053462983,
      "num_tokens": 105345706.0,
      "step": 51985
    },
    {
      "entropy": 5.70572304725647,
      "epoch": 4.543388971423578,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003105054953118591,
      "loss": 5.1356,
      "mean_token_accuracy": 0.21930849850177764,
      "num_tokens": 105356354.0,
      "step": 51990
    },
    {
      "entropy": 5.735745573043824,
      "epoch": 4.543825919776283,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031047472429890263,
      "loss": 5.1751,
      "mean_token_accuracy": 0.21446937769651414,
      "num_tokens": 105367133.0,
      "step": 51995
    },
    {
      "entropy": 5.8002299785614015,
      "epoch": 4.544262868128987,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003104439526055223,
      "loss": 5.2918,
      "mean_token_accuracy": 0.20708059519529343,
      "num_tokens": 105376480.0,
      "step": 52000
    },
    {
      "entropy": 5.729370212554931,
      "epoch": 4.544699816481692,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003104131802323084,
      "loss": 5.2231,
      "mean_token_accuracy": 0.20795288681983948,
      "num_tokens": 105386237.0,
      "step": 52005
    },
    {
      "entropy": 5.722386646270752,
      "epoch": 4.545136764834396,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031038240717985125,
      "loss": 5.1719,
      "mean_token_accuracy": 0.21280520111322404,
      "num_tokens": 105395687.0,
      "step": 52010
    },
    {
      "entropy": 5.752387237548828,
      "epoch": 4.5455737131871015,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003103516334487409,
      "loss": 5.1381,
      "mean_token_accuracy": 0.21317062377929688,
      "num_tokens": 105405465.0,
      "step": 52015
    },
    {
      "entropy": 5.805533409118652,
      "epoch": 4.546010661539806,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003103208590395676,
      "loss": 5.226,
      "mean_token_accuracy": 0.21617645025253296,
      "num_tokens": 105415234.0,
      "step": 52020
    },
    {
      "entropy": 5.7090325355529785,
      "epoch": 4.54644760989251,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003102900839529219,
      "loss": 5.1957,
      "mean_token_accuracy": 0.2137721821665764,
      "num_tokens": 105425773.0,
      "step": 52025
    },
    {
      "entropy": 5.777948617935181,
      "epoch": 4.546884558245216,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003102593081893938,
      "loss": 5.2738,
      "mean_token_accuracy": 0.20648631155490876,
      "num_tokens": 105436247.0,
      "step": 52030
    },
    {
      "entropy": 5.761267566680909,
      "epoch": 4.54732150659792,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031022853174957375,
      "loss": 5.1415,
      "mean_token_accuracy": 0.22004460990428926,
      "num_tokens": 105446227.0,
      "step": 52035
    },
    {
      "entropy": 5.756596183776855,
      "epoch": 4.547758454950625,
      "grad_norm": 1.234375,
      "learning_rate": 0.000310197754634052,
      "loss": 5.2596,
      "mean_token_accuracy": 0.20890765488147736,
      "num_tokens": 105456368.0,
      "step": 52040
    },
    {
      "entropy": 5.731490278244019,
      "epoch": 4.548195403303329,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003101669768434188,
      "loss": 5.241,
      "mean_token_accuracy": 0.20816704630851746,
      "num_tokens": 105465530.0,
      "step": 52045
    },
    {
      "entropy": 5.803909969329834,
      "epoch": 4.5486323516560345,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003101361983782646,
      "loss": 5.2471,
      "mean_token_accuracy": 0.20123950988054276,
      "num_tokens": 105475723.0,
      "step": 52050
    },
    {
      "entropy": 5.829819679260254,
      "epoch": 4.549069300008739,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031010541923917964,
      "loss": 5.2131,
      "mean_token_accuracy": 0.2144630879163742,
      "num_tokens": 105484810.0,
      "step": 52055
    },
    {
      "entropy": 5.762195444107055,
      "epoch": 4.5495062483614435,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031007463942675446,
      "loss": 5.1918,
      "mean_token_accuracy": 0.211574225127697,
      "num_tokens": 105494748.0,
      "step": 52060
    },
    {
      "entropy": 5.750945520401001,
      "epoch": 4.549943196714148,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003100438589415793,
      "loss": 5.2093,
      "mean_token_accuracy": 0.2155514806509018,
      "num_tokens": 105505752.0,
      "step": 52065
    },
    {
      "entropy": 5.755007314682007,
      "epoch": 4.550380145066853,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003100130777842445,
      "loss": 5.1739,
      "mean_token_accuracy": 0.21015915125608445,
      "num_tokens": 105515842.0,
      "step": 52070
    },
    {
      "entropy": 5.6983558177948,
      "epoch": 4.550817093419558,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003099822959553405,
      "loss": 5.1037,
      "mean_token_accuracy": 0.22237931340932846,
      "num_tokens": 105525786.0,
      "step": 52075
    },
    {
      "entropy": 5.73206467628479,
      "epoch": 4.551254041772262,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030995151345545777,
      "loss": 5.1657,
      "mean_token_accuracy": 0.21303317248821257,
      "num_tokens": 105535034.0,
      "step": 52080
    },
    {
      "entropy": 5.648330974578857,
      "epoch": 4.551690990124968,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003099207302851868,
      "loss": 5.2045,
      "mean_token_accuracy": 0.2140020713210106,
      "num_tokens": 105545313.0,
      "step": 52085
    },
    {
      "entropy": 5.860350465774536,
      "epoch": 4.552127938477672,
      "grad_norm": 1.25,
      "learning_rate": 0.00030988994644511774,
      "loss": 5.2726,
      "mean_token_accuracy": 0.20426394045352936,
      "num_tokens": 105555034.0,
      "step": 52090
    },
    {
      "entropy": 5.78893632888794,
      "epoch": 4.5525648868303765,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030985916193584137,
      "loss": 5.158,
      "mean_token_accuracy": 0.2157311260700226,
      "num_tokens": 105564694.0,
      "step": 52095
    },
    {
      "entropy": 5.714910364151001,
      "epoch": 4.553001835183081,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030982837675794787,
      "loss": 5.2578,
      "mean_token_accuracy": 0.21147603690624237,
      "num_tokens": 105575052.0,
      "step": 52100
    },
    {
      "entropy": 5.6966331005096436,
      "epoch": 4.553438783535786,
      "grad_norm": 0.984375,
      "learning_rate": 0.000309797590912028,
      "loss": 5.1641,
      "mean_token_accuracy": 0.22062388211488723,
      "num_tokens": 105585529.0,
      "step": 52105
    },
    {
      "entropy": 5.803965997695923,
      "epoch": 4.553875731888491,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003097668043986721,
      "loss": 5.2147,
      "mean_token_accuracy": 0.21085613816976548,
      "num_tokens": 105596464.0,
      "step": 52110
    },
    {
      "entropy": 5.829441785812378,
      "epoch": 4.554312680241195,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030973601721847054,
      "loss": 5.2074,
      "mean_token_accuracy": 0.20634986460208893,
      "num_tokens": 105605493.0,
      "step": 52115
    },
    {
      "entropy": 5.695996856689453,
      "epoch": 4.5547496285939,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000309705229372014,
      "loss": 5.2107,
      "mean_token_accuracy": 0.2126711517572403,
      "num_tokens": 105616265.0,
      "step": 52120
    },
    {
      "entropy": 5.698868179321289,
      "epoch": 4.555186576946605,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030967444085989304,
      "loss": 5.2138,
      "mean_token_accuracy": 0.21008913069963456,
      "num_tokens": 105625889.0,
      "step": 52125
    },
    {
      "entropy": 5.7680236339569095,
      "epoch": 4.55562352529931,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030964365168269816,
      "loss": 5.2535,
      "mean_token_accuracy": 0.20917645394802092,
      "num_tokens": 105635821.0,
      "step": 52130
    },
    {
      "entropy": 5.728595161437989,
      "epoch": 4.556060473652014,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030961286184101984,
      "loss": 5.2068,
      "mean_token_accuracy": 0.2099240764975548,
      "num_tokens": 105646787.0,
      "step": 52135
    },
    {
      "entropy": 5.748929262161255,
      "epoch": 4.556497422004719,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003095820713354487,
      "loss": 5.2209,
      "mean_token_accuracy": 0.21089154183864595,
      "num_tokens": 105658001.0,
      "step": 52140
    },
    {
      "entropy": 5.735169696807861,
      "epoch": 4.556934370357424,
      "grad_norm": 1.125,
      "learning_rate": 0.0003095512801665753,
      "loss": 5.2023,
      "mean_token_accuracy": 0.21023706197738648,
      "num_tokens": 105667543.0,
      "step": 52145
    },
    {
      "entropy": 5.78099799156189,
      "epoch": 4.557371318710128,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030952048833499026,
      "loss": 5.1938,
      "mean_token_accuracy": 0.20926388800144197,
      "num_tokens": 105677365.0,
      "step": 52150
    },
    {
      "entropy": 5.813456439971924,
      "epoch": 4.557808267062833,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003094896958412841,
      "loss": 5.1931,
      "mean_token_accuracy": 0.2089049905538559,
      "num_tokens": 105688241.0,
      "step": 52155
    },
    {
      "entropy": 5.735379409790039,
      "epoch": 4.558245215415538,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003094589026860476,
      "loss": 5.1371,
      "mean_token_accuracy": 0.21648413985967635,
      "num_tokens": 105697719.0,
      "step": 52160
    },
    {
      "entropy": 5.7226982593536375,
      "epoch": 4.558682163768243,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003094281088698712,
      "loss": 5.249,
      "mean_token_accuracy": 0.1991396352648735,
      "num_tokens": 105708268.0,
      "step": 52165
    },
    {
      "entropy": 5.808789539337158,
      "epoch": 4.559119112120947,
      "grad_norm": 1.0,
      "learning_rate": 0.0003093973143933456,
      "loss": 5.3162,
      "mean_token_accuracy": 0.2027392640709877,
      "num_tokens": 105718668.0,
      "step": 52170
    },
    {
      "entropy": 5.797886705398559,
      "epoch": 4.559556060473652,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030936651925706166,
      "loss": 5.2804,
      "mean_token_accuracy": 0.20784709900617598,
      "num_tokens": 105730904.0,
      "step": 52175
    },
    {
      "entropy": 5.684429073333741,
      "epoch": 4.559993008826357,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003093357234616098,
      "loss": 5.1305,
      "mean_token_accuracy": 0.21628961563110352,
      "num_tokens": 105742146.0,
      "step": 52180
    },
    {
      "entropy": 5.74501600265503,
      "epoch": 4.560429957179061,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003093049270075807,
      "loss": 5.2439,
      "mean_token_accuracy": 0.20970468521118163,
      "num_tokens": 105753218.0,
      "step": 52185
    },
    {
      "entropy": 5.725484466552734,
      "epoch": 4.560866905531766,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003092741298955651,
      "loss": 5.1263,
      "mean_token_accuracy": 0.2116120308637619,
      "num_tokens": 105763133.0,
      "step": 52190
    },
    {
      "entropy": 5.74096450805664,
      "epoch": 4.561303853884471,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030924333212615384,
      "loss": 5.2643,
      "mean_token_accuracy": 0.20589857697486877,
      "num_tokens": 105773810.0,
      "step": 52195
    },
    {
      "entropy": 5.774216413497925,
      "epoch": 4.561740802237176,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003092125336999375,
      "loss": 5.2143,
      "mean_token_accuracy": 0.21195551753044128,
      "num_tokens": 105783918.0,
      "step": 52200
    },
    {
      "entropy": 5.732764434814453,
      "epoch": 4.56217775058988,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003091817346175068,
      "loss": 5.1694,
      "mean_token_accuracy": 0.21581699848175048,
      "num_tokens": 105793799.0,
      "step": 52205
    },
    {
      "entropy": 5.743086624145508,
      "epoch": 4.562614698942585,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030915093487945256,
      "loss": 5.2132,
      "mean_token_accuracy": 0.20848000198602676,
      "num_tokens": 105804311.0,
      "step": 52210
    },
    {
      "entropy": 5.6841799259185795,
      "epoch": 4.56305164729529,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003091201344863655,
      "loss": 5.1903,
      "mean_token_accuracy": 0.2158287838101387,
      "num_tokens": 105814132.0,
      "step": 52215
    },
    {
      "entropy": 5.73252854347229,
      "epoch": 4.5634885956479945,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030908933343883635,
      "loss": 5.1917,
      "mean_token_accuracy": 0.21278193295001985,
      "num_tokens": 105824693.0,
      "step": 52220
    },
    {
      "entropy": 5.845236825942993,
      "epoch": 4.563925544000699,
      "grad_norm": 1.125,
      "learning_rate": 0.00030905853173745595,
      "loss": 5.3313,
      "mean_token_accuracy": 0.19810967296361923,
      "num_tokens": 105835024.0,
      "step": 52225
    },
    {
      "entropy": 5.750909757614136,
      "epoch": 4.564362492353403,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000309027729382815,
      "loss": 5.1815,
      "mean_token_accuracy": 0.21295072734355927,
      "num_tokens": 105844555.0,
      "step": 52230
    },
    {
      "entropy": 5.749283266067505,
      "epoch": 4.564799440706109,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003089969263755045,
      "loss": 5.2484,
      "mean_token_accuracy": 0.2127447873353958,
      "num_tokens": 105854408.0,
      "step": 52235
    },
    {
      "entropy": 5.7072672843933105,
      "epoch": 4.565236389058813,
      "grad_norm": 1.03125,
      "learning_rate": 0.000308966122716115,
      "loss": 5.1289,
      "mean_token_accuracy": 0.21772652715444565,
      "num_tokens": 105863491.0,
      "step": 52240
    },
    {
      "entropy": 5.735282897949219,
      "epoch": 4.565673337411518,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003089353184052376,
      "loss": 5.1589,
      "mean_token_accuracy": 0.21347038745880126,
      "num_tokens": 105873001.0,
      "step": 52245
    },
    {
      "entropy": 5.643208408355713,
      "epoch": 4.566110285764223,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030890451344346286,
      "loss": 5.1398,
      "mean_token_accuracy": 0.22039569169282913,
      "num_tokens": 105883200.0,
      "step": 52250
    },
    {
      "entropy": 5.720789957046509,
      "epoch": 4.5665472341169275,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030887370783138187,
      "loss": 5.2106,
      "mean_token_accuracy": 0.20726726055145264,
      "num_tokens": 105893558.0,
      "step": 52255
    },
    {
      "entropy": 5.7411760807037355,
      "epoch": 4.566984182469632,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003088429015695854,
      "loss": 5.2086,
      "mean_token_accuracy": 0.21217572093009948,
      "num_tokens": 105903945.0,
      "step": 52260
    },
    {
      "entropy": 5.784400892257691,
      "epoch": 4.5674211308223365,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003088120946586643,
      "loss": 5.2045,
      "mean_token_accuracy": 0.2098671391606331,
      "num_tokens": 105915028.0,
      "step": 52265
    },
    {
      "entropy": 5.744619417190552,
      "epoch": 4.567858079175042,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030878128709920964,
      "loss": 5.2406,
      "mean_token_accuracy": 0.20732791125774383,
      "num_tokens": 105925881.0,
      "step": 52270
    },
    {
      "entropy": 5.751769208908081,
      "epoch": 4.568295027527746,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030875047889181205,
      "loss": 5.2987,
      "mean_token_accuracy": 0.2053053632378578,
      "num_tokens": 105936924.0,
      "step": 52275
    },
    {
      "entropy": 5.789592599868774,
      "epoch": 4.568731975880451,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003087196700370626,
      "loss": 5.2505,
      "mean_token_accuracy": 0.20831457078456878,
      "num_tokens": 105947507.0,
      "step": 52280
    },
    {
      "entropy": 5.719886207580567,
      "epoch": 4.569168924233155,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003086888605355523,
      "loss": 5.175,
      "mean_token_accuracy": 0.21455345451831817,
      "num_tokens": 105957146.0,
      "step": 52285
    },
    {
      "entropy": 5.681102085113525,
      "epoch": 4.569605872585861,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000308658050387872,
      "loss": 5.1336,
      "mean_token_accuracy": 0.21831993013620377,
      "num_tokens": 105967339.0,
      "step": 52290
    },
    {
      "entropy": 5.732712984085083,
      "epoch": 4.570042820938565,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030862723959461266,
      "loss": 5.1709,
      "mean_token_accuracy": 0.22163686603307725,
      "num_tokens": 105977280.0,
      "step": 52295
    },
    {
      "entropy": 5.731552696228027,
      "epoch": 4.5704797692912695,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003085964281563652,
      "loss": 5.1895,
      "mean_token_accuracy": 0.21554525345563888,
      "num_tokens": 105986691.0,
      "step": 52300
    },
    {
      "entropy": 5.74252724647522,
      "epoch": 4.570916717643975,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030856561607372064,
      "loss": 5.1899,
      "mean_token_accuracy": 0.2073956698179245,
      "num_tokens": 105997589.0,
      "step": 52305
    },
    {
      "entropy": 5.682808971405029,
      "epoch": 4.571353665996679,
      "grad_norm": 1.0,
      "learning_rate": 0.0003085348033472699,
      "loss": 5.1885,
      "mean_token_accuracy": 0.21010738462209702,
      "num_tokens": 106007537.0,
      "step": 52310
    },
    {
      "entropy": 5.723634481430054,
      "epoch": 4.571790614349384,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030850398997760414,
      "loss": 5.1307,
      "mean_token_accuracy": 0.21462692469358444,
      "num_tokens": 106017731.0,
      "step": 52315
    },
    {
      "entropy": 5.761875009536743,
      "epoch": 4.572227562702088,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003084731759653144,
      "loss": 5.3007,
      "mean_token_accuracy": 0.20496968179941177,
      "num_tokens": 106028670.0,
      "step": 52320
    },
    {
      "entropy": 5.768000268936158,
      "epoch": 4.572664511054794,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030844236131099146,
      "loss": 5.1736,
      "mean_token_accuracy": 0.21322368383407592,
      "num_tokens": 106038167.0,
      "step": 52325
    },
    {
      "entropy": 5.765533351898194,
      "epoch": 4.573101459407498,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030841154601522655,
      "loss": 5.253,
      "mean_token_accuracy": 0.2102886289358139,
      "num_tokens": 106048733.0,
      "step": 52330
    },
    {
      "entropy": 5.722169876098633,
      "epoch": 4.573538407760203,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030838073007861066,
      "loss": 5.0994,
      "mean_token_accuracy": 0.2187839761376381,
      "num_tokens": 106058643.0,
      "step": 52335
    },
    {
      "entropy": 5.729326391220093,
      "epoch": 4.573975356112907,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030834991350173484,
      "loss": 5.2716,
      "mean_token_accuracy": 0.20670784562826156,
      "num_tokens": 106068796.0,
      "step": 52340
    },
    {
      "entropy": 5.696891736984253,
      "epoch": 4.574412304465612,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030831909628519026,
      "loss": 5.2235,
      "mean_token_accuracy": 0.21828722953796387,
      "num_tokens": 106078901.0,
      "step": 52345
    },
    {
      "entropy": 5.775319671630859,
      "epoch": 4.574849252818317,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030828827842956796,
      "loss": 5.2477,
      "mean_token_accuracy": 0.20738012194633484,
      "num_tokens": 106088249.0,
      "step": 52350
    },
    {
      "entropy": 5.796012258529663,
      "epoch": 4.575286201171021,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000308257459935459,
      "loss": 5.1976,
      "mean_token_accuracy": 0.20668897926807403,
      "num_tokens": 106098421.0,
      "step": 52355
    },
    {
      "entropy": 5.665899658203125,
      "epoch": 4.575723149523727,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003082266408034546,
      "loss": 5.1319,
      "mean_token_accuracy": 0.21274561882019044,
      "num_tokens": 106108988.0,
      "step": 52360
    },
    {
      "entropy": 5.746353387832642,
      "epoch": 4.576160097876431,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003081958210341458,
      "loss": 5.246,
      "mean_token_accuracy": 0.2050361379981041,
      "num_tokens": 106119925.0,
      "step": 52365
    },
    {
      "entropy": 5.710245847702026,
      "epoch": 4.576597046229136,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030816500062812375,
      "loss": 5.1481,
      "mean_token_accuracy": 0.2104862317442894,
      "num_tokens": 106129575.0,
      "step": 52370
    },
    {
      "entropy": 5.711358642578125,
      "epoch": 4.57703399458184,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030813417958597955,
      "loss": 5.2203,
      "mean_token_accuracy": 0.21137610375881194,
      "num_tokens": 106139338.0,
      "step": 52375
    },
    {
      "entropy": 5.743761014938355,
      "epoch": 4.5774709429345455,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030810335790830445,
      "loss": 5.163,
      "mean_token_accuracy": 0.2109944298863411,
      "num_tokens": 106149558.0,
      "step": 52380
    },
    {
      "entropy": 5.750881052017212,
      "epoch": 4.57790789128725,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003080725355956897,
      "loss": 5.2467,
      "mean_token_accuracy": 0.20791172683238984,
      "num_tokens": 106160018.0,
      "step": 52385
    },
    {
      "entropy": 5.718981409072876,
      "epoch": 4.578344839639954,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003080417126487263,
      "loss": 5.1948,
      "mean_token_accuracy": 0.21608973145484925,
      "num_tokens": 106171027.0,
      "step": 52390
    },
    {
      "entropy": 5.751422595977783,
      "epoch": 4.578781787992659,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003080108890680056,
      "loss": 5.1518,
      "mean_token_accuracy": 0.21494151502847672,
      "num_tokens": 106180625.0,
      "step": 52395
    },
    {
      "entropy": 5.566308259963989,
      "epoch": 4.579218736345364,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003079800648541186,
      "loss": 5.0895,
      "mean_token_accuracy": 0.22569560110569,
      "num_tokens": 106190318.0,
      "step": 52400
    },
    {
      "entropy": 5.710084581375122,
      "epoch": 4.579655684698069,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003079492400076568,
      "loss": 5.1894,
      "mean_token_accuracy": 0.2077864319086075,
      "num_tokens": 106200145.0,
      "step": 52405
    },
    {
      "entropy": 5.765716695785523,
      "epoch": 4.580092633050773,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030791841452921127,
      "loss": 5.1791,
      "mean_token_accuracy": 0.2193007469177246,
      "num_tokens": 106210861.0,
      "step": 52410
    },
    {
      "entropy": 5.785117197036743,
      "epoch": 4.5805295814034785,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030788758841937335,
      "loss": 5.2261,
      "mean_token_accuracy": 0.20834635943174362,
      "num_tokens": 106221810.0,
      "step": 52415
    },
    {
      "entropy": 5.81148681640625,
      "epoch": 4.580966529756183,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003078567616787342,
      "loss": 5.2597,
      "mean_token_accuracy": 0.2050096794962883,
      "num_tokens": 106232332.0,
      "step": 52420
    },
    {
      "entropy": 5.79560546875,
      "epoch": 4.5814034781088875,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003078259343078852,
      "loss": 5.2196,
      "mean_token_accuracy": 0.20755250602960587,
      "num_tokens": 106243243.0,
      "step": 52425
    },
    {
      "entropy": 5.697043180465698,
      "epoch": 4.581840426461592,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003077951063074175,
      "loss": 5.1482,
      "mean_token_accuracy": 0.21546910256147384,
      "num_tokens": 106252346.0,
      "step": 52430
    },
    {
      "entropy": 5.725067281723023,
      "epoch": 4.582277374814297,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003077642776779225,
      "loss": 5.2163,
      "mean_token_accuracy": 0.21373579502105713,
      "num_tokens": 106261849.0,
      "step": 52435
    },
    {
      "entropy": 5.711783170700073,
      "epoch": 4.582714323167002,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030773344841999154,
      "loss": 5.2534,
      "mean_token_accuracy": 0.2110290512442589,
      "num_tokens": 106272846.0,
      "step": 52440
    },
    {
      "entropy": 5.731394577026367,
      "epoch": 4.583151271519706,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030770261853421585,
      "loss": 5.1419,
      "mean_token_accuracy": 0.220708529651165,
      "num_tokens": 106282625.0,
      "step": 52445
    },
    {
      "entropy": 5.681801509857178,
      "epoch": 4.583588219872411,
      "grad_norm": 1.125,
      "learning_rate": 0.0003076717880211868,
      "loss": 5.1523,
      "mean_token_accuracy": 0.22156188786029815,
      "num_tokens": 106292340.0,
      "step": 52450
    },
    {
      "entropy": 5.793785476684571,
      "epoch": 4.584025168225116,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030764095688149567,
      "loss": 5.2599,
      "mean_token_accuracy": 0.20784602016210557,
      "num_tokens": 106301795.0,
      "step": 52455
    },
    {
      "entropy": 5.834345865249634,
      "epoch": 4.5844621165778205,
      "grad_norm": 1.125,
      "learning_rate": 0.00030761012511573406,
      "loss": 5.2516,
      "mean_token_accuracy": 0.21017199158668518,
      "num_tokens": 106313640.0,
      "step": 52460
    },
    {
      "entropy": 5.772209644317627,
      "epoch": 4.584899064930525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030757929272449296,
      "loss": 5.1789,
      "mean_token_accuracy": 0.2076279416680336,
      "num_tokens": 106323627.0,
      "step": 52465
    },
    {
      "entropy": 5.752750205993652,
      "epoch": 4.58533601328323,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003075484597083641,
      "loss": 5.287,
      "mean_token_accuracy": 0.2068563625216484,
      "num_tokens": 106334495.0,
      "step": 52470
    },
    {
      "entropy": 5.7637707710266115,
      "epoch": 4.585772961635935,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030751762606793864,
      "loss": 5.2471,
      "mean_token_accuracy": 0.2103014186024666,
      "num_tokens": 106345209.0,
      "step": 52475
    },
    {
      "entropy": 5.799087858200073,
      "epoch": 4.586209909988639,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003074867918038081,
      "loss": 5.2401,
      "mean_token_accuracy": 0.2034217670559883,
      "num_tokens": 106355294.0,
      "step": 52480
    },
    {
      "entropy": 5.783552789688111,
      "epoch": 4.586646858341344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003074559569165639,
      "loss": 5.2569,
      "mean_token_accuracy": 0.20661344677209853,
      "num_tokens": 106366340.0,
      "step": 52485
    },
    {
      "entropy": 5.7572181224823,
      "epoch": 4.587083806694049,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003074251214067974,
      "loss": 5.1873,
      "mean_token_accuracy": 0.21199725270271302,
      "num_tokens": 106376270.0,
      "step": 52490
    },
    {
      "entropy": 5.741531229019165,
      "epoch": 4.587520755046754,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003073942852751001,
      "loss": 5.2182,
      "mean_token_accuracy": 0.21319472938776016,
      "num_tokens": 106386231.0,
      "step": 52495
    },
    {
      "entropy": 5.736451244354248,
      "epoch": 4.587957703399458,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030736344852206346,
      "loss": 5.1353,
      "mean_token_accuracy": 0.21465612351894378,
      "num_tokens": 106396109.0,
      "step": 52500
    },
    {
      "entropy": 5.702086925506592,
      "epoch": 4.5883946517521625,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030733261114827885,
      "loss": 5.1791,
      "mean_token_accuracy": 0.21245138645172118,
      "num_tokens": 106406404.0,
      "step": 52505
    },
    {
      "entropy": 5.728720569610596,
      "epoch": 4.588831600104868,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030730177315433787,
      "loss": 5.17,
      "mean_token_accuracy": 0.22089006900787353,
      "num_tokens": 106416762.0,
      "step": 52510
    },
    {
      "entropy": 5.748462533950805,
      "epoch": 4.589268548457572,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030727093454083193,
      "loss": 5.2401,
      "mean_token_accuracy": 0.20743171870708466,
      "num_tokens": 106427095.0,
      "step": 52515
    },
    {
      "entropy": 5.65178337097168,
      "epoch": 4.589705496810277,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030724009530835254,
      "loss": 5.1996,
      "mean_token_accuracy": 0.2259836345911026,
      "num_tokens": 106437576.0,
      "step": 52520
    },
    {
      "entropy": 5.622574090957642,
      "epoch": 4.590142445162982,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003072092554574912,
      "loss": 5.0907,
      "mean_token_accuracy": 0.2214026615023613,
      "num_tokens": 106448735.0,
      "step": 52525
    },
    {
      "entropy": 5.726290082931518,
      "epoch": 4.590579393515687,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030717841498883964,
      "loss": 5.2241,
      "mean_token_accuracy": 0.21078450381755828,
      "num_tokens": 106459039.0,
      "step": 52530
    },
    {
      "entropy": 5.701711893081665,
      "epoch": 4.591016341868391,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000307147573902989,
      "loss": 5.1639,
      "mean_token_accuracy": 0.2140205442905426,
      "num_tokens": 106469257.0,
      "step": 52535
    },
    {
      "entropy": 5.72127251625061,
      "epoch": 4.591453290221096,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003071167322005312,
      "loss": 5.1662,
      "mean_token_accuracy": 0.21569748371839523,
      "num_tokens": 106479564.0,
      "step": 52540
    },
    {
      "entropy": 5.790946006774902,
      "epoch": 4.591890238573801,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030708588988205755,
      "loss": 5.1988,
      "mean_token_accuracy": 0.2115831196308136,
      "num_tokens": 106489606.0,
      "step": 52545
    },
    {
      "entropy": 5.765503311157227,
      "epoch": 4.592327186926505,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030705504694815986,
      "loss": 5.1426,
      "mean_token_accuracy": 0.2188754692673683,
      "num_tokens": 106500850.0,
      "step": 52550
    },
    {
      "entropy": 5.744914102554321,
      "epoch": 4.59276413527921,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030702420339942944,
      "loss": 5.2488,
      "mean_token_accuracy": 0.2065680742263794,
      "num_tokens": 106510563.0,
      "step": 52555
    },
    {
      "entropy": 5.718670892715454,
      "epoch": 4.593201083631914,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030699335923645805,
      "loss": 5.2399,
      "mean_token_accuracy": 0.20493504405021667,
      "num_tokens": 106520709.0,
      "step": 52560
    },
    {
      "entropy": 5.7718456268310545,
      "epoch": 4.59363803198462,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030696251445983723,
      "loss": 5.2382,
      "mean_token_accuracy": 0.20641193091869353,
      "num_tokens": 106530946.0,
      "step": 52565
    },
    {
      "entropy": 5.768351125717163,
      "epoch": 4.594074980337324,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003069316690701587,
      "loss": 5.1815,
      "mean_token_accuracy": 0.2116088792681694,
      "num_tokens": 106540735.0,
      "step": 52570
    },
    {
      "entropy": 5.7333043098449705,
      "epoch": 4.594511928690029,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000306900823068014,
      "loss": 5.2077,
      "mean_token_accuracy": 0.21675915271043777,
      "num_tokens": 106551128.0,
      "step": 52575
    },
    {
      "entropy": 5.730377626419068,
      "epoch": 4.594948877042734,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030686997645399484,
      "loss": 5.2614,
      "mean_token_accuracy": 0.20755353718996047,
      "num_tokens": 106562602.0,
      "step": 52580
    },
    {
      "entropy": 5.74985671043396,
      "epoch": 4.5953858253954385,
      "grad_norm": 1.0,
      "learning_rate": 0.00030683912922869274,
      "loss": 5.1775,
      "mean_token_accuracy": 0.21642849147319793,
      "num_tokens": 106573409.0,
      "step": 52585
    },
    {
      "entropy": 5.838548803329468,
      "epoch": 4.595822773748143,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003068082813926995,
      "loss": 5.2758,
      "mean_token_accuracy": 0.20156539380550384,
      "num_tokens": 106584342.0,
      "step": 52590
    },
    {
      "entropy": 5.822613143920899,
      "epoch": 4.596259722100847,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003067774329466068,
      "loss": 5.3059,
      "mean_token_accuracy": 0.1995843917131424,
      "num_tokens": 106596154.0,
      "step": 52595
    },
    {
      "entropy": 5.673223161697388,
      "epoch": 4.596696670453552,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030674658389100623,
      "loss": 5.0926,
      "mean_token_accuracy": 0.22482538521289824,
      "num_tokens": 106605842.0,
      "step": 52600
    },
    {
      "entropy": 5.757203149795532,
      "epoch": 4.597133618806257,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030671573422648955,
      "loss": 5.2388,
      "mean_token_accuracy": 0.21332545578479767,
      "num_tokens": 106616319.0,
      "step": 52605
    },
    {
      "entropy": 5.791321229934693,
      "epoch": 4.597570567158962,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003066848839536485,
      "loss": 5.2512,
      "mean_token_accuracy": 0.2107214957475662,
      "num_tokens": 106627259.0,
      "step": 52610
    },
    {
      "entropy": 5.781975841522216,
      "epoch": 4.598007515511666,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003066540330730747,
      "loss": 5.2287,
      "mean_token_accuracy": 0.21189174503087999,
      "num_tokens": 106637712.0,
      "step": 52615
    },
    {
      "entropy": 5.783879804611206,
      "epoch": 4.598444463864372,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030662318158536004,
      "loss": 5.1874,
      "mean_token_accuracy": 0.22261080890893936,
      "num_tokens": 106648896.0,
      "step": 52620
    },
    {
      "entropy": 5.730694913864136,
      "epoch": 4.598881412217076,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030659232949109614,
      "loss": 5.1991,
      "mean_token_accuracy": 0.2185831516981125,
      "num_tokens": 106660016.0,
      "step": 52625
    },
    {
      "entropy": 5.687144088745117,
      "epoch": 4.5993183605697805,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003065614767908749,
      "loss": 5.1951,
      "mean_token_accuracy": 0.21384121775627135,
      "num_tokens": 106669631.0,
      "step": 52630
    },
    {
      "entropy": 5.729037141799926,
      "epoch": 4.599755308922486,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030653062348528793,
      "loss": 5.2243,
      "mean_token_accuracy": 0.21106260716915132,
      "num_tokens": 106679218.0,
      "step": 52635
    },
    {
      "entropy": 5.795842933654785,
      "epoch": 4.60019225727519,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000306499769574927,
      "loss": 5.1384,
      "mean_token_accuracy": 0.21642330586910247,
      "num_tokens": 106688694.0,
      "step": 52640
    },
    {
      "entropy": 5.738474798202515,
      "epoch": 4.600629205627895,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003064689150603841,
      "loss": 5.1787,
      "mean_token_accuracy": 0.2160705015063286,
      "num_tokens": 106698044.0,
      "step": 52645
    },
    {
      "entropy": 5.696801662445068,
      "epoch": 4.601066153980599,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003064380599422509,
      "loss": 5.1746,
      "mean_token_accuracy": 0.21119610518217086,
      "num_tokens": 106707973.0,
      "step": 52650
    },
    {
      "entropy": 5.681399202346801,
      "epoch": 4.601503102333304,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003064072042211192,
      "loss": 5.0753,
      "mean_token_accuracy": 0.22188300639390945,
      "num_tokens": 106717827.0,
      "step": 52655
    },
    {
      "entropy": 5.684625768661499,
      "epoch": 4.601940050686009,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003063763478975809,
      "loss": 5.1539,
      "mean_token_accuracy": 0.22639009356498718,
      "num_tokens": 106727168.0,
      "step": 52660
    },
    {
      "entropy": 5.789338874816894,
      "epoch": 4.602376999038714,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003063454909722278,
      "loss": 5.2467,
      "mean_token_accuracy": 0.21070023775100707,
      "num_tokens": 106738686.0,
      "step": 52665
    },
    {
      "entropy": 5.777047777175904,
      "epoch": 4.602813947391418,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003063146334456517,
      "loss": 5.2652,
      "mean_token_accuracy": 0.2066127523779869,
      "num_tokens": 106748326.0,
      "step": 52670
    },
    {
      "entropy": 5.729000425338745,
      "epoch": 4.603250895744123,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003062837753184446,
      "loss": 5.1516,
      "mean_token_accuracy": 0.2202035218477249,
      "num_tokens": 106758255.0,
      "step": 52675
    },
    {
      "entropy": 5.704656744003296,
      "epoch": 4.603687844096828,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030625291659119825,
      "loss": 5.2127,
      "mean_token_accuracy": 0.2152406543493271,
      "num_tokens": 106767671.0,
      "step": 52680
    },
    {
      "entropy": 5.686500310897827,
      "epoch": 4.604124792449532,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003062220572645046,
      "loss": 5.1403,
      "mean_token_accuracy": 0.2178018108010292,
      "num_tokens": 106778217.0,
      "step": 52685
    },
    {
      "entropy": 5.766377735137939,
      "epoch": 4.604561740802237,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003061911973389555,
      "loss": 5.1423,
      "mean_token_accuracy": 0.21748314946889877,
      "num_tokens": 106788134.0,
      "step": 52690
    },
    {
      "entropy": 5.692324924468994,
      "epoch": 4.604998689154942,
      "grad_norm": 1.015625,
      "learning_rate": 0.000306160336815143,
      "loss": 5.1208,
      "mean_token_accuracy": 0.21228015869855882,
      "num_tokens": 106797933.0,
      "step": 52695
    },
    {
      "entropy": 5.6882579803466795,
      "epoch": 4.605435637507647,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030612947569365876,
      "loss": 5.1861,
      "mean_token_accuracy": 0.214193195104599,
      "num_tokens": 106808175.0,
      "step": 52700
    },
    {
      "entropy": 5.746799659729004,
      "epoch": 4.605872585860351,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000306098613975095,
      "loss": 5.2425,
      "mean_token_accuracy": 0.2072489634156227,
      "num_tokens": 106819197.0,
      "step": 52705
    },
    {
      "entropy": 5.7102550029754635,
      "epoch": 4.606309534213056,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030606775166004343,
      "loss": 5.1482,
      "mean_token_accuracy": 0.21851257234811783,
      "num_tokens": 106830527.0,
      "step": 52710
    },
    {
      "entropy": 5.749148797988892,
      "epoch": 4.606746482565761,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003060368887490961,
      "loss": 5.2324,
      "mean_token_accuracy": 0.20171142369508743,
      "num_tokens": 106840208.0,
      "step": 52715
    },
    {
      "entropy": 5.71410231590271,
      "epoch": 4.607183430918465,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030600602524284503,
      "loss": 5.1895,
      "mean_token_accuracy": 0.21367186307907104,
      "num_tokens": 106850144.0,
      "step": 52720
    },
    {
      "entropy": 5.749515676498413,
      "epoch": 4.60762037927117,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030597516114188217,
      "loss": 5.1912,
      "mean_token_accuracy": 0.21569785177707673,
      "num_tokens": 106860585.0,
      "step": 52725
    },
    {
      "entropy": 5.735503911972046,
      "epoch": 4.608057327623875,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030594429644679945,
      "loss": 5.1763,
      "mean_token_accuracy": 0.21011535674333573,
      "num_tokens": 106871098.0,
      "step": 52730
    },
    {
      "entropy": 5.776554822921753,
      "epoch": 4.60849427597658,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030591343115818885,
      "loss": 5.282,
      "mean_token_accuracy": 0.20650826245546341,
      "num_tokens": 106880921.0,
      "step": 52735
    },
    {
      "entropy": 5.748296642303467,
      "epoch": 4.608931224329284,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003058825652766426,
      "loss": 5.1914,
      "mean_token_accuracy": 0.20994327515363692,
      "num_tokens": 106890490.0,
      "step": 52740
    },
    {
      "entropy": 5.731605434417725,
      "epoch": 4.609368172681989,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003058516988027524,
      "loss": 5.2336,
      "mean_token_accuracy": 0.21235451996326446,
      "num_tokens": 106900200.0,
      "step": 52745
    },
    {
      "entropy": 5.732453441619873,
      "epoch": 4.609805121034694,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030582083173711055,
      "loss": 5.2683,
      "mean_token_accuracy": 0.20443932265043258,
      "num_tokens": 106910354.0,
      "step": 52750
    },
    {
      "entropy": 5.749299669265747,
      "epoch": 4.6102420693873984,
      "grad_norm": 1.078125,
      "learning_rate": 0.000305789964080309,
      "loss": 5.1061,
      "mean_token_accuracy": 0.21728386133909225,
      "num_tokens": 106920669.0,
      "step": 52755
    },
    {
      "entropy": 5.749757623672485,
      "epoch": 4.610679017740103,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030575909583293974,
      "loss": 5.21,
      "mean_token_accuracy": 0.21070388555526734,
      "num_tokens": 106930163.0,
      "step": 52760
    },
    {
      "entropy": 5.751344347000122,
      "epoch": 4.611115966092807,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000305728226995595,
      "loss": 5.2195,
      "mean_token_accuracy": 0.2040123611688614,
      "num_tokens": 106939588.0,
      "step": 52765
    },
    {
      "entropy": 5.726315975189209,
      "epoch": 4.611552914445513,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003056973575688667,
      "loss": 5.2296,
      "mean_token_accuracy": 0.20611241161823274,
      "num_tokens": 106950044.0,
      "step": 52770
    },
    {
      "entropy": 5.702286243438721,
      "epoch": 4.611989862798217,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030566648755334704,
      "loss": 5.1318,
      "mean_token_accuracy": 0.21845499575138091,
      "num_tokens": 106959970.0,
      "step": 52775
    },
    {
      "entropy": 5.7033661842346195,
      "epoch": 4.612426811150922,
      "grad_norm": 1.0625,
      "learning_rate": 0.000305635616949628,
      "loss": 5.1156,
      "mean_token_accuracy": 0.21997424513101577,
      "num_tokens": 106970055.0,
      "step": 52780
    },
    {
      "entropy": 5.687873411178589,
      "epoch": 4.612863759503627,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003056047457583018,
      "loss": 5.1612,
      "mean_token_accuracy": 0.22132312059402465,
      "num_tokens": 106979598.0,
      "step": 52785
    },
    {
      "entropy": 5.696700239181519,
      "epoch": 4.6133007078563315,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003055738739799606,
      "loss": 5.1865,
      "mean_token_accuracy": 0.21728538125753402,
      "num_tokens": 106990662.0,
      "step": 52790
    },
    {
      "entropy": 5.686467170715332,
      "epoch": 4.613737656209036,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030554300161519645,
      "loss": 5.2031,
      "mean_token_accuracy": 0.21742510050535202,
      "num_tokens": 107000942.0,
      "step": 52795
    },
    {
      "entropy": 5.795657682418823,
      "epoch": 4.6141746045617404,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030551212866460146,
      "loss": 5.2839,
      "mean_token_accuracy": 0.2046491250395775,
      "num_tokens": 107010928.0,
      "step": 52800
    },
    {
      "entropy": 5.682701683044433,
      "epoch": 4.614611552914446,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003054812551287679,
      "loss": 5.1499,
      "mean_token_accuracy": 0.21599683165550232,
      "num_tokens": 107020563.0,
      "step": 52805
    },
    {
      "entropy": 5.724782943725586,
      "epoch": 4.61504850126715,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030545038100828793,
      "loss": 5.2179,
      "mean_token_accuracy": 0.21690996140241622,
      "num_tokens": 107032260.0,
      "step": 52810
    },
    {
      "entropy": 5.748409080505371,
      "epoch": 4.615485449619855,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030541950630375367,
      "loss": 5.245,
      "mean_token_accuracy": 0.20057404041290283,
      "num_tokens": 107041866.0,
      "step": 52815
    },
    {
      "entropy": 5.763511323928833,
      "epoch": 4.615922397972559,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003053886310157573,
      "loss": 5.2759,
      "mean_token_accuracy": 0.20579042583703994,
      "num_tokens": 107052621.0,
      "step": 52820
    },
    {
      "entropy": 5.771842336654663,
      "epoch": 4.616359346325265,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003053577551448912,
      "loss": 5.2084,
      "mean_token_accuracy": 0.20828789472579956,
      "num_tokens": 107061634.0,
      "step": 52825
    },
    {
      "entropy": 5.723611497879029,
      "epoch": 4.616796294677969,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003053268786917473,
      "loss": 5.2273,
      "mean_token_accuracy": 0.20962940752506257,
      "num_tokens": 107071683.0,
      "step": 52830
    },
    {
      "entropy": 5.669399261474609,
      "epoch": 4.6172332430306735,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000305296001656918,
      "loss": 5.1385,
      "mean_token_accuracy": 0.22132404297590255,
      "num_tokens": 107081654.0,
      "step": 52835
    },
    {
      "entropy": 5.654354524612427,
      "epoch": 4.617670191383379,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030526512404099556,
      "loss": 5.1221,
      "mean_token_accuracy": 0.21959281116724014,
      "num_tokens": 107091853.0,
      "step": 52840
    },
    {
      "entropy": 5.716493940353393,
      "epoch": 4.618107139736083,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030523424584457217,
      "loss": 5.1403,
      "mean_token_accuracy": 0.21683947294950484,
      "num_tokens": 107101485.0,
      "step": 52845
    },
    {
      "entropy": 5.735020160675049,
      "epoch": 4.618544088088788,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003052033670682401,
      "loss": 5.2914,
      "mean_token_accuracy": 0.20684437751770018,
      "num_tokens": 107112051.0,
      "step": 52850
    },
    {
      "entropy": 5.81099681854248,
      "epoch": 4.618981036441492,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003051724877125916,
      "loss": 5.2527,
      "mean_token_accuracy": 0.20540691316127777,
      "num_tokens": 107121954.0,
      "step": 52855
    },
    {
      "entropy": 5.715100431442261,
      "epoch": 4.619417984794198,
      "grad_norm": 1.078125,
      "learning_rate": 0.000305141607778219,
      "loss": 5.1176,
      "mean_token_accuracy": 0.22624235451221467,
      "num_tokens": 107132521.0,
      "step": 52860
    },
    {
      "entropy": 5.729641056060791,
      "epoch": 4.619854933146902,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003051107272657146,
      "loss": 5.1432,
      "mean_token_accuracy": 0.21381037086248397,
      "num_tokens": 107142795.0,
      "step": 52865
    },
    {
      "entropy": 5.709730911254883,
      "epoch": 4.620291881499607,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003050798461756706,
      "loss": 5.2413,
      "mean_token_accuracy": 0.20769782662391661,
      "num_tokens": 107153660.0,
      "step": 52870
    },
    {
      "entropy": 5.759910726547242,
      "epoch": 4.620728829852311,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030504896450867934,
      "loss": 5.2356,
      "mean_token_accuracy": 0.21086053550243378,
      "num_tokens": 107163840.0,
      "step": 52875
    },
    {
      "entropy": 5.762806749343872,
      "epoch": 4.621165778205016,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030501808226533327,
      "loss": 5.2163,
      "mean_token_accuracy": 0.21401909738779068,
      "num_tokens": 107173471.0,
      "step": 52880
    },
    {
      "entropy": 5.759777641296386,
      "epoch": 4.621602726557721,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003049871994462247,
      "loss": 5.1366,
      "mean_token_accuracy": 0.21026961654424667,
      "num_tokens": 107183077.0,
      "step": 52885
    },
    {
      "entropy": 5.703420114517212,
      "epoch": 4.622039674910425,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030495631605194584,
      "loss": 5.1239,
      "mean_token_accuracy": 0.2240644723176956,
      "num_tokens": 107193330.0,
      "step": 52890
    },
    {
      "entropy": 5.662634229660034,
      "epoch": 4.622476623263131,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003049254320830892,
      "loss": 5.1807,
      "mean_token_accuracy": 0.2164678767323494,
      "num_tokens": 107203528.0,
      "step": 52895
    },
    {
      "entropy": 5.711431217193604,
      "epoch": 4.622913571615835,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003048945475402471,
      "loss": 5.2279,
      "mean_token_accuracy": 0.20092480331659318,
      "num_tokens": 107213195.0,
      "step": 52900
    },
    {
      "entropy": 5.71175799369812,
      "epoch": 4.62335051996854,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003048636624240119,
      "loss": 5.1174,
      "mean_token_accuracy": 0.22319078892469407,
      "num_tokens": 107222178.0,
      "step": 52905
    },
    {
      "entropy": 5.732737445831299,
      "epoch": 4.623787468321244,
      "grad_norm": 1.125,
      "learning_rate": 0.00030483277673497603,
      "loss": 5.2172,
      "mean_token_accuracy": 0.21622426807880402,
      "num_tokens": 107232982.0,
      "step": 52910
    },
    {
      "entropy": 5.678849506378174,
      "epoch": 4.6242244166739495,
      "grad_norm": 1.125,
      "learning_rate": 0.00030480189047373186,
      "loss": 5.1037,
      "mean_token_accuracy": 0.21687746196985244,
      "num_tokens": 107241851.0,
      "step": 52915
    },
    {
      "entropy": 5.70654182434082,
      "epoch": 4.624661365026654,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003047710036408718,
      "loss": 5.1323,
      "mean_token_accuracy": 0.20782055109739303,
      "num_tokens": 107251170.0,
      "step": 52920
    },
    {
      "entropy": 5.810210752487182,
      "epoch": 4.625098313379358,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030474011623698826,
      "loss": 5.2145,
      "mean_token_accuracy": 0.20941513776779175,
      "num_tokens": 107260944.0,
      "step": 52925
    },
    {
      "entropy": 5.780109405517578,
      "epoch": 4.625535261732063,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003047092282626738,
      "loss": 5.2428,
      "mean_token_accuracy": 0.20969585776329042,
      "num_tokens": 107271181.0,
      "step": 52930
    },
    {
      "entropy": 5.6893235206604,
      "epoch": 4.625972210084768,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003046783397185207,
      "loss": 5.2136,
      "mean_token_accuracy": 0.21359445601701738,
      "num_tokens": 107280809.0,
      "step": 52935
    },
    {
      "entropy": 5.666692924499512,
      "epoch": 4.626409158437473,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003046474506051214,
      "loss": 5.1043,
      "mean_token_accuracy": 0.22076445817947388,
      "num_tokens": 107290653.0,
      "step": 52940
    },
    {
      "entropy": 5.741936922073364,
      "epoch": 4.626846106790177,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003046165609230686,
      "loss": 5.2251,
      "mean_token_accuracy": 0.21095598638057708,
      "num_tokens": 107300518.0,
      "step": 52945
    },
    {
      "entropy": 5.699287605285645,
      "epoch": 4.6272830551428825,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003045856706729547,
      "loss": 5.1092,
      "mean_token_accuracy": 0.21700287163257598,
      "num_tokens": 107310514.0,
      "step": 52950
    },
    {
      "entropy": 5.722618198394775,
      "epoch": 4.627720003495587,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000304554779855372,
      "loss": 5.2133,
      "mean_token_accuracy": 0.21906280070543288,
      "num_tokens": 107320383.0,
      "step": 52955
    },
    {
      "entropy": 5.712089538574219,
      "epoch": 4.6281569518482915,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030452388847091317,
      "loss": 5.2146,
      "mean_token_accuracy": 0.21854099482297898,
      "num_tokens": 107331228.0,
      "step": 52960
    },
    {
      "entropy": 5.6876836776733395,
      "epoch": 4.628593900200996,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003044929965201707,
      "loss": 5.1792,
      "mean_token_accuracy": 0.2157617285847664,
      "num_tokens": 107340543.0,
      "step": 52965
    },
    {
      "entropy": 5.661074256896972,
      "epoch": 4.629030848553701,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003044621040037371,
      "loss": 5.2093,
      "mean_token_accuracy": 0.21501528173685075,
      "num_tokens": 107351657.0,
      "step": 52970
    },
    {
      "entropy": 5.818478393554687,
      "epoch": 4.629467796906406,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030443121092220486,
      "loss": 5.3143,
      "mean_token_accuracy": 0.200690396130085,
      "num_tokens": 107361630.0,
      "step": 52975
    },
    {
      "entropy": 5.794460821151733,
      "epoch": 4.62990474525911,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003044003172761667,
      "loss": 5.1953,
      "mean_token_accuracy": 0.21408091783523558,
      "num_tokens": 107371488.0,
      "step": 52980
    },
    {
      "entropy": 5.710949468612671,
      "epoch": 4.630341693611815,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030436942306621495,
      "loss": 5.1798,
      "mean_token_accuracy": 0.21209385246038437,
      "num_tokens": 107381906.0,
      "step": 52985
    },
    {
      "entropy": 5.7378880977630615,
      "epoch": 4.63077864196452,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003043385282929423,
      "loss": 5.2369,
      "mean_token_accuracy": 0.20893661677837372,
      "num_tokens": 107392488.0,
      "step": 52990
    },
    {
      "entropy": 5.731385612487793,
      "epoch": 4.6312155903172245,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003043076329569412,
      "loss": 5.1273,
      "mean_token_accuracy": 0.21953654587268828,
      "num_tokens": 107402479.0,
      "step": 52995
    },
    {
      "entropy": 5.7605994701385494,
      "epoch": 4.631652538669929,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030427673705880453,
      "loss": 5.3276,
      "mean_token_accuracy": 0.21044564247131348,
      "num_tokens": 107412870.0,
      "step": 53000
    },
    {
      "entropy": 5.766379356384277,
      "epoch": 4.632089487022634,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003042458405991246,
      "loss": 5.1754,
      "mean_token_accuracy": 0.21481914073228836,
      "num_tokens": 107422923.0,
      "step": 53005
    },
    {
      "entropy": 5.686727571487427,
      "epoch": 4.632526435375339,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003042149435784942,
      "loss": 5.1647,
      "mean_token_accuracy": 0.21064198166131973,
      "num_tokens": 107432233.0,
      "step": 53010
    },
    {
      "entropy": 5.693139553070068,
      "epoch": 4.632963383728043,
      "grad_norm": 1.0,
      "learning_rate": 0.00030418404599750576,
      "loss": 5.1927,
      "mean_token_accuracy": 0.21521875262260437,
      "num_tokens": 107442838.0,
      "step": 53015
    },
    {
      "entropy": 5.7620445728302006,
      "epoch": 4.633400332080748,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030415314785675206,
      "loss": 5.1955,
      "mean_token_accuracy": 0.2148850753903389,
      "num_tokens": 107453542.0,
      "step": 53020
    },
    {
      "entropy": 5.751940822601318,
      "epoch": 4.633837280433453,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003041222491568258,
      "loss": 5.1866,
      "mean_token_accuracy": 0.21874081939458848,
      "num_tokens": 107463139.0,
      "step": 53025
    },
    {
      "entropy": 5.640073204040528,
      "epoch": 4.634274228786158,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030409134989831945,
      "loss": 5.1765,
      "mean_token_accuracy": 0.21117463558912278,
      "num_tokens": 107474213.0,
      "step": 53030
    },
    {
      "entropy": 5.768244028091431,
      "epoch": 4.634711177138862,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030406045008182574,
      "loss": 5.2485,
      "mean_token_accuracy": 0.2106889307498932,
      "num_tokens": 107485317.0,
      "step": 53035
    },
    {
      "entropy": 5.832401466369629,
      "epoch": 4.6351481254915665,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030402954970793747,
      "loss": 5.2225,
      "mean_token_accuracy": 0.209103000164032,
      "num_tokens": 107494460.0,
      "step": 53040
    },
    {
      "entropy": 5.6930357933044435,
      "epoch": 4.635585073844272,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003039986487772471,
      "loss": 5.1767,
      "mean_token_accuracy": 0.21412280797958375,
      "num_tokens": 107504042.0,
      "step": 53045
    },
    {
      "entropy": 5.708801698684693,
      "epoch": 4.636022022196976,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003039677472903476,
      "loss": 5.1915,
      "mean_token_accuracy": 0.2150069758296013,
      "num_tokens": 107513642.0,
      "step": 53050
    },
    {
      "entropy": 5.7648704051971436,
      "epoch": 4.636458970549681,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030393684524783135,
      "loss": 5.2163,
      "mean_token_accuracy": 0.20368503481149675,
      "num_tokens": 107523060.0,
      "step": 53055
    },
    {
      "entropy": 5.69281530380249,
      "epoch": 4.636895918902386,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030390594265029136,
      "loss": 5.1712,
      "mean_token_accuracy": 0.22253578752279282,
      "num_tokens": 107533238.0,
      "step": 53060
    },
    {
      "entropy": 5.55787353515625,
      "epoch": 4.637332867255091,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00030387503949832025,
      "loss": 5.0625,
      "mean_token_accuracy": 0.23121343851089476,
      "num_tokens": 107543076.0,
      "step": 53065
    },
    {
      "entropy": 5.701804780960083,
      "epoch": 4.637769815607795,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030384413579251075,
      "loss": 5.2225,
      "mean_token_accuracy": 0.2141531378030777,
      "num_tokens": 107554288.0,
      "step": 53070
    },
    {
      "entropy": 5.719631862640381,
      "epoch": 4.6382067639605,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003038132315334556,
      "loss": 5.1244,
      "mean_token_accuracy": 0.2125426098704338,
      "num_tokens": 107564346.0,
      "step": 53075
    },
    {
      "entropy": 5.700748062133789,
      "epoch": 4.638643712313205,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003037823267217475,
      "loss": 5.111,
      "mean_token_accuracy": 0.21416828483343125,
      "num_tokens": 107573941.0,
      "step": 53080
    },
    {
      "entropy": 5.66527099609375,
      "epoch": 4.639080660665909,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003037514213579793,
      "loss": 5.1999,
      "mean_token_accuracy": 0.2077791467308998,
      "num_tokens": 107583812.0,
      "step": 53085
    },
    {
      "entropy": 5.716320991516113,
      "epoch": 4.639517609018614,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003037205154427439,
      "loss": 5.1679,
      "mean_token_accuracy": 0.21456888914108277,
      "num_tokens": 107593377.0,
      "step": 53090
    },
    {
      "entropy": 5.665588903427124,
      "epoch": 4.639954557371318,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030368960897663393,
      "loss": 5.218,
      "mean_token_accuracy": 0.2257832556962967,
      "num_tokens": 107604765.0,
      "step": 53095
    },
    {
      "entropy": 5.763782644271851,
      "epoch": 4.640391505724024,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030365870196024216,
      "loss": 5.2455,
      "mean_token_accuracy": 0.20536858439445496,
      "num_tokens": 107615976.0,
      "step": 53100
    },
    {
      "entropy": 5.742709493637085,
      "epoch": 4.640828454076728,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003036277943941615,
      "loss": 5.14,
      "mean_token_accuracy": 0.2158385306596756,
      "num_tokens": 107625098.0,
      "step": 53105
    },
    {
      "entropy": 5.672360801696778,
      "epoch": 4.641265402429433,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030359688627898466,
      "loss": 5.1728,
      "mean_token_accuracy": 0.21663118302822112,
      "num_tokens": 107634488.0,
      "step": 53110
    },
    {
      "entropy": 5.720689010620117,
      "epoch": 4.641702350782138,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003035659776153047,
      "loss": 5.1907,
      "mean_token_accuracy": 0.21449453979730607,
      "num_tokens": 107644030.0,
      "step": 53115
    },
    {
      "entropy": 5.735622930526733,
      "epoch": 4.6421392991348425,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003035350684037143,
      "loss": 5.1316,
      "mean_token_accuracy": 0.21952962577342988,
      "num_tokens": 107653790.0,
      "step": 53120
    },
    {
      "entropy": 5.6977219581604,
      "epoch": 4.642576247487547,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003035041586448063,
      "loss": 5.1071,
      "mean_token_accuracy": 0.21206732988357543,
      "num_tokens": 107664133.0,
      "step": 53125
    },
    {
      "entropy": 5.7059461116790775,
      "epoch": 4.643013195840251,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030347324833917357,
      "loss": 5.2039,
      "mean_token_accuracy": 0.2140579417347908,
      "num_tokens": 107673259.0,
      "step": 53130
    },
    {
      "entropy": 5.759308767318726,
      "epoch": 4.643450144192957,
      "grad_norm": 1.125,
      "learning_rate": 0.000303442337487409,
      "loss": 5.2289,
      "mean_token_accuracy": 0.2080240935087204,
      "num_tokens": 107683250.0,
      "step": 53135
    },
    {
      "entropy": 5.65732855796814,
      "epoch": 4.643887092545661,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003034114260901056,
      "loss": 5.1085,
      "mean_token_accuracy": 0.22180522680282594,
      "num_tokens": 107692166.0,
      "step": 53140
    },
    {
      "entropy": 5.688459014892578,
      "epoch": 4.644324040898366,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003033805141478561,
      "loss": 5.1843,
      "mean_token_accuracy": 0.21701654940843582,
      "num_tokens": 107701906.0,
      "step": 53145
    },
    {
      "entropy": 5.733008670806885,
      "epoch": 4.64476098925107,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030334960166125344,
      "loss": 5.2191,
      "mean_token_accuracy": 0.2110680714249611,
      "num_tokens": 107712139.0,
      "step": 53150
    },
    {
      "entropy": 5.7813914775848385,
      "epoch": 4.6451979376037755,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030331868863089055,
      "loss": 5.2092,
      "mean_token_accuracy": 0.20836686193943024,
      "num_tokens": 107721836.0,
      "step": 53155
    },
    {
      "entropy": 5.73735818862915,
      "epoch": 4.64563488595648,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030328777505736037,
      "loss": 5.2129,
      "mean_token_accuracy": 0.2140532612800598,
      "num_tokens": 107731682.0,
      "step": 53160
    },
    {
      "entropy": 5.688370656967163,
      "epoch": 4.6460718343091845,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003032568609412559,
      "loss": 5.1531,
      "mean_token_accuracy": 0.22001426219940184,
      "num_tokens": 107741870.0,
      "step": 53165
    },
    {
      "entropy": 5.689209222793579,
      "epoch": 4.64650878266189,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030322594628317005,
      "loss": 5.2158,
      "mean_token_accuracy": 0.21168898344039916,
      "num_tokens": 107752507.0,
      "step": 53170
    },
    {
      "entropy": 5.735220098495484,
      "epoch": 4.646945731014594,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003031950310836957,
      "loss": 5.1764,
      "mean_token_accuracy": 0.2137208327651024,
      "num_tokens": 107762842.0,
      "step": 53175
    },
    {
      "entropy": 5.774090766906738,
      "epoch": 4.647382679367299,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003031641153434259,
      "loss": 5.1695,
      "mean_token_accuracy": 0.20858625173568726,
      "num_tokens": 107771796.0,
      "step": 53180
    },
    {
      "entropy": 5.722891759872437,
      "epoch": 4.647819627720003,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003031331990629536,
      "loss": 5.234,
      "mean_token_accuracy": 0.21636386662721635,
      "num_tokens": 107781264.0,
      "step": 53185
    },
    {
      "entropy": 5.7025820255279545,
      "epoch": 4.648256576072709,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003031022822428719,
      "loss": 5.1776,
      "mean_token_accuracy": 0.2155936285853386,
      "num_tokens": 107791405.0,
      "step": 53190
    },
    {
      "entropy": 5.799831771850586,
      "epoch": 4.648693524425413,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030307136488377356,
      "loss": 5.2731,
      "mean_token_accuracy": 0.2035892829298973,
      "num_tokens": 107802168.0,
      "step": 53195
    },
    {
      "entropy": 5.782447624206543,
      "epoch": 4.6491304727781175,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003030404469862518,
      "loss": 5.2468,
      "mean_token_accuracy": 0.20546018630266188,
      "num_tokens": 107811900.0,
      "step": 53200
    },
    {
      "entropy": 5.724066686630249,
      "epoch": 4.649567421130822,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003030095285508995,
      "loss": 5.2205,
      "mean_token_accuracy": 0.20669845789670943,
      "num_tokens": 107821906.0,
      "step": 53205
    },
    {
      "entropy": 5.737465476989746,
      "epoch": 4.650004369483527,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030297860957830984,
      "loss": 5.2112,
      "mean_token_accuracy": 0.20865466743707656,
      "num_tokens": 107831268.0,
      "step": 53210
    },
    {
      "entropy": 5.787882614135742,
      "epoch": 4.650441317836232,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030294769006907585,
      "loss": 5.2872,
      "mean_token_accuracy": 0.20929391980171203,
      "num_tokens": 107841284.0,
      "step": 53215
    },
    {
      "entropy": 5.770891380310059,
      "epoch": 4.650878266188936,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003029167700237904,
      "loss": 5.2174,
      "mean_token_accuracy": 0.2153395816683769,
      "num_tokens": 107853119.0,
      "step": 53220
    },
    {
      "entropy": 5.783117866516113,
      "epoch": 4.651315214541642,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030288584944304674,
      "loss": 5.183,
      "mean_token_accuracy": 0.21534324139356614,
      "num_tokens": 107864162.0,
      "step": 53225
    },
    {
      "entropy": 5.733302974700928,
      "epoch": 4.651752162894346,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003028549283274378,
      "loss": 5.1437,
      "mean_token_accuracy": 0.21579121351242064,
      "num_tokens": 107874921.0,
      "step": 53230
    },
    {
      "entropy": 5.721620845794678,
      "epoch": 4.652189111247051,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003028240066775568,
      "loss": 5.26,
      "mean_token_accuracy": 0.2036773681640625,
      "num_tokens": 107885841.0,
      "step": 53235
    },
    {
      "entropy": 5.72726845741272,
      "epoch": 4.652626059599755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003027930844939968,
      "loss": 5.2192,
      "mean_token_accuracy": 0.2095681235194206,
      "num_tokens": 107896651.0,
      "step": 53240
    },
    {
      "entropy": 5.7515030860900875,
      "epoch": 4.6530630079524595,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030276216177735083,
      "loss": 5.2407,
      "mean_token_accuracy": 0.20577987432479858,
      "num_tokens": 107907540.0,
      "step": 53245
    },
    {
      "entropy": 5.780457639694214,
      "epoch": 4.653499956305165,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030273123852821203,
      "loss": 5.146,
      "mean_token_accuracy": 0.21241679191589355,
      "num_tokens": 107916894.0,
      "step": 53250
    },
    {
      "entropy": 5.740335178375244,
      "epoch": 4.653936904657869,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030270031474717357,
      "loss": 5.2589,
      "mean_token_accuracy": 0.20401499122381211,
      "num_tokens": 107927160.0,
      "step": 53255
    },
    {
      "entropy": 5.749779462814331,
      "epoch": 4.654373853010574,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003026693904348286,
      "loss": 5.2627,
      "mean_token_accuracy": 0.2019418865442276,
      "num_tokens": 107938511.0,
      "step": 53260
    },
    {
      "entropy": 5.747396850585938,
      "epoch": 4.654810801363279,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003026384655917702,
      "loss": 5.1732,
      "mean_token_accuracy": 0.21075353473424913,
      "num_tokens": 107948942.0,
      "step": 53265
    },
    {
      "entropy": 5.774463748931884,
      "epoch": 4.655247749715984,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030260754021859147,
      "loss": 5.2491,
      "mean_token_accuracy": 0.21039875000715255,
      "num_tokens": 107959768.0,
      "step": 53270
    },
    {
      "entropy": 5.778703212738037,
      "epoch": 4.655684698068688,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003025766143158858,
      "loss": 5.2003,
      "mean_token_accuracy": 0.21168750673532485,
      "num_tokens": 107970956.0,
      "step": 53275
    },
    {
      "entropy": 5.781962823867798,
      "epoch": 4.6561216464213935,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003025456878842461,
      "loss": 5.2368,
      "mean_token_accuracy": 0.21273786574602127,
      "num_tokens": 107980786.0,
      "step": 53280
    },
    {
      "entropy": 5.724166917800903,
      "epoch": 4.656558594774098,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030251476092426575,
      "loss": 5.2638,
      "mean_token_accuracy": 0.2062551736831665,
      "num_tokens": 107990578.0,
      "step": 53285
    },
    {
      "entropy": 5.726508665084839,
      "epoch": 4.656995543126802,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003024838334365378,
      "loss": 5.2391,
      "mean_token_accuracy": 0.20704239308834077,
      "num_tokens": 108000417.0,
      "step": 53290
    },
    {
      "entropy": 5.7329202651977536,
      "epoch": 4.657432491479507,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030245290542165555,
      "loss": 5.1937,
      "mean_token_accuracy": 0.21342357397079467,
      "num_tokens": 108010145.0,
      "step": 53295
    },
    {
      "entropy": 5.695567512512207,
      "epoch": 4.657869439832211,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003024219768802122,
      "loss": 5.1829,
      "mean_token_accuracy": 0.21349081248044968,
      "num_tokens": 108020510.0,
      "step": 53300
    },
    {
      "entropy": 5.7397544384002686,
      "epoch": 4.658306388184917,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003023910478128009,
      "loss": 5.2266,
      "mean_token_accuracy": 0.2134527638554573,
      "num_tokens": 108029763.0,
      "step": 53305
    },
    {
      "entropy": 5.792538261413574,
      "epoch": 4.658743336537621,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030236011822001505,
      "loss": 5.2892,
      "mean_token_accuracy": 0.20400997549295424,
      "num_tokens": 108039944.0,
      "step": 53310
    },
    {
      "entropy": 5.802494955062866,
      "epoch": 4.659180284890326,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003023291881024477,
      "loss": 5.2606,
      "mean_token_accuracy": 0.21040269434452058,
      "num_tokens": 108051030.0,
      "step": 53315
    },
    {
      "entropy": 5.729314661026001,
      "epoch": 4.659617233243031,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030229825746069225,
      "loss": 5.1318,
      "mean_token_accuracy": 0.22166870385408402,
      "num_tokens": 108061537.0,
      "step": 53320
    },
    {
      "entropy": 5.7056684494018555,
      "epoch": 4.6600541815957355,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003022673262953419,
      "loss": 5.1613,
      "mean_token_accuracy": 0.21465088278055192,
      "num_tokens": 108072142.0,
      "step": 53325
    },
    {
      "entropy": 5.697365140914917,
      "epoch": 4.66049112994844,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030223639460699,
      "loss": 5.2188,
      "mean_token_accuracy": 0.20691795498132706,
      "num_tokens": 108083201.0,
      "step": 53330
    },
    {
      "entropy": 5.72548999786377,
      "epoch": 4.660928078301144,
      "grad_norm": 1.125,
      "learning_rate": 0.0003022054623962297,
      "loss": 5.1888,
      "mean_token_accuracy": 0.21307158917188646,
      "num_tokens": 108093175.0,
      "step": 53335
    },
    {
      "entropy": 5.739467668533325,
      "epoch": 4.66136502665385,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030217452966365446,
      "loss": 5.1531,
      "mean_token_accuracy": 0.22367108166217803,
      "num_tokens": 108103918.0,
      "step": 53340
    },
    {
      "entropy": 5.701328611373901,
      "epoch": 4.661801975006554,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003021435964098574,
      "loss": 5.1933,
      "mean_token_accuracy": 0.2154175817966461,
      "num_tokens": 108115060.0,
      "step": 53345
    },
    {
      "entropy": 5.705983066558838,
      "epoch": 4.662238923359259,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030211266263543204,
      "loss": 5.1308,
      "mean_token_accuracy": 0.21867304295301437,
      "num_tokens": 108124262.0,
      "step": 53350
    },
    {
      "entropy": 5.715392827987671,
      "epoch": 4.662675871711963,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030208172834097157,
      "loss": 5.1548,
      "mean_token_accuracy": 0.2172219783067703,
      "num_tokens": 108132852.0,
      "step": 53355
    },
    {
      "entropy": 5.65280499458313,
      "epoch": 4.6631128200646685,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030205079352706935,
      "loss": 5.1457,
      "mean_token_accuracy": 0.2194264993071556,
      "num_tokens": 108142533.0,
      "step": 53360
    },
    {
      "entropy": 5.692742586135864,
      "epoch": 4.663549768417373,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003020198581943188,
      "loss": 5.1528,
      "mean_token_accuracy": 0.21569646149873734,
      "num_tokens": 108151781.0,
      "step": 53365
    },
    {
      "entropy": 5.8207732200622555,
      "epoch": 4.6639867167700775,
      "grad_norm": 1.0,
      "learning_rate": 0.0003019889223433131,
      "loss": 5.2698,
      "mean_token_accuracy": 0.20277624875307082,
      "num_tokens": 108162005.0,
      "step": 53370
    },
    {
      "entropy": 5.782744884490967,
      "epoch": 4.664423665122783,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030195798597464573,
      "loss": 5.2799,
      "mean_token_accuracy": 0.21114180237054825,
      "num_tokens": 108172009.0,
      "step": 53375
    },
    {
      "entropy": 5.75708384513855,
      "epoch": 4.664860613475487,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003019270490889101,
      "loss": 5.217,
      "mean_token_accuracy": 0.21136425137519838,
      "num_tokens": 108181584.0,
      "step": 53380
    },
    {
      "entropy": 5.761505270004273,
      "epoch": 4.665297561828192,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003018961116866996,
      "loss": 5.212,
      "mean_token_accuracy": 0.21280044913291932,
      "num_tokens": 108190846.0,
      "step": 53385
    },
    {
      "entropy": 5.762305831909179,
      "epoch": 4.665734510180896,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030186517376860753,
      "loss": 5.1507,
      "mean_token_accuracy": 0.2137746572494507,
      "num_tokens": 108199650.0,
      "step": 53390
    },
    {
      "entropy": 5.788375949859619,
      "epoch": 4.666171458533602,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003018342353352274,
      "loss": 5.3071,
      "mean_token_accuracy": 0.20400397330522538,
      "num_tokens": 108210294.0,
      "step": 53395
    },
    {
      "entropy": 5.804459953308106,
      "epoch": 4.666608406886306,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030180329638715246,
      "loss": 5.2557,
      "mean_token_accuracy": 0.20529559701681138,
      "num_tokens": 108218981.0,
      "step": 53400
    },
    {
      "entropy": 5.737042379379273,
      "epoch": 4.6670453552390105,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003017723569249764,
      "loss": 5.174,
      "mean_token_accuracy": 0.2166825130581856,
      "num_tokens": 108229613.0,
      "step": 53405
    },
    {
      "entropy": 5.711775875091552,
      "epoch": 4.667482303591715,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030174141694929243,
      "loss": 5.1945,
      "mean_token_accuracy": 0.21470102667808533,
      "num_tokens": 108239269.0,
      "step": 53410
    },
    {
      "entropy": 5.698837089538574,
      "epoch": 4.66791925194442,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030171047646069404,
      "loss": 5.1857,
      "mean_token_accuracy": 0.2080681398510933,
      "num_tokens": 108248942.0,
      "step": 53415
    },
    {
      "entropy": 5.757335519790649,
      "epoch": 4.668356200297125,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003016795354597748,
      "loss": 5.2256,
      "mean_token_accuracy": 0.20746346861124038,
      "num_tokens": 108260479.0,
      "step": 53420
    },
    {
      "entropy": 5.719735193252563,
      "epoch": 4.668793148649829,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003016485939471281,
      "loss": 5.195,
      "mean_token_accuracy": 0.216010981798172,
      "num_tokens": 108269530.0,
      "step": 53425
    },
    {
      "entropy": 5.732980489730835,
      "epoch": 4.669230097002535,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030161765192334735,
      "loss": 5.2068,
      "mean_token_accuracy": 0.21978131234645842,
      "num_tokens": 108280650.0,
      "step": 53430
    },
    {
      "entropy": 5.79211049079895,
      "epoch": 4.669667045355239,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030158670938902605,
      "loss": 5.1664,
      "mean_token_accuracy": 0.21517403870821,
      "num_tokens": 108290266.0,
      "step": 53435
    },
    {
      "entropy": 5.648149108886718,
      "epoch": 4.670103993707944,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003015557663447578,
      "loss": 5.1448,
      "mean_token_accuracy": 0.21989131420850755,
      "num_tokens": 108300911.0,
      "step": 53440
    },
    {
      "entropy": 5.662575960159302,
      "epoch": 4.670540942060648,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000301524822791136,
      "loss": 5.1146,
      "mean_token_accuracy": 0.22117913514375687,
      "num_tokens": 108311000.0,
      "step": 53445
    },
    {
      "entropy": 5.729001474380493,
      "epoch": 4.670977890413353,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003014938787287542,
      "loss": 5.1845,
      "mean_token_accuracy": 0.21641404628753663,
      "num_tokens": 108321817.0,
      "step": 53450
    },
    {
      "entropy": 5.79900050163269,
      "epoch": 4.671414838766058,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030146293415820593,
      "loss": 5.3148,
      "mean_token_accuracy": 0.20372175574302673,
      "num_tokens": 108332917.0,
      "step": 53455
    },
    {
      "entropy": 5.713973522186279,
      "epoch": 4.671851787118762,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003014319890800847,
      "loss": 5.1147,
      "mean_token_accuracy": 0.2188531845808029,
      "num_tokens": 108344013.0,
      "step": 53460
    },
    {
      "entropy": 5.808024930953979,
      "epoch": 4.672288735471467,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003014010434949841,
      "loss": 5.2984,
      "mean_token_accuracy": 0.2020520895719528,
      "num_tokens": 108353849.0,
      "step": 53465
    },
    {
      "entropy": 5.772624158859253,
      "epoch": 4.672725683824172,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030137009740349767,
      "loss": 5.2582,
      "mean_token_accuracy": 0.21039140671491624,
      "num_tokens": 108363437.0,
      "step": 53470
    },
    {
      "entropy": 5.7100941181182865,
      "epoch": 4.673162632176877,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030133915080621894,
      "loss": 5.1915,
      "mean_token_accuracy": 0.21500123292207718,
      "num_tokens": 108372477.0,
      "step": 53475
    },
    {
      "entropy": 5.746136903762817,
      "epoch": 4.673599580529581,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030130820370374145,
      "loss": 5.2219,
      "mean_token_accuracy": 0.21267807930707933,
      "num_tokens": 108382693.0,
      "step": 53480
    },
    {
      "entropy": 5.700245666503906,
      "epoch": 4.6740365288822865,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003012772560966588,
      "loss": 5.1025,
      "mean_token_accuracy": 0.22137167006731034,
      "num_tokens": 108392044.0,
      "step": 53485
    },
    {
      "entropy": 5.679703187942505,
      "epoch": 4.674473477234991,
      "grad_norm": 1.0,
      "learning_rate": 0.00030124630798556475,
      "loss": 5.1094,
      "mean_token_accuracy": 0.2172573134303093,
      "num_tokens": 108401541.0,
      "step": 53490
    },
    {
      "entropy": 5.666904354095459,
      "epoch": 4.674910425587695,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003012153593710527,
      "loss": 5.1274,
      "mean_token_accuracy": 0.22033784091472625,
      "num_tokens": 108412337.0,
      "step": 53495
    },
    {
      "entropy": 5.748858118057251,
      "epoch": 4.6753473739404,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003011844102537163,
      "loss": 5.2235,
      "mean_token_accuracy": 0.207165989279747,
      "num_tokens": 108421965.0,
      "step": 53500
    },
    {
      "entropy": 5.650922727584839,
      "epoch": 4.675784322293105,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003011534606341492,
      "loss": 5.1263,
      "mean_token_accuracy": 0.22389203310012817,
      "num_tokens": 108432380.0,
      "step": 53505
    },
    {
      "entropy": 5.627721977233887,
      "epoch": 4.67622127064581,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000301122510512945,
      "loss": 5.1071,
      "mean_token_accuracy": 0.21570996791124344,
      "num_tokens": 108441893.0,
      "step": 53510
    },
    {
      "entropy": 5.688503074645996,
      "epoch": 4.676658218998514,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003010915598906973,
      "loss": 5.1333,
      "mean_token_accuracy": 0.22515340149402618,
      "num_tokens": 108452188.0,
      "step": 53515
    },
    {
      "entropy": 5.744128084182739,
      "epoch": 4.677095167351219,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030106060876799995,
      "loss": 5.1545,
      "mean_token_accuracy": 0.22231631875038146,
      "num_tokens": 108462071.0,
      "step": 53520
    },
    {
      "entropy": 5.6377912044525145,
      "epoch": 4.677532115703924,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003010296571454464,
      "loss": 5.0815,
      "mean_token_accuracy": 0.2269311398267746,
      "num_tokens": 108470988.0,
      "step": 53525
    },
    {
      "entropy": 5.722366046905518,
      "epoch": 4.6779690640566285,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030099870502363037,
      "loss": 5.2202,
      "mean_token_accuracy": 0.20798081308603286,
      "num_tokens": 108480534.0,
      "step": 53530
    },
    {
      "entropy": 5.805191516876221,
      "epoch": 4.678406012409333,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003009677524031456,
      "loss": 5.2464,
      "mean_token_accuracy": 0.209509913623333,
      "num_tokens": 108491311.0,
      "step": 53535
    },
    {
      "entropy": 5.824076461791992,
      "epoch": 4.678842960762038,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003009367992845857,
      "loss": 5.2235,
      "mean_token_accuracy": 0.21897540539503096,
      "num_tokens": 108501407.0,
      "step": 53540
    },
    {
      "entropy": 5.757824277877807,
      "epoch": 4.679279909114743,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003009058456685444,
      "loss": 5.2791,
      "mean_token_accuracy": 0.20089850723743438,
      "num_tokens": 108511149.0,
      "step": 53545
    },
    {
      "entropy": 5.704613161087036,
      "epoch": 4.679716857467447,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003008748915556154,
      "loss": 5.2048,
      "mean_token_accuracy": 0.2114155724644661,
      "num_tokens": 108521224.0,
      "step": 53550
    },
    {
      "entropy": 5.746368646621704,
      "epoch": 4.680153805820152,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003008439369463924,
      "loss": 5.1791,
      "mean_token_accuracy": 0.213474839925766,
      "num_tokens": 108530860.0,
      "step": 53555
    },
    {
      "entropy": 5.70315408706665,
      "epoch": 4.680590754172857,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003008129818414692,
      "loss": 5.2478,
      "mean_token_accuracy": 0.20861099362373353,
      "num_tokens": 108540460.0,
      "step": 53560
    },
    {
      "entropy": 5.771579170227051,
      "epoch": 4.6810277025255616,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030078202624143945,
      "loss": 5.204,
      "mean_token_accuracy": 0.20722212940454482,
      "num_tokens": 108550588.0,
      "step": 53565
    },
    {
      "entropy": 5.753455924987793,
      "epoch": 4.681464650878266,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030075107014689695,
      "loss": 5.2153,
      "mean_token_accuracy": 0.21603113263845444,
      "num_tokens": 108561107.0,
      "step": 53570
    },
    {
      "entropy": 5.765408992767334,
      "epoch": 4.6819015992309705,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003007201135584354,
      "loss": 5.2309,
      "mean_token_accuracy": 0.20575778186321259,
      "num_tokens": 108571834.0,
      "step": 53575
    },
    {
      "entropy": 5.804080009460449,
      "epoch": 4.682338547583676,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003006891564766486,
      "loss": 5.2542,
      "mean_token_accuracy": 0.2055194556713104,
      "num_tokens": 108581264.0,
      "step": 53580
    },
    {
      "entropy": 5.83553261756897,
      "epoch": 4.68277549593638,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030065819890213037,
      "loss": 5.2943,
      "mean_token_accuracy": 0.2038124457001686,
      "num_tokens": 108591302.0,
      "step": 53585
    },
    {
      "entropy": 5.7351672649383545,
      "epoch": 4.683212444289085,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030062724083547437,
      "loss": 5.1777,
      "mean_token_accuracy": 0.21657567471265793,
      "num_tokens": 108601834.0,
      "step": 53590
    },
    {
      "entropy": 5.782975053787231,
      "epoch": 4.68364939264179,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003005962822772745,
      "loss": 5.265,
      "mean_token_accuracy": 0.20796941220760345,
      "num_tokens": 108612455.0,
      "step": 53595
    },
    {
      "entropy": 5.742186164855957,
      "epoch": 4.684086340994495,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003005653232281245,
      "loss": 5.1772,
      "mean_token_accuracy": 0.2143015131354332,
      "num_tokens": 108622999.0,
      "step": 53600
    },
    {
      "entropy": 5.72601432800293,
      "epoch": 4.684523289347199,
      "grad_norm": 1.0,
      "learning_rate": 0.0003005343636886182,
      "loss": 5.1781,
      "mean_token_accuracy": 0.21491851955652236,
      "num_tokens": 108633710.0,
      "step": 53605
    },
    {
      "entropy": 5.679214000701904,
      "epoch": 4.6849602376999036,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003005034036593495,
      "loss": 5.1186,
      "mean_token_accuracy": 0.20851499885320662,
      "num_tokens": 108644128.0,
      "step": 53610
    },
    {
      "entropy": 5.720923376083374,
      "epoch": 4.685397186052609,
      "grad_norm": 1.125,
      "learning_rate": 0.0003004724431409121,
      "loss": 5.16,
      "mean_token_accuracy": 0.2147548407316208,
      "num_tokens": 108654259.0,
      "step": 53615
    },
    {
      "entropy": 5.738364410400391,
      "epoch": 4.685834134405313,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003004414821338999,
      "loss": 5.1846,
      "mean_token_accuracy": 0.21341679990291595,
      "num_tokens": 108664835.0,
      "step": 53620
    },
    {
      "entropy": 5.68439621925354,
      "epoch": 4.686271082758018,
      "grad_norm": 1.125,
      "learning_rate": 0.0003004105206389067,
      "loss": 5.1731,
      "mean_token_accuracy": 0.21480657011270524,
      "num_tokens": 108675083.0,
      "step": 53625
    },
    {
      "entropy": 5.735680866241455,
      "epoch": 4.686708031110722,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030037955865652645,
      "loss": 5.1893,
      "mean_token_accuracy": 0.21124779134988786,
      "num_tokens": 108685295.0,
      "step": 53630
    },
    {
      "entropy": 5.677998971939087,
      "epoch": 4.687144979463428,
      "grad_norm": 1.140625,
      "learning_rate": 0.000300348596187353,
      "loss": 5.1669,
      "mean_token_accuracy": 0.21731216758489608,
      "num_tokens": 108694957.0,
      "step": 53635
    },
    {
      "entropy": 5.761869096755982,
      "epoch": 4.687581927816132,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030031763323198007,
      "loss": 5.2569,
      "mean_token_accuracy": 0.21621566712856294,
      "num_tokens": 108705410.0,
      "step": 53640
    },
    {
      "entropy": 5.781038379669189,
      "epoch": 4.688018876168837,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003002866697910018,
      "loss": 5.1831,
      "mean_token_accuracy": 0.21870249956846238,
      "num_tokens": 108715189.0,
      "step": 53645
    },
    {
      "entropy": 5.769473361968994,
      "epoch": 4.688455824521542,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003002557058650118,
      "loss": 5.2711,
      "mean_token_accuracy": 0.20502174347639085,
      "num_tokens": 108725225.0,
      "step": 53650
    },
    {
      "entropy": 5.789599132537842,
      "epoch": 4.688892772874246,
      "grad_norm": 1.0,
      "learning_rate": 0.0003002247414546043,
      "loss": 5.2659,
      "mean_token_accuracy": 0.20529907047748566,
      "num_tokens": 108735933.0,
      "step": 53655
    },
    {
      "entropy": 5.789352083206177,
      "epoch": 4.689329721226951,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030019377656037295,
      "loss": 5.2315,
      "mean_token_accuracy": 0.20910128057003022,
      "num_tokens": 108745336.0,
      "step": 53660
    },
    {
      "entropy": 5.73815131187439,
      "epoch": 4.689766669579655,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003001628111829118,
      "loss": 5.2131,
      "mean_token_accuracy": 0.20633162707090377,
      "num_tokens": 108756583.0,
      "step": 53665
    },
    {
      "entropy": 5.719290924072266,
      "epoch": 4.690203617932361,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030013184532281465,
      "loss": 5.2254,
      "mean_token_accuracy": 0.2095789432525635,
      "num_tokens": 108766606.0,
      "step": 53670
    },
    {
      "entropy": 5.636775159835816,
      "epoch": 4.690640566285065,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030010087898067566,
      "loss": 5.139,
      "mean_token_accuracy": 0.22526713013648986,
      "num_tokens": 108776803.0,
      "step": 53675
    },
    {
      "entropy": 5.7240620136260985,
      "epoch": 4.69107751463777,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030006991215708855,
      "loss": 5.219,
      "mean_token_accuracy": 0.21394059360027312,
      "num_tokens": 108787033.0,
      "step": 53680
    },
    {
      "entropy": 5.703849220275879,
      "epoch": 4.691514462990474,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030003894485264754,
      "loss": 5.1065,
      "mean_token_accuracy": 0.22339381128549576,
      "num_tokens": 108797593.0,
      "step": 53685
    },
    {
      "entropy": 5.699637985229492,
      "epoch": 4.6919514113431795,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030000797706794633,
      "loss": 5.176,
      "mean_token_accuracy": 0.2156512290239334,
      "num_tokens": 108807065.0,
      "step": 53690
    },
    {
      "entropy": 5.704955053329468,
      "epoch": 4.692388359695884,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029997700880357906,
      "loss": 5.1443,
      "mean_token_accuracy": 0.2192848652601242,
      "num_tokens": 108816689.0,
      "step": 53695
    },
    {
      "entropy": 5.714033317565918,
      "epoch": 4.692825308048588,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029994604006013966,
      "loss": 5.1913,
      "mean_token_accuracy": 0.220462828874588,
      "num_tokens": 108826630.0,
      "step": 53700
    },
    {
      "entropy": 5.68388295173645,
      "epoch": 4.693262256401294,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002999150708382221,
      "loss": 5.1254,
      "mean_token_accuracy": 0.22242858558893203,
      "num_tokens": 108836268.0,
      "step": 53705
    },
    {
      "entropy": 5.758709144592285,
      "epoch": 4.693699204753998,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002998841011384205,
      "loss": 5.1549,
      "mean_token_accuracy": 0.2163284480571747,
      "num_tokens": 108845865.0,
      "step": 53710
    },
    {
      "entropy": 5.751278114318848,
      "epoch": 4.694136153106703,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002998531309613288,
      "loss": 5.172,
      "mean_token_accuracy": 0.21026254296302796,
      "num_tokens": 108855628.0,
      "step": 53715
    },
    {
      "entropy": 5.729598045349121,
      "epoch": 4.694573101459407,
      "grad_norm": 1.125,
      "learning_rate": 0.00029982216030754094,
      "loss": 5.2204,
      "mean_token_accuracy": 0.20747970640659333,
      "num_tokens": 108866323.0,
      "step": 53720
    },
    {
      "entropy": 5.8481401920318605,
      "epoch": 4.695010049812113,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002997911891776511,
      "loss": 5.2868,
      "mean_token_accuracy": 0.2042912483215332,
      "num_tokens": 108876783.0,
      "step": 53725
    },
    {
      "entropy": 5.721418762207032,
      "epoch": 4.695446998164817,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029976021757225317,
      "loss": 5.1625,
      "mean_token_accuracy": 0.21694762110710145,
      "num_tokens": 108887171.0,
      "step": 53730
    },
    {
      "entropy": 5.721096229553223,
      "epoch": 4.6958839465175215,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002997292454919415,
      "loss": 5.112,
      "mean_token_accuracy": 0.21422039270401,
      "num_tokens": 108897441.0,
      "step": 53735
    },
    {
      "entropy": 5.7190906524658205,
      "epoch": 4.696320894870226,
      "grad_norm": 1.125,
      "learning_rate": 0.00029969827293730973,
      "loss": 5.1664,
      "mean_token_accuracy": 0.21071786284446717,
      "num_tokens": 108906608.0,
      "step": 53740
    },
    {
      "entropy": 5.771175765991211,
      "epoch": 4.696757843222931,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029966729990895226,
      "loss": 5.2239,
      "mean_token_accuracy": 0.21549385339021682,
      "num_tokens": 108917038.0,
      "step": 53745
    },
    {
      "entropy": 5.785371541976929,
      "epoch": 4.697194791575636,
      "grad_norm": 1.171875,
      "learning_rate": 0.000299636326407463,
      "loss": 5.2538,
      "mean_token_accuracy": 0.20671392381191253,
      "num_tokens": 108926682.0,
      "step": 53750
    },
    {
      "entropy": 5.688692665100097,
      "epoch": 4.69763173992834,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029960535243343616,
      "loss": 5.1815,
      "mean_token_accuracy": 0.21813137978315353,
      "num_tokens": 108936936.0,
      "step": 53755
    },
    {
      "entropy": 5.777227210998535,
      "epoch": 4.698068688281046,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002995743779874657,
      "loss": 5.2213,
      "mean_token_accuracy": 0.20931578576564788,
      "num_tokens": 108947635.0,
      "step": 53760
    },
    {
      "entropy": 5.733997774124146,
      "epoch": 4.69850563663375,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002995434030701457,
      "loss": 5.1404,
      "mean_token_accuracy": 0.2182831346988678,
      "num_tokens": 108957085.0,
      "step": 53765
    },
    {
      "entropy": 5.702873182296753,
      "epoch": 4.698942584986455,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002995124276820705,
      "loss": 5.2191,
      "mean_token_accuracy": 0.20866167694330215,
      "num_tokens": 108966639.0,
      "step": 53770
    },
    {
      "entropy": 5.742829370498657,
      "epoch": 4.699379533339159,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029948145182383414,
      "loss": 5.1449,
      "mean_token_accuracy": 0.2180673062801361,
      "num_tokens": 108977300.0,
      "step": 53775
    },
    {
      "entropy": 5.7496785640716555,
      "epoch": 4.699816481691864,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002994504754960306,
      "loss": 5.2387,
      "mean_token_accuracy": 0.21482320129871368,
      "num_tokens": 108987281.0,
      "step": 53780
    },
    {
      "entropy": 5.776463508605957,
      "epoch": 4.700253430044569,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029941949869925424,
      "loss": 5.2182,
      "mean_token_accuracy": 0.21054184734821318,
      "num_tokens": 108996676.0,
      "step": 53785
    },
    {
      "entropy": 5.715813779830933,
      "epoch": 4.700690378397273,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029938852143409896,
      "loss": 5.2164,
      "mean_token_accuracy": 0.21470339447259904,
      "num_tokens": 109006633.0,
      "step": 53790
    },
    {
      "entropy": 5.721436929702759,
      "epoch": 4.701127326749978,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029935754370115906,
      "loss": 5.1302,
      "mean_token_accuracy": 0.2131837785243988,
      "num_tokens": 109016876.0,
      "step": 53795
    },
    {
      "entropy": 5.734802961349487,
      "epoch": 4.701564275102683,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029932656550102886,
      "loss": 5.1515,
      "mean_token_accuracy": 0.21687671840190886,
      "num_tokens": 109027107.0,
      "step": 53800
    },
    {
      "entropy": 5.817874813079834,
      "epoch": 4.702001223455388,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002992955868343023,
      "loss": 5.2601,
      "mean_token_accuracy": 0.20413547456264497,
      "num_tokens": 109037809.0,
      "step": 53805
    },
    {
      "entropy": 5.717866992950439,
      "epoch": 4.702438171808092,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002992646077015737,
      "loss": 5.2272,
      "mean_token_accuracy": 0.20713620632886887,
      "num_tokens": 109048913.0,
      "step": 53810
    },
    {
      "entropy": 5.699159049987793,
      "epoch": 4.7028751201607975,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029923362810343714,
      "loss": 5.1613,
      "mean_token_accuracy": 0.2105081483721733,
      "num_tokens": 109059482.0,
      "step": 53815
    },
    {
      "entropy": 5.7557721614837645,
      "epoch": 4.703312068513502,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029920264804048687,
      "loss": 5.2497,
      "mean_token_accuracy": 0.20907963067293167,
      "num_tokens": 109069236.0,
      "step": 53820
    },
    {
      "entropy": 5.745034980773926,
      "epoch": 4.703749016866206,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002991716675133172,
      "loss": 5.2148,
      "mean_token_accuracy": 0.20643844306468964,
      "num_tokens": 109079215.0,
      "step": 53825
    },
    {
      "entropy": 5.73418869972229,
      "epoch": 4.704185965218911,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029914068652252225,
      "loss": 5.2086,
      "mean_token_accuracy": 0.20743298083543776,
      "num_tokens": 109088695.0,
      "step": 53830
    },
    {
      "entropy": 5.758302450180054,
      "epoch": 4.704622913571616,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029910970506869634,
      "loss": 5.2722,
      "mean_token_accuracy": 0.20754646360874177,
      "num_tokens": 109099203.0,
      "step": 53835
    },
    {
      "entropy": 5.868995332717896,
      "epoch": 4.705059861924321,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029907872315243354,
      "loss": 5.3492,
      "mean_token_accuracy": 0.19957054555416107,
      "num_tokens": 109110917.0,
      "step": 53840
    },
    {
      "entropy": 5.773767137527466,
      "epoch": 4.705496810277025,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029904774077432826,
      "loss": 5.2056,
      "mean_token_accuracy": 0.21079551726579665,
      "num_tokens": 109121277.0,
      "step": 53845
    },
    {
      "entropy": 5.719046545028687,
      "epoch": 4.70593375862973,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002990167579349748,
      "loss": 5.1474,
      "mean_token_accuracy": 0.21297656297683715,
      "num_tokens": 109130485.0,
      "step": 53850
    },
    {
      "entropy": 5.677299308776855,
      "epoch": 4.706370706982435,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002989857746349672,
      "loss": 5.1347,
      "mean_token_accuracy": 0.21545421779155732,
      "num_tokens": 109139872.0,
      "step": 53855
    },
    {
      "entropy": 5.75372200012207,
      "epoch": 4.7068076553351395,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029895479087489995,
      "loss": 5.2329,
      "mean_token_accuracy": 0.2078666567802429,
      "num_tokens": 109150004.0,
      "step": 53860
    },
    {
      "entropy": 5.743498706817627,
      "epoch": 4.707244603687844,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029892380665536727,
      "loss": 5.23,
      "mean_token_accuracy": 0.20837990790605546,
      "num_tokens": 109160078.0,
      "step": 53865
    },
    {
      "entropy": 5.781103849411011,
      "epoch": 4.707681552040549,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002988928219769634,
      "loss": 5.1669,
      "mean_token_accuracy": 0.21253733783960344,
      "num_tokens": 109168829.0,
      "step": 53870
    },
    {
      "entropy": 5.623613739013672,
      "epoch": 4.708118500393254,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029886183684028276,
      "loss": 5.1042,
      "mean_token_accuracy": 0.22429848462343216,
      "num_tokens": 109178909.0,
      "step": 53875
    },
    {
      "entropy": 5.769679164886474,
      "epoch": 4.708555448745958,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002988308512459196,
      "loss": 5.2376,
      "mean_token_accuracy": 0.20964901447296141,
      "num_tokens": 109188156.0,
      "step": 53880
    },
    {
      "entropy": 5.806567049026489,
      "epoch": 4.708992397098663,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029879986519446814,
      "loss": 5.2627,
      "mean_token_accuracy": 0.20743064284324647,
      "num_tokens": 109198777.0,
      "step": 53885
    },
    {
      "entropy": 5.7335868835449215,
      "epoch": 4.709429345451367,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029876887868652283,
      "loss": 5.1983,
      "mean_token_accuracy": 0.2121279090642929,
      "num_tokens": 109207942.0,
      "step": 53890
    },
    {
      "entropy": 5.684366607666016,
      "epoch": 4.7098662938040725,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029873789172267794,
      "loss": 5.1554,
      "mean_token_accuracy": 0.21445282995700837,
      "num_tokens": 109217636.0,
      "step": 53895
    },
    {
      "entropy": 5.739268589019775,
      "epoch": 4.710303242156777,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002987069043035279,
      "loss": 5.2073,
      "mean_token_accuracy": 0.2121238201856613,
      "num_tokens": 109228379.0,
      "step": 53900
    },
    {
      "entropy": 5.722348022460937,
      "epoch": 4.7107401905094815,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029867591642966694,
      "loss": 5.1613,
      "mean_token_accuracy": 0.21996962577104567,
      "num_tokens": 109237750.0,
      "step": 53905
    },
    {
      "entropy": 5.773005485534668,
      "epoch": 4.711177138862187,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029864492810168945,
      "loss": 5.2282,
      "mean_token_accuracy": 0.20871153026819228,
      "num_tokens": 109248411.0,
      "step": 53910
    },
    {
      "entropy": 5.7534647464752195,
      "epoch": 4.711614087214891,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029861393932018997,
      "loss": 5.2056,
      "mean_token_accuracy": 0.2118563562631607,
      "num_tokens": 109257495.0,
      "step": 53915
    },
    {
      "entropy": 5.708182048797608,
      "epoch": 4.712051035567596,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002985829500857628,
      "loss": 5.1959,
      "mean_token_accuracy": 0.22012839615345,
      "num_tokens": 109267003.0,
      "step": 53920
    },
    {
      "entropy": 5.673859119415283,
      "epoch": 4.712487983920301,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029855196039900217,
      "loss": 5.1408,
      "mean_token_accuracy": 0.21906914561986923,
      "num_tokens": 109276336.0,
      "step": 53925
    },
    {
      "entropy": 5.781564807891845,
      "epoch": 4.712924932273006,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002985209702605027,
      "loss": 5.2403,
      "mean_token_accuracy": 0.2075158029794693,
      "num_tokens": 109286639.0,
      "step": 53930
    },
    {
      "entropy": 5.7563597679138185,
      "epoch": 4.71336188062571,
      "grad_norm": 1.125,
      "learning_rate": 0.0002984899796708586,
      "loss": 5.1359,
      "mean_token_accuracy": 0.21931229084730147,
      "num_tokens": 109296994.0,
      "step": 53935
    },
    {
      "entropy": 5.756929159164429,
      "epoch": 4.7137988289784145,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002984589886306644,
      "loss": 5.25,
      "mean_token_accuracy": 0.20511816442012787,
      "num_tokens": 109306729.0,
      "step": 53940
    },
    {
      "entropy": 5.707530117034912,
      "epoch": 4.714235777331119,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002984279971405145,
      "loss": 5.2697,
      "mean_token_accuracy": 0.2074722722172737,
      "num_tokens": 109318664.0,
      "step": 53945
    },
    {
      "entropy": 5.814471244812012,
      "epoch": 4.714672725683824,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002983970052010033,
      "loss": 5.2112,
      "mean_token_accuracy": 0.2074819579720497,
      "num_tokens": 109327950.0,
      "step": 53950
    },
    {
      "entropy": 5.754384517669678,
      "epoch": 4.715109674036529,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029836601281272543,
      "loss": 5.164,
      "mean_token_accuracy": 0.21457447111606598,
      "num_tokens": 109338046.0,
      "step": 53955
    },
    {
      "entropy": 5.7314332008361815,
      "epoch": 4.715546622389233,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000298335019976275,
      "loss": 5.1698,
      "mean_token_accuracy": 0.2159167066216469,
      "num_tokens": 109348738.0,
      "step": 53960
    },
    {
      "entropy": 5.796203136444092,
      "epoch": 4.715983570741939,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002983040266922469,
      "loss": 5.24,
      "mean_token_accuracy": 0.20493671745061875,
      "num_tokens": 109358729.0,
      "step": 53965
    },
    {
      "entropy": 5.765652513504028,
      "epoch": 4.716420519094643,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002982730329612352,
      "loss": 5.2073,
      "mean_token_accuracy": 0.20828462094068528,
      "num_tokens": 109368912.0,
      "step": 53970
    },
    {
      "entropy": 5.738517141342163,
      "epoch": 4.716857467447348,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029824203878383454,
      "loss": 5.1957,
      "mean_token_accuracy": 0.21299307644367219,
      "num_tokens": 109378703.0,
      "step": 53975
    },
    {
      "entropy": 5.788786602020264,
      "epoch": 4.717294415800053,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002982110441606394,
      "loss": 5.2563,
      "mean_token_accuracy": 0.2094627231359482,
      "num_tokens": 109388839.0,
      "step": 53980
    },
    {
      "entropy": 5.744126987457276,
      "epoch": 4.717731364152757,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029818004909224426,
      "loss": 5.0934,
      "mean_token_accuracy": 0.2197301894426346,
      "num_tokens": 109399661.0,
      "step": 53985
    },
    {
      "entropy": 5.641302537918091,
      "epoch": 4.718168312505462,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029814905357924374,
      "loss": 5.0511,
      "mean_token_accuracy": 0.22493021786212922,
      "num_tokens": 109409755.0,
      "step": 53990
    },
    {
      "entropy": 5.690918254852295,
      "epoch": 4.718605260858166,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002981180576222321,
      "loss": 5.1613,
      "mean_token_accuracy": 0.2132044956088066,
      "num_tokens": 109421119.0,
      "step": 53995
    },
    {
      "entropy": 5.728606843948365,
      "epoch": 4.719042209210871,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029808706122180404,
      "loss": 5.1641,
      "mean_token_accuracy": 0.21734662652015685,
      "num_tokens": 109431136.0,
      "step": 54000
    },
    {
      "epoch": 4.719042209210871,
      "eval_entropy": 5.4584173502302695,
      "eval_loss": 5.274385452270508,
      "eval_mean_token_accuracy": 0.21693351117180315,
      "eval_num_tokens": 109431136.0,
      "eval_runtime": 28.2534,
      "eval_samples_per_second": 1310.072,
      "eval_steps_per_second": 163.768,
      "step": 54000
    },
    {
      "entropy": 5.726720333099365,
      "epoch": 4.719479157563576,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002980560643785541,
      "loss": 5.1781,
      "mean_token_accuracy": 0.21036430597305297,
      "num_tokens": 109441898.0,
      "step": 54005
    },
    {
      "entropy": 5.784595489501953,
      "epoch": 4.719916105916281,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002980250670930767,
      "loss": 5.2313,
      "mean_token_accuracy": 0.21318596601486206,
      "num_tokens": 109451067.0,
      "step": 54010
    },
    {
      "entropy": 5.691265869140625,
      "epoch": 4.720353054268985,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002979940693659666,
      "loss": 5.0954,
      "mean_token_accuracy": 0.22460405826568602,
      "num_tokens": 109461131.0,
      "step": 54015
    },
    {
      "entropy": 5.702071046829223,
      "epoch": 4.7207900026216905,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029796307119781803,
      "loss": 5.2123,
      "mean_token_accuracy": 0.2165055215358734,
      "num_tokens": 109471495.0,
      "step": 54020
    },
    {
      "entropy": 5.672268629074097,
      "epoch": 4.721226950974395,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029793207258922575,
      "loss": 5.1555,
      "mean_token_accuracy": 0.2156777411699295,
      "num_tokens": 109481571.0,
      "step": 54025
    },
    {
      "entropy": 5.731678581237793,
      "epoch": 4.721663899327099,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029790107354078435,
      "loss": 5.1166,
      "mean_token_accuracy": 0.21239448785781861,
      "num_tokens": 109491233.0,
      "step": 54030
    },
    {
      "entropy": 5.711422967910766,
      "epoch": 4.722100847679804,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029787007405308825,
      "loss": 5.1748,
      "mean_token_accuracy": 0.2155521482229233,
      "num_tokens": 109501722.0,
      "step": 54035
    },
    {
      "entropy": 5.722124147415161,
      "epoch": 4.722537796032509,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029783907412673236,
      "loss": 5.162,
      "mean_token_accuracy": 0.21621166169643402,
      "num_tokens": 109510423.0,
      "step": 54040
    },
    {
      "entropy": 5.693440628051758,
      "epoch": 4.722974744385214,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002978080737623108,
      "loss": 5.2101,
      "mean_token_accuracy": 0.20988955348730087,
      "num_tokens": 109520746.0,
      "step": 54045
    },
    {
      "entropy": 5.733924436569214,
      "epoch": 4.723411692737918,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002977770729604186,
      "loss": 5.2126,
      "mean_token_accuracy": 0.21155020147562026,
      "num_tokens": 109530677.0,
      "step": 54050
    },
    {
      "entropy": 5.7246256351470945,
      "epoch": 4.723848641090623,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002977460717216502,
      "loss": 5.1757,
      "mean_token_accuracy": 0.21681201756000518,
      "num_tokens": 109539923.0,
      "step": 54055
    },
    {
      "entropy": 5.672190952301025,
      "epoch": 4.724285589443328,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002977150700466002,
      "loss": 5.1343,
      "mean_token_accuracy": 0.22405881732702254,
      "num_tokens": 109548925.0,
      "step": 54060
    },
    {
      "entropy": 5.671202802658081,
      "epoch": 4.7247225377960325,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029768406793586325,
      "loss": 5.1256,
      "mean_token_accuracy": 0.21457713544368745,
      "num_tokens": 109558842.0,
      "step": 54065
    },
    {
      "entropy": 5.738715696334839,
      "epoch": 4.725159486148737,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029765306539003393,
      "loss": 5.0949,
      "mean_token_accuracy": 0.22195217311382293,
      "num_tokens": 109567757.0,
      "step": 54070
    },
    {
      "entropy": 5.775686836242675,
      "epoch": 4.725596434501442,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000297622062409707,
      "loss": 5.2167,
      "mean_token_accuracy": 0.2096917673945427,
      "num_tokens": 109577635.0,
      "step": 54075
    },
    {
      "entropy": 5.722248601913452,
      "epoch": 4.726033382854147,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002975910589954771,
      "loss": 5.1704,
      "mean_token_accuracy": 0.20974294394254683,
      "num_tokens": 109586800.0,
      "step": 54080
    },
    {
      "entropy": 5.712167596817016,
      "epoch": 4.726470331206851,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029756005514793883,
      "loss": 5.1271,
      "mean_token_accuracy": 0.2163490906357765,
      "num_tokens": 109596681.0,
      "step": 54085
    },
    {
      "entropy": 5.740663719177246,
      "epoch": 4.726907279559556,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002975290508676869,
      "loss": 5.2354,
      "mean_token_accuracy": 0.21355683356523514,
      "num_tokens": 109606829.0,
      "step": 54090
    },
    {
      "entropy": 5.760685873031616,
      "epoch": 4.727344227912261,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029749804615531594,
      "loss": 5.226,
      "mean_token_accuracy": 0.21144738495349885,
      "num_tokens": 109616521.0,
      "step": 54095
    },
    {
      "entropy": 5.7781013488769535,
      "epoch": 4.7277811762649655,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029746704101142065,
      "loss": 5.2012,
      "mean_token_accuracy": 0.20763325542211533,
      "num_tokens": 109627184.0,
      "step": 54100
    },
    {
      "entropy": 5.729838848114014,
      "epoch": 4.72821812461767,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002974360354365957,
      "loss": 5.0858,
      "mean_token_accuracy": 0.2255094975233078,
      "num_tokens": 109637680.0,
      "step": 54105
    },
    {
      "entropy": 5.738041687011719,
      "epoch": 4.7286550729703745,
      "grad_norm": 1.0,
      "learning_rate": 0.00029740502943143596,
      "loss": 5.1826,
      "mean_token_accuracy": 0.21053774654865265,
      "num_tokens": 109647521.0,
      "step": 54110
    },
    {
      "entropy": 5.691545820236206,
      "epoch": 4.72909202132308,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029737402299653587,
      "loss": 5.1508,
      "mean_token_accuracy": 0.21933820992708206,
      "num_tokens": 109656889.0,
      "step": 54115
    },
    {
      "entropy": 5.667005825042724,
      "epoch": 4.729528969675784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029734301613249035,
      "loss": 5.0597,
      "mean_token_accuracy": 0.22366696000099182,
      "num_tokens": 109666681.0,
      "step": 54120
    },
    {
      "entropy": 5.7600785255432125,
      "epoch": 4.729965918028489,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029731200883989414,
      "loss": 5.3071,
      "mean_token_accuracy": 0.2065662696957588,
      "num_tokens": 109677657.0,
      "step": 54125
    },
    {
      "entropy": 5.7408239364624025,
      "epoch": 4.730402866381194,
      "grad_norm": 1.0,
      "learning_rate": 0.00029728100111934186,
      "loss": 5.158,
      "mean_token_accuracy": 0.212483187019825,
      "num_tokens": 109688031.0,
      "step": 54130
    },
    {
      "entropy": 5.730578279495239,
      "epoch": 4.730839814733899,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002972499929714283,
      "loss": 5.2469,
      "mean_token_accuracy": 0.21175612807273864,
      "num_tokens": 109699652.0,
      "step": 54135
    },
    {
      "entropy": 5.689077711105346,
      "epoch": 4.731276763086603,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002972189843967482,
      "loss": 5.081,
      "mean_token_accuracy": 0.21995254456996918,
      "num_tokens": 109709674.0,
      "step": 54140
    },
    {
      "entropy": 5.648576402664185,
      "epoch": 4.7317137114393075,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002971879753958963,
      "loss": 5.1028,
      "mean_token_accuracy": 0.22019697725772858,
      "num_tokens": 109720374.0,
      "step": 54145
    },
    {
      "entropy": 5.7176995277404785,
      "epoch": 4.732150659792013,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002971569659694675,
      "loss": 5.1897,
      "mean_token_accuracy": 0.20844198018312454,
      "num_tokens": 109730613.0,
      "step": 54150
    },
    {
      "entropy": 5.732439708709717,
      "epoch": 4.732587608144717,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002971259561180565,
      "loss": 5.1764,
      "mean_token_accuracy": 0.2154079869389534,
      "num_tokens": 109741593.0,
      "step": 54155
    },
    {
      "entropy": 5.802332592010498,
      "epoch": 4.733024556497422,
      "grad_norm": 1.140625,
      "learning_rate": 0.000297094945842258,
      "loss": 5.3344,
      "mean_token_accuracy": 0.19870188534259797,
      "num_tokens": 109752178.0,
      "step": 54160
    },
    {
      "entropy": 5.829834747314453,
      "epoch": 4.733461504850126,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002970639351426669,
      "loss": 5.2569,
      "mean_token_accuracy": 0.2063933402299881,
      "num_tokens": 109762833.0,
      "step": 54165
    },
    {
      "entropy": 5.766874504089356,
      "epoch": 4.733898453202832,
      "grad_norm": 1.0,
      "learning_rate": 0.00029703292401987784,
      "loss": 5.2067,
      "mean_token_accuracy": 0.21456831246614455,
      "num_tokens": 109773449.0,
      "step": 54170
    },
    {
      "entropy": 5.733742189407349,
      "epoch": 4.734335401555536,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029700191247448586,
      "loss": 5.2027,
      "mean_token_accuracy": 0.20932964086532593,
      "num_tokens": 109783637.0,
      "step": 54175
    },
    {
      "entropy": 5.764841508865357,
      "epoch": 4.734772349908241,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029697090050708575,
      "loss": 5.1467,
      "mean_token_accuracy": 0.21877555996179582,
      "num_tokens": 109793443.0,
      "step": 54180
    },
    {
      "entropy": 5.751561546325684,
      "epoch": 4.735209298260946,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002969398881182721,
      "loss": 5.2178,
      "mean_token_accuracy": 0.2064864695072174,
      "num_tokens": 109804253.0,
      "step": 54185
    },
    {
      "entropy": 5.747650194168091,
      "epoch": 4.73564624661365,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029690887530864,
      "loss": 5.1542,
      "mean_token_accuracy": 0.21392280608415604,
      "num_tokens": 109814827.0,
      "step": 54190
    },
    {
      "entropy": 5.742172336578369,
      "epoch": 4.736083194966355,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029687786207878414,
      "loss": 5.1691,
      "mean_token_accuracy": 0.21218350380659104,
      "num_tokens": 109824388.0,
      "step": 54195
    },
    {
      "entropy": 5.759196138381958,
      "epoch": 4.736520143319059,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002968468484292995,
      "loss": 5.294,
      "mean_token_accuracy": 0.2080315187573433,
      "num_tokens": 109833967.0,
      "step": 54200
    },
    {
      "entropy": 5.6453722476959225,
      "epoch": 4.736957091671765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029681583436078074,
      "loss": 5.0702,
      "mean_token_accuracy": 0.2361208751797676,
      "num_tokens": 109842522.0,
      "step": 54205
    },
    {
      "entropy": 5.753962469100952,
      "epoch": 4.737394040024469,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029678481987382294,
      "loss": 5.1313,
      "mean_token_accuracy": 0.21880011558532714,
      "num_tokens": 109853015.0,
      "step": 54210
    },
    {
      "entropy": 5.834231805801392,
      "epoch": 4.737830988377174,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002967538049690208,
      "loss": 5.1975,
      "mean_token_accuracy": 0.2090085491538048,
      "num_tokens": 109862801.0,
      "step": 54215
    },
    {
      "entropy": 5.742019844055176,
      "epoch": 4.738267936729878,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029672278964696936,
      "loss": 5.1557,
      "mean_token_accuracy": 0.21484316438436507,
      "num_tokens": 109871888.0,
      "step": 54220
    },
    {
      "entropy": 5.6391377449035645,
      "epoch": 4.7387048850825835,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002966917739082634,
      "loss": 5.178,
      "mean_token_accuracy": 0.21258931010961532,
      "num_tokens": 109881933.0,
      "step": 54225
    },
    {
      "entropy": 5.717382431030273,
      "epoch": 4.739141833435288,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029666075775349783,
      "loss": 5.1526,
      "mean_token_accuracy": 0.2170711115002632,
      "num_tokens": 109892210.0,
      "step": 54230
    },
    {
      "entropy": 5.779815053939819,
      "epoch": 4.739578781787992,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029662974118326756,
      "loss": 5.1967,
      "mean_token_accuracy": 0.21634069532155992,
      "num_tokens": 109902507.0,
      "step": 54235
    },
    {
      "entropy": 5.75586256980896,
      "epoch": 4.740015730140698,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002965987241981676,
      "loss": 5.2112,
      "mean_token_accuracy": 0.21564542949199678,
      "num_tokens": 109912052.0,
      "step": 54240
    },
    {
      "entropy": 5.721407747268676,
      "epoch": 4.740452678493402,
      "grad_norm": 1.25,
      "learning_rate": 0.0002965677067987927,
      "loss": 5.1307,
      "mean_token_accuracy": 0.21666416227817537,
      "num_tokens": 109922591.0,
      "step": 54245
    },
    {
      "entropy": 5.767399024963379,
      "epoch": 4.740889626846107,
      "grad_norm": 1.140625,
      "learning_rate": 0.000296536688985738,
      "loss": 5.1981,
      "mean_token_accuracy": 0.20571983158588408,
      "num_tokens": 109932015.0,
      "step": 54250
    },
    {
      "entropy": 5.741322803497314,
      "epoch": 4.741326575198811,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002965056707595982,
      "loss": 5.2091,
      "mean_token_accuracy": 0.21373498439788818,
      "num_tokens": 109943184.0,
      "step": 54255
    },
    {
      "entropy": 5.710617923736573,
      "epoch": 4.7417635235515165,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002964746521209684,
      "loss": 5.1721,
      "mean_token_accuracy": 0.22192997634410858,
      "num_tokens": 109952013.0,
      "step": 54260
    },
    {
      "entropy": 5.698227119445801,
      "epoch": 4.742200471904221,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029644363307044353,
      "loss": 5.1098,
      "mean_token_accuracy": 0.2173192784190178,
      "num_tokens": 109962163.0,
      "step": 54265
    },
    {
      "entropy": 5.675943851470947,
      "epoch": 4.7426374202569255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002964126136086185,
      "loss": 5.2354,
      "mean_token_accuracy": 0.21781871616840362,
      "num_tokens": 109971479.0,
      "step": 54270
    },
    {
      "entropy": 5.718510150909424,
      "epoch": 4.74307436860963,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002963815937360884,
      "loss": 5.1219,
      "mean_token_accuracy": 0.21841199696063995,
      "num_tokens": 109981407.0,
      "step": 54275
    },
    {
      "entropy": 5.765867137908936,
      "epoch": 4.743511316962335,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002963505734534481,
      "loss": 5.2492,
      "mean_token_accuracy": 0.20954883694648743,
      "num_tokens": 109991404.0,
      "step": 54280
    },
    {
      "entropy": 5.73490343093872,
      "epoch": 4.74394826531504,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029631955276129265,
      "loss": 5.1518,
      "mean_token_accuracy": 0.21781228333711625,
      "num_tokens": 110001483.0,
      "step": 54285
    },
    {
      "entropy": 5.807419157028198,
      "epoch": 4.744385213667744,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029628853166021696,
      "loss": 5.2175,
      "mean_token_accuracy": 0.20848336070775986,
      "num_tokens": 110011460.0,
      "step": 54290
    },
    {
      "entropy": 5.675145626068115,
      "epoch": 4.74482216202045,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002962575101508161,
      "loss": 5.1131,
      "mean_token_accuracy": 0.21518518626689911,
      "num_tokens": 110021199.0,
      "step": 54295
    },
    {
      "entropy": 5.746891927719116,
      "epoch": 4.745259110373154,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000296226488233685,
      "loss": 5.2411,
      "mean_token_accuracy": 0.20529608875513078,
      "num_tokens": 110031127.0,
      "step": 54300
    },
    {
      "entropy": 5.730842781066895,
      "epoch": 4.7456960587258585,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002961954659094187,
      "loss": 5.182,
      "mean_token_accuracy": 0.21693075001239776,
      "num_tokens": 110041541.0,
      "step": 54305
    },
    {
      "entropy": 5.768270826339721,
      "epoch": 4.746133007078563,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002961644431786124,
      "loss": 5.1985,
      "mean_token_accuracy": 0.21373618841171266,
      "num_tokens": 110050882.0,
      "step": 54310
    },
    {
      "entropy": 5.773413610458374,
      "epoch": 4.746569955431268,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002961334200418609,
      "loss": 5.2673,
      "mean_token_accuracy": 0.2078141301870346,
      "num_tokens": 110062510.0,
      "step": 54315
    },
    {
      "entropy": 5.771555328369141,
      "epoch": 4.747006903783973,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002961023964997594,
      "loss": 5.1919,
      "mean_token_accuracy": 0.2127928152680397,
      "num_tokens": 110072625.0,
      "step": 54320
    },
    {
      "entropy": 5.730978870391846,
      "epoch": 4.747443852136677,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029607137255290277,
      "loss": 5.1519,
      "mean_token_accuracy": 0.21643544733524323,
      "num_tokens": 110082525.0,
      "step": 54325
    },
    {
      "entropy": 5.7053985595703125,
      "epoch": 4.747880800489382,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029604034820188623,
      "loss": 5.2083,
      "mean_token_accuracy": 0.21409086585044862,
      "num_tokens": 110092137.0,
      "step": 54330
    },
    {
      "entropy": 5.74933271408081,
      "epoch": 4.748317748842087,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0002960093234473048,
      "loss": 5.235,
      "mean_token_accuracy": 0.21112782657146453,
      "num_tokens": 110103677.0,
      "step": 54335
    },
    {
      "entropy": 5.7417090892791744,
      "epoch": 4.748754697194792,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029597829828975354,
      "loss": 5.1501,
      "mean_token_accuracy": 0.21896735727787017,
      "num_tokens": 110113131.0,
      "step": 54340
    },
    {
      "entropy": 5.7547242641448975,
      "epoch": 4.749191645547496,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002959472727298275,
      "loss": 5.2294,
      "mean_token_accuracy": 0.21338543593883513,
      "num_tokens": 110124777.0,
      "step": 54345
    },
    {
      "entropy": 5.7676187515258786,
      "epoch": 4.749628593900201,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002959162467681218,
      "loss": 5.1721,
      "mean_token_accuracy": 0.213430255651474,
      "num_tokens": 110134234.0,
      "step": 54350
    },
    {
      "entropy": 5.72623085975647,
      "epoch": 4.750065542252906,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002958852204052315,
      "loss": 5.2103,
      "mean_token_accuracy": 0.20687133073806763,
      "num_tokens": 110144135.0,
      "step": 54355
    },
    {
      "entropy": 5.6572585105896,
      "epoch": 4.75050249060561,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029585419364175174,
      "loss": 5.1116,
      "mean_token_accuracy": 0.21158581972122192,
      "num_tokens": 110154009.0,
      "step": 54360
    },
    {
      "entropy": 5.788527345657348,
      "epoch": 4.750939438958315,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029582316647827774,
      "loss": 5.2519,
      "mean_token_accuracy": 0.20739710032939912,
      "num_tokens": 110163628.0,
      "step": 54365
    },
    {
      "entropy": 5.746222639083863,
      "epoch": 4.75137638731102,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029579213891540437,
      "loss": 5.1716,
      "mean_token_accuracy": 0.21665374785661698,
      "num_tokens": 110174321.0,
      "step": 54370
    },
    {
      "entropy": 5.681610727310181,
      "epoch": 4.751813335663725,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029576111095372685,
      "loss": 5.107,
      "mean_token_accuracy": 0.2257288798689842,
      "num_tokens": 110184674.0,
      "step": 54375
    },
    {
      "entropy": 5.631223821640015,
      "epoch": 4.752250284016429,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029573008259384037,
      "loss": 5.1063,
      "mean_token_accuracy": 0.22483773976564408,
      "num_tokens": 110195090.0,
      "step": 54380
    },
    {
      "entropy": 5.757683753967285,
      "epoch": 4.752687232369134,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002956990538363401,
      "loss": 5.1693,
      "mean_token_accuracy": 0.21394706666469573,
      "num_tokens": 110206237.0,
      "step": 54385
    },
    {
      "entropy": 5.761229228973389,
      "epoch": 4.753124180721839,
      "grad_norm": 1.125,
      "learning_rate": 0.0002956680246818211,
      "loss": 5.2233,
      "mean_token_accuracy": 0.21703388243913652,
      "num_tokens": 110216434.0,
      "step": 54390
    },
    {
      "entropy": 5.731501722335816,
      "epoch": 4.753561129074543,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002956369951308786,
      "loss": 5.2295,
      "mean_token_accuracy": 0.20859227627515792,
      "num_tokens": 110226847.0,
      "step": 54395
    },
    {
      "entropy": 5.631533241271972,
      "epoch": 4.753998077427248,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029560596518410766,
      "loss": 5.0429,
      "mean_token_accuracy": 0.23109820038080214,
      "num_tokens": 110238113.0,
      "step": 54400
    },
    {
      "entropy": 5.759077501296997,
      "epoch": 4.754435025779953,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002955749348421035,
      "loss": 5.188,
      "mean_token_accuracy": 0.21620623171329498,
      "num_tokens": 110248338.0,
      "step": 54405
    },
    {
      "entropy": 5.6586137294769285,
      "epoch": 4.754871974132658,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029554390410546134,
      "loss": 5.1353,
      "mean_token_accuracy": 0.2142074942588806,
      "num_tokens": 110259193.0,
      "step": 54410
    },
    {
      "entropy": 5.773079776763916,
      "epoch": 4.755308922485362,
      "grad_norm": 1.125,
      "learning_rate": 0.0002955128729747764,
      "loss": 5.1912,
      "mean_token_accuracy": 0.20614926666021346,
      "num_tokens": 110269929.0,
      "step": 54415
    },
    {
      "entropy": 5.642238569259644,
      "epoch": 4.755745870838067,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002954818414506437,
      "loss": 5.0571,
      "mean_token_accuracy": 0.22226547449827194,
      "num_tokens": 110279591.0,
      "step": 54420
    },
    {
      "entropy": 5.599048233032226,
      "epoch": 4.756182819190772,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002954508095336586,
      "loss": 5.0567,
      "mean_token_accuracy": 0.22818756699562073,
      "num_tokens": 110289781.0,
      "step": 54425
    },
    {
      "entropy": 5.726693773269654,
      "epoch": 4.7566197675434765,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002954197772244162,
      "loss": 5.1942,
      "mean_token_accuracy": 0.21426362097263335,
      "num_tokens": 110299581.0,
      "step": 54430
    },
    {
      "entropy": 5.748695659637451,
      "epoch": 4.757056715896181,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029538874452351185,
      "loss": 5.2223,
      "mean_token_accuracy": 0.20819367468357086,
      "num_tokens": 110309343.0,
      "step": 54435
    },
    {
      "entropy": 5.748353672027588,
      "epoch": 4.757493664248885,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029535771143154074,
      "loss": 5.2216,
      "mean_token_accuracy": 0.21350135058164596,
      "num_tokens": 110320346.0,
      "step": 54440
    },
    {
      "entropy": 5.706883478164673,
      "epoch": 4.757930612601591,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029532667794909795,
      "loss": 5.1925,
      "mean_token_accuracy": 0.21372003704309464,
      "num_tokens": 110330775.0,
      "step": 54445
    },
    {
      "entropy": 5.7500323295593265,
      "epoch": 4.758367560954295,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029529564407677883,
      "loss": 5.195,
      "mean_token_accuracy": 0.21675880998373032,
      "num_tokens": 110340372.0,
      "step": 54450
    },
    {
      "entropy": 5.709173727035522,
      "epoch": 4.758804509307,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002952646098151786,
      "loss": 5.1504,
      "mean_token_accuracy": 0.21643224358558655,
      "num_tokens": 110349988.0,
      "step": 54455
    },
    {
      "entropy": 5.752673768997193,
      "epoch": 4.759241457659705,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029523357516489265,
      "loss": 5.2122,
      "mean_token_accuracy": 0.2125256285071373,
      "num_tokens": 110360072.0,
      "step": 54460
    },
    {
      "entropy": 5.735281705856323,
      "epoch": 4.7596784060124095,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002952025401265161,
      "loss": 5.1607,
      "mean_token_accuracy": 0.22166652977466583,
      "num_tokens": 110370289.0,
      "step": 54465
    },
    {
      "entropy": 5.782227897644043,
      "epoch": 4.760115354365114,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002951715047006441,
      "loss": 5.2177,
      "mean_token_accuracy": 0.21341247409582137,
      "num_tokens": 110380777.0,
      "step": 54470
    },
    {
      "entropy": 5.773083829879761,
      "epoch": 4.7605523027178185,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002951404688878722,
      "loss": 5.1868,
      "mean_token_accuracy": 0.21501149088144303,
      "num_tokens": 110390800.0,
      "step": 54475
    },
    {
      "entropy": 5.70566029548645,
      "epoch": 4.760989251070524,
      "grad_norm": 1.125,
      "learning_rate": 0.0002951094326887955,
      "loss": 5.2703,
      "mean_token_accuracy": 0.20075815469026564,
      "num_tokens": 110400582.0,
      "step": 54480
    },
    {
      "entropy": 5.761270380020141,
      "epoch": 4.761426199423228,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029507839610400945,
      "loss": 5.1963,
      "mean_token_accuracy": 0.2079466164112091,
      "num_tokens": 110410341.0,
      "step": 54485
    },
    {
      "entropy": 5.7486724853515625,
      "epoch": 4.761863147775933,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002950473591341091,
      "loss": 5.1862,
      "mean_token_accuracy": 0.2120512157678604,
      "num_tokens": 110420804.0,
      "step": 54490
    },
    {
      "entropy": 5.7009893417358395,
      "epoch": 4.762300096128637,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029501632177969,
      "loss": 5.0783,
      "mean_token_accuracy": 0.22182923108339309,
      "num_tokens": 110431485.0,
      "step": 54495
    },
    {
      "entropy": 5.706374740600586,
      "epoch": 4.762737044481343,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002949852840413473,
      "loss": 5.1566,
      "mean_token_accuracy": 0.2169412687420845,
      "num_tokens": 110442368.0,
      "step": 54500
    },
    {
      "entropy": 5.619919347763061,
      "epoch": 4.763173992834047,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029495424591967644,
      "loss": 5.0834,
      "mean_token_accuracy": 0.22909644842147828,
      "num_tokens": 110453087.0,
      "step": 54505
    },
    {
      "entropy": 5.761839246749878,
      "epoch": 4.7636109411867515,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002949232074152726,
      "loss": 5.2284,
      "mean_token_accuracy": 0.2081569418311119,
      "num_tokens": 110463086.0,
      "step": 54510
    },
    {
      "entropy": 5.735861158370971,
      "epoch": 4.764047889539457,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002948921685287313,
      "loss": 5.2051,
      "mean_token_accuracy": 0.21154484301805496,
      "num_tokens": 110472918.0,
      "step": 54515
    },
    {
      "entropy": 5.7844490051269535,
      "epoch": 4.764484837892161,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029486112926064775,
      "loss": 5.2026,
      "mean_token_accuracy": 0.20568778365850449,
      "num_tokens": 110482966.0,
      "step": 54520
    },
    {
      "entropy": 5.737463521957397,
      "epoch": 4.764921786244866,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002948300896116173,
      "loss": 5.1963,
      "mean_token_accuracy": 0.2184051051735878,
      "num_tokens": 110493343.0,
      "step": 54525
    },
    {
      "entropy": 5.727660417556763,
      "epoch": 4.76535873459757,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002947990495822354,
      "loss": 5.0986,
      "mean_token_accuracy": 0.2234789550304413,
      "num_tokens": 110503978.0,
      "step": 54530
    },
    {
      "entropy": 5.733267688751221,
      "epoch": 4.765795682950275,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029476800917309733,
      "loss": 5.1541,
      "mean_token_accuracy": 0.21811143010854722,
      "num_tokens": 110513689.0,
      "step": 54535
    },
    {
      "entropy": 5.714434623718262,
      "epoch": 4.76623263130298,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029473696838479843,
      "loss": 5.1767,
      "mean_token_accuracy": 0.21874364465475082,
      "num_tokens": 110523002.0,
      "step": 54540
    },
    {
      "entropy": 5.733756875991821,
      "epoch": 4.766669579655685,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029470592721793414,
      "loss": 5.2658,
      "mean_token_accuracy": 0.20826193690299988,
      "num_tokens": 110533834.0,
      "step": 54545
    },
    {
      "entropy": 5.704041290283203,
      "epoch": 4.767106528008389,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029467488567309983,
      "loss": 5.0681,
      "mean_token_accuracy": 0.22915772795677186,
      "num_tokens": 110544331.0,
      "step": 54550
    },
    {
      "entropy": 5.783283376693726,
      "epoch": 4.767543476361094,
      "grad_norm": 1.046875,
      "learning_rate": 0.000294643843750891,
      "loss": 5.2391,
      "mean_token_accuracy": 0.20491552501916885,
      "num_tokens": 110554356.0,
      "step": 54555
    },
    {
      "entropy": 5.759527206420898,
      "epoch": 4.767980424713799,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002946128014519029,
      "loss": 5.2924,
      "mean_token_accuracy": 0.20502231270074844,
      "num_tokens": 110564794.0,
      "step": 54560
    },
    {
      "entropy": 5.7778887271881105,
      "epoch": 4.768417373066503,
      "grad_norm": 1.140625,
      "learning_rate": 0.000294581758776731,
      "loss": 5.2229,
      "mean_token_accuracy": 0.20733482241630555,
      "num_tokens": 110574523.0,
      "step": 54565
    },
    {
      "entropy": 5.709337043762207,
      "epoch": 4.768854321419209,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002945507157259707,
      "loss": 5.1535,
      "mean_token_accuracy": 0.22032321542501448,
      "num_tokens": 110585051.0,
      "step": 54570
    },
    {
      "entropy": 5.727132272720337,
      "epoch": 4.769291269771913,
      "grad_norm": 1.0,
      "learning_rate": 0.0002945196723002174,
      "loss": 5.1431,
      "mean_token_accuracy": 0.21971799731254577,
      "num_tokens": 110595987.0,
      "step": 54575
    },
    {
      "entropy": 5.796589517593384,
      "epoch": 4.769728218124618,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029448862850006664,
      "loss": 5.2718,
      "mean_token_accuracy": 0.2053923025727272,
      "num_tokens": 110605986.0,
      "step": 54580
    },
    {
      "entropy": 5.792552804946899,
      "epoch": 4.770165166477322,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029445758432611373,
      "loss": 5.2386,
      "mean_token_accuracy": 0.20992051512002946,
      "num_tokens": 110616243.0,
      "step": 54585
    },
    {
      "entropy": 5.7852729797363285,
      "epoch": 4.770602114830027,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002944265397789541,
      "loss": 5.2076,
      "mean_token_accuracy": 0.2102820858359337,
      "num_tokens": 110625752.0,
      "step": 54590
    },
    {
      "entropy": 5.764336395263672,
      "epoch": 4.771039063182732,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029439549485918334,
      "loss": 5.2391,
      "mean_token_accuracy": 0.21789452582597732,
      "num_tokens": 110635607.0,
      "step": 54595
    },
    {
      "entropy": 5.8019603252410885,
      "epoch": 4.771476011535436,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029436444956739674,
      "loss": 5.2209,
      "mean_token_accuracy": 0.21321318298578262,
      "num_tokens": 110646376.0,
      "step": 54600
    },
    {
      "entropy": 5.751847505569458,
      "epoch": 4.771912959888141,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029433340390418996,
      "loss": 5.2274,
      "mean_token_accuracy": 0.20436671376228333,
      "num_tokens": 110656697.0,
      "step": 54605
    },
    {
      "entropy": 5.7375623226165775,
      "epoch": 4.772349908240846,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002943023578701583,
      "loss": 5.2438,
      "mean_token_accuracy": 0.2174438640475273,
      "num_tokens": 110665897.0,
      "step": 54610
    },
    {
      "entropy": 5.688907098770142,
      "epoch": 4.772786856593551,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002942713114658973,
      "loss": 5.1396,
      "mean_token_accuracy": 0.22499750554561615,
      "num_tokens": 110676282.0,
      "step": 54615
    },
    {
      "entropy": 5.76478796005249,
      "epoch": 4.773223804946255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029424026469200245,
      "loss": 5.2618,
      "mean_token_accuracy": 0.21144620776176454,
      "num_tokens": 110686097.0,
      "step": 54620
    },
    {
      "entropy": 5.716355848312378,
      "epoch": 4.773660753298961,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029420921754906923,
      "loss": 5.1753,
      "mean_token_accuracy": 0.21529625058174134,
      "num_tokens": 110696378.0,
      "step": 54625
    },
    {
      "entropy": 5.779963636398316,
      "epoch": 4.774097701651665,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002941781700376932,
      "loss": 5.24,
      "mean_token_accuracy": 0.2102825716137886,
      "num_tokens": 110707144.0,
      "step": 54630
    },
    {
      "entropy": 5.7097447395324705,
      "epoch": 4.7745346500043695,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029414712215846973,
      "loss": 5.1333,
      "mean_token_accuracy": 0.2194504424929619,
      "num_tokens": 110715993.0,
      "step": 54635
    },
    {
      "entropy": 5.782987356185913,
      "epoch": 4.774971598357074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029411607391199445,
      "loss": 5.2278,
      "mean_token_accuracy": 0.20377079844474794,
      "num_tokens": 110726627.0,
      "step": 54640
    },
    {
      "entropy": 5.798917245864868,
      "epoch": 4.775408546709778,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029408502529886286,
      "loss": 5.2778,
      "mean_token_accuracy": 0.21158662736415862,
      "num_tokens": 110736958.0,
      "step": 54645
    },
    {
      "entropy": 5.6950360298156735,
      "epoch": 4.775845495062484,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002940539763196706,
      "loss": 5.1238,
      "mean_token_accuracy": 0.217495958507061,
      "num_tokens": 110747057.0,
      "step": 54650
    },
    {
      "entropy": 5.7088175296783445,
      "epoch": 4.776282443415188,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029402292697501293,
      "loss": 5.2421,
      "mean_token_accuracy": 0.21085168421268463,
      "num_tokens": 110756430.0,
      "step": 54655
    },
    {
      "entropy": 5.726967668533325,
      "epoch": 4.776719391767893,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029399187726548564,
      "loss": 5.146,
      "mean_token_accuracy": 0.21567347943782805,
      "num_tokens": 110766566.0,
      "step": 54660
    },
    {
      "entropy": 5.672745656967163,
      "epoch": 4.777156340120598,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002939608271916841,
      "loss": 5.142,
      "mean_token_accuracy": 0.21944948136806489,
      "num_tokens": 110775934.0,
      "step": 54665
    },
    {
      "entropy": 5.791196393966675,
      "epoch": 4.777593288473303,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002939297767542041,
      "loss": 5.1979,
      "mean_token_accuracy": 0.2110932797193527,
      "num_tokens": 110787608.0,
      "step": 54670
    },
    {
      "entropy": 5.756317138671875,
      "epoch": 4.778030236826007,
      "grad_norm": 1.078125,
      "learning_rate": 0.000293898725953641,
      "loss": 5.1761,
      "mean_token_accuracy": 0.21492077708244323,
      "num_tokens": 110797265.0,
      "step": 54675
    },
    {
      "entropy": 5.64783878326416,
      "epoch": 4.7784671851787115,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002938676747905904,
      "loss": 5.0961,
      "mean_token_accuracy": 0.21868561953306198,
      "num_tokens": 110807059.0,
      "step": 54680
    },
    {
      "entropy": 5.721025085449218,
      "epoch": 4.778904133531417,
      "grad_norm": 1.109375,
      "learning_rate": 0.000293836623265648,
      "loss": 5.12,
      "mean_token_accuracy": 0.21526452600955964,
      "num_tokens": 110816056.0,
      "step": 54685
    },
    {
      "entropy": 5.808204460144043,
      "epoch": 4.779341081884121,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029380557137940924,
      "loss": 5.2231,
      "mean_token_accuracy": 0.2110515609383583,
      "num_tokens": 110825601.0,
      "step": 54690
    },
    {
      "entropy": 5.739305925369263,
      "epoch": 4.779778030236826,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029377451913246977,
      "loss": 5.1018,
      "mean_token_accuracy": 0.219736148416996,
      "num_tokens": 110835435.0,
      "step": 54695
    },
    {
      "entropy": 5.753665113449097,
      "epoch": 4.78021497858953,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002937434665254252,
      "loss": 5.176,
      "mean_token_accuracy": 0.2171602040529251,
      "num_tokens": 110845607.0,
      "step": 54700
    },
    {
      "entropy": 5.729040908813476,
      "epoch": 4.780651926942236,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002937124135588711,
      "loss": 5.1828,
      "mean_token_accuracy": 0.2169637694954872,
      "num_tokens": 110856089.0,
      "step": 54705
    },
    {
      "entropy": 5.789300918579102,
      "epoch": 4.78108887529494,
      "grad_norm": 1.0,
      "learning_rate": 0.000293681360233403,
      "loss": 5.2672,
      "mean_token_accuracy": 0.20934372693300246,
      "num_tokens": 110865491.0,
      "step": 54710
    },
    {
      "entropy": 5.748550271987915,
      "epoch": 4.781525823647645,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002936503065496167,
      "loss": 5.2336,
      "mean_token_accuracy": 0.2104157865047455,
      "num_tokens": 110875579.0,
      "step": 54715
    },
    {
      "entropy": 5.699601793289185,
      "epoch": 4.78196277200035,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029361925250810786,
      "loss": 5.1832,
      "mean_token_accuracy": 0.2163658246397972,
      "num_tokens": 110886452.0,
      "step": 54720
    },
    {
      "entropy": 5.814985466003418,
      "epoch": 4.782399720353054,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002935881981094719,
      "loss": 5.2661,
      "mean_token_accuracy": 0.2047137051820755,
      "num_tokens": 110897854.0,
      "step": 54725
    },
    {
      "entropy": 5.77573528289795,
      "epoch": 4.782836668705759,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029355714335430456,
      "loss": 5.152,
      "mean_token_accuracy": 0.21426525712013245,
      "num_tokens": 110907942.0,
      "step": 54730
    },
    {
      "entropy": 5.755763244628906,
      "epoch": 4.783273617058463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002935260882432015,
      "loss": 5.2296,
      "mean_token_accuracy": 0.21525953263044356,
      "num_tokens": 110917381.0,
      "step": 54735
    },
    {
      "entropy": 5.7931800365448,
      "epoch": 4.783710565411169,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002934950327767583,
      "loss": 5.2198,
      "mean_token_accuracy": 0.20847998559474945,
      "num_tokens": 110927343.0,
      "step": 54740
    },
    {
      "entropy": 5.775851631164551,
      "epoch": 4.784147513763873,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002934639769555708,
      "loss": 5.1911,
      "mean_token_accuracy": 0.2150707647204399,
      "num_tokens": 110937475.0,
      "step": 54745
    },
    {
      "entropy": 5.714669179916382,
      "epoch": 4.784584462116578,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002934329207802345,
      "loss": 5.1248,
      "mean_token_accuracy": 0.22608604282140732,
      "num_tokens": 110947411.0,
      "step": 54750
    },
    {
      "entropy": 5.698912715911865,
      "epoch": 4.785021410469282,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029340186425134507,
      "loss": 5.1543,
      "mean_token_accuracy": 0.22095881998538972,
      "num_tokens": 110957625.0,
      "step": 54755
    },
    {
      "entropy": 5.749842548370362,
      "epoch": 4.7854583588219874,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002933708073694983,
      "loss": 5.2399,
      "mean_token_accuracy": 0.21154245287179946,
      "num_tokens": 110967848.0,
      "step": 54760
    },
    {
      "entropy": 5.760981512069702,
      "epoch": 4.785895307174692,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002933397501352898,
      "loss": 5.1824,
      "mean_token_accuracy": 0.21746697574853896,
      "num_tokens": 110978109.0,
      "step": 54765
    },
    {
      "entropy": 5.7900902271270756,
      "epoch": 4.786332255527396,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002933086925493153,
      "loss": 5.1527,
      "mean_token_accuracy": 0.21675294488668442,
      "num_tokens": 110987803.0,
      "step": 54770
    },
    {
      "entropy": 5.694940805435181,
      "epoch": 4.786769203880102,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002932776346121704,
      "loss": 5.1465,
      "mean_token_accuracy": 0.21856310814619065,
      "num_tokens": 110997568.0,
      "step": 54775
    },
    {
      "entropy": 5.724542331695557,
      "epoch": 4.787206152232806,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000293246576324451,
      "loss": 5.208,
      "mean_token_accuracy": 0.21526411026716233,
      "num_tokens": 111006922.0,
      "step": 54780
    },
    {
      "entropy": 5.788908243179321,
      "epoch": 4.787643100585511,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029321551768675256,
      "loss": 5.1836,
      "mean_token_accuracy": 0.209272800385952,
      "num_tokens": 111017455.0,
      "step": 54785
    },
    {
      "entropy": 5.760396242141724,
      "epoch": 4.788080048938215,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002931844586996711,
      "loss": 5.2306,
      "mean_token_accuracy": 0.20745979696512223,
      "num_tokens": 111027668.0,
      "step": 54790
    },
    {
      "entropy": 5.736910200119018,
      "epoch": 4.7885169972909205,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029315339936380216,
      "loss": 5.2031,
      "mean_token_accuracy": 0.21653015911579132,
      "num_tokens": 111038199.0,
      "step": 54795
    },
    {
      "entropy": 5.799084186553955,
      "epoch": 4.788953945643625,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002931223396797415,
      "loss": 5.3229,
      "mean_token_accuracy": 0.2031595900654793,
      "num_tokens": 111047721.0,
      "step": 54800
    },
    {
      "entropy": 5.751490688323974,
      "epoch": 4.7893908939963294,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002930912796480848,
      "loss": 5.1637,
      "mean_token_accuracy": 0.22055369466543198,
      "num_tokens": 111057771.0,
      "step": 54805
    },
    {
      "entropy": 5.766078090667724,
      "epoch": 4.789827842349034,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029306021926942796,
      "loss": 5.1901,
      "mean_token_accuracy": 0.21203308552503586,
      "num_tokens": 111066994.0,
      "step": 54810
    },
    {
      "entropy": 5.721041917800903,
      "epoch": 4.790264790701739,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002930291585443667,
      "loss": 5.1368,
      "mean_token_accuracy": 0.21312694102525712,
      "num_tokens": 111076280.0,
      "step": 54815
    },
    {
      "entropy": 5.690176963806152,
      "epoch": 4.790701739054444,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002929980974734966,
      "loss": 5.2155,
      "mean_token_accuracy": 0.21299009323120116,
      "num_tokens": 111087034.0,
      "step": 54820
    },
    {
      "entropy": 5.707128477096558,
      "epoch": 4.791138687407148,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002929670360574136,
      "loss": 5.0777,
      "mean_token_accuracy": 0.2214274898171425,
      "num_tokens": 111097896.0,
      "step": 54825
    },
    {
      "entropy": 5.744340991973877,
      "epoch": 4.791575635759854,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029293597429671343,
      "loss": 5.2042,
      "mean_token_accuracy": 0.20874832421541215,
      "num_tokens": 111108464.0,
      "step": 54830
    },
    {
      "entropy": 5.728462600708008,
      "epoch": 4.792012584112558,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002929049121919919,
      "loss": 5.2207,
      "mean_token_accuracy": 0.21435820311307907,
      "num_tokens": 111118670.0,
      "step": 54835
    },
    {
      "entropy": 5.759822702407837,
      "epoch": 4.7924495324652625,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029287384974384476,
      "loss": 5.1948,
      "mean_token_accuracy": 0.2114403650164604,
      "num_tokens": 111129451.0,
      "step": 54840
    },
    {
      "entropy": 5.774183225631714,
      "epoch": 4.792886480817967,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002928427869528678,
      "loss": 5.2781,
      "mean_token_accuracy": 0.19965946823358535,
      "num_tokens": 111139852.0,
      "step": 54845
    },
    {
      "entropy": 5.821495199203492,
      "epoch": 4.793323429170672,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029281172381965685,
      "loss": 5.2179,
      "mean_token_accuracy": 0.20556462407112122,
      "num_tokens": 111149408.0,
      "step": 54850
    },
    {
      "entropy": 5.743552923202515,
      "epoch": 4.793760377523377,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002927806603448077,
      "loss": 5.1985,
      "mean_token_accuracy": 0.20828363299369812,
      "num_tokens": 111159182.0,
      "step": 54855
    },
    {
      "entropy": 5.69180908203125,
      "epoch": 4.794197325876081,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029274959652891624,
      "loss": 5.1771,
      "mean_token_accuracy": 0.212380750477314,
      "num_tokens": 111168747.0,
      "step": 54860
    },
    {
      "entropy": 5.762465572357177,
      "epoch": 4.794634274228786,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002927185323725782,
      "loss": 5.2672,
      "mean_token_accuracy": 0.20185347497463227,
      "num_tokens": 111178850.0,
      "step": 54865
    },
    {
      "entropy": 5.735583782196045,
      "epoch": 4.795071222581491,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029268746787638935,
      "loss": 5.1709,
      "mean_token_accuracy": 0.2177328884601593,
      "num_tokens": 111188173.0,
      "step": 54870
    },
    {
      "entropy": 5.762389373779297,
      "epoch": 4.795508170934196,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002926564030409456,
      "loss": 5.208,
      "mean_token_accuracy": 0.201837719976902,
      "num_tokens": 111198564.0,
      "step": 54875
    },
    {
      "entropy": 5.707560634613037,
      "epoch": 4.7959451192869,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029262533786684276,
      "loss": 5.1379,
      "mean_token_accuracy": 0.216363126039505,
      "num_tokens": 111208546.0,
      "step": 54880
    },
    {
      "entropy": 5.706278944015503,
      "epoch": 4.796382067639605,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002925942723546768,
      "loss": 5.1112,
      "mean_token_accuracy": 0.22125234752893447,
      "num_tokens": 111218903.0,
      "step": 54885
    },
    {
      "entropy": 5.699239683151245,
      "epoch": 4.79681901599231,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002925632065050434,
      "loss": 5.143,
      "mean_token_accuracy": 0.21721666753292085,
      "num_tokens": 111229420.0,
      "step": 54890
    },
    {
      "entropy": 5.774221992492675,
      "epoch": 4.797255964345014,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00029253214031853855,
      "loss": 5.1579,
      "mean_token_accuracy": 0.21768647581338882,
      "num_tokens": 111238764.0,
      "step": 54895
    },
    {
      "entropy": 5.749040985107422,
      "epoch": 4.797692912697719,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002925010737957581,
      "loss": 5.2803,
      "mean_token_accuracy": 0.203656005859375,
      "num_tokens": 111249082.0,
      "step": 54900
    },
    {
      "entropy": 5.75835428237915,
      "epoch": 4.798129861050424,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002924700069372978,
      "loss": 5.2784,
      "mean_token_accuracy": 0.20617249459028245,
      "num_tokens": 111259141.0,
      "step": 54905
    },
    {
      "entropy": 5.722598028182984,
      "epoch": 4.798566809403129,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002924389397437537,
      "loss": 5.1768,
      "mean_token_accuracy": 0.21778105050325394,
      "num_tokens": 111269356.0,
      "step": 54910
    },
    {
      "entropy": 5.747978687286377,
      "epoch": 4.799003757755833,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029240787221572153,
      "loss": 5.0913,
      "mean_token_accuracy": 0.22195762693881987,
      "num_tokens": 111278550.0,
      "step": 54915
    },
    {
      "entropy": 5.701876926422119,
      "epoch": 4.799440706108538,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002923768043537972,
      "loss": 5.1962,
      "mean_token_accuracy": 0.21322462111711502,
      "num_tokens": 111289231.0,
      "step": 54920
    },
    {
      "entropy": 5.754762268066406,
      "epoch": 4.799877654461243,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029234573615857676,
      "loss": 5.2553,
      "mean_token_accuracy": 0.2021104335784912,
      "num_tokens": 111298401.0,
      "step": 54925
    },
    {
      "entropy": 5.734134912490845,
      "epoch": 4.800314602813947,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029231466763065595,
      "loss": 5.1969,
      "mean_token_accuracy": 0.21694670170545577,
      "num_tokens": 111309752.0,
      "step": 54930
    },
    {
      "entropy": 5.714988088607788,
      "epoch": 4.800751551166652,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002922835987706308,
      "loss": 5.2028,
      "mean_token_accuracy": 0.2123669356107712,
      "num_tokens": 111319852.0,
      "step": 54935
    },
    {
      "entropy": 5.6966606140136715,
      "epoch": 4.801188499519357,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002922525295790971,
      "loss": 5.1325,
      "mean_token_accuracy": 0.22106748819351196,
      "num_tokens": 111329465.0,
      "step": 54940
    },
    {
      "entropy": 5.783967304229736,
      "epoch": 4.801625447872062,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002922214600566508,
      "loss": 5.2177,
      "mean_token_accuracy": 0.207373209297657,
      "num_tokens": 111338949.0,
      "step": 54945
    },
    {
      "entropy": 5.799101161956787,
      "epoch": 4.802062396224766,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029219039020388803,
      "loss": 5.2603,
      "mean_token_accuracy": 0.20902719646692275,
      "num_tokens": 111349451.0,
      "step": 54950
    },
    {
      "entropy": 5.686794662475586,
      "epoch": 4.802499344577471,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002921593200214045,
      "loss": 5.1137,
      "mean_token_accuracy": 0.22035885006189346,
      "num_tokens": 111359349.0,
      "step": 54955
    },
    {
      "entropy": 5.7304291248321535,
      "epoch": 4.802936292930176,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002921282495097962,
      "loss": 5.2531,
      "mean_token_accuracy": 0.21215543895959854,
      "num_tokens": 111369618.0,
      "step": 54960
    },
    {
      "entropy": 5.72594518661499,
      "epoch": 4.8033732412828805,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029209717866965904,
      "loss": 5.1744,
      "mean_token_accuracy": 0.21802212446928024,
      "num_tokens": 111379395.0,
      "step": 54965
    },
    {
      "entropy": 5.781170749664307,
      "epoch": 4.803810189635585,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002920661075015891,
      "loss": 5.2893,
      "mean_token_accuracy": 0.21043953001499177,
      "num_tokens": 111389722.0,
      "step": 54970
    },
    {
      "entropy": 5.75532751083374,
      "epoch": 4.804247137988289,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002920350360061823,
      "loss": 5.1891,
      "mean_token_accuracy": 0.21226043552160262,
      "num_tokens": 111400470.0,
      "step": 54975
    },
    {
      "entropy": 5.740268135070801,
      "epoch": 4.804684086340995,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029200396418403463,
      "loss": 5.18,
      "mean_token_accuracy": 0.21291449069976806,
      "num_tokens": 111410296.0,
      "step": 54980
    },
    {
      "entropy": 5.733100128173828,
      "epoch": 4.805121034693699,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002919728920357419,
      "loss": 5.1425,
      "mean_token_accuracy": 0.22292421758174896,
      "num_tokens": 111421790.0,
      "step": 54985
    },
    {
      "entropy": 5.760402679443359,
      "epoch": 4.805557983046404,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029194181956190027,
      "loss": 5.2697,
      "mean_token_accuracy": 0.20733244121074676,
      "num_tokens": 111431764.0,
      "step": 54990
    },
    {
      "entropy": 5.738229608535766,
      "epoch": 4.805994931399109,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002919107467631056,
      "loss": 5.1909,
      "mean_token_accuracy": 0.2123204290866852,
      "num_tokens": 111441334.0,
      "step": 54995
    },
    {
      "entropy": 5.783652067184448,
      "epoch": 4.8064318797518135,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000291879673639954,
      "loss": 5.2622,
      "mean_token_accuracy": 0.2097567453980446,
      "num_tokens": 111451123.0,
      "step": 55000
    },
    {
      "entropy": 5.75184268951416,
      "epoch": 4.806868828104518,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029184860019304136,
      "loss": 5.1582,
      "mean_token_accuracy": 0.22108163088560104,
      "num_tokens": 111461157.0,
      "step": 55005
    },
    {
      "entropy": 5.731890487670898,
      "epoch": 4.8073057764572225,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002918175264229638,
      "loss": 5.2145,
      "mean_token_accuracy": 0.21133464872837066,
      "num_tokens": 111471164.0,
      "step": 55010
    },
    {
      "entropy": 5.796941757202148,
      "epoch": 4.807742724809928,
      "grad_norm": 0.9375,
      "learning_rate": 0.0002917864523303172,
      "loss": 5.3236,
      "mean_token_accuracy": 0.20262140929698944,
      "num_tokens": 111481870.0,
      "step": 55015
    },
    {
      "entropy": 5.750057935714722,
      "epoch": 4.808179673162632,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029175537791569775,
      "loss": 5.1983,
      "mean_token_accuracy": 0.22452501505613326,
      "num_tokens": 111491321.0,
      "step": 55020
    },
    {
      "entropy": 5.74499192237854,
      "epoch": 4.808616621515337,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029172430317970133,
      "loss": 5.1775,
      "mean_token_accuracy": 0.2113295629620552,
      "num_tokens": 111500518.0,
      "step": 55025
    },
    {
      "entropy": 5.742605686187744,
      "epoch": 4.809053569868041,
      "grad_norm": 1.28125,
      "learning_rate": 0.000291693228122924,
      "loss": 5.1529,
      "mean_token_accuracy": 0.2174870938062668,
      "num_tokens": 111510452.0,
      "step": 55030
    },
    {
      "entropy": 5.671249628067017,
      "epoch": 4.809490518220747,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029166215274596186,
      "loss": 5.0424,
      "mean_token_accuracy": 0.22614084780216218,
      "num_tokens": 111520930.0,
      "step": 55035
    },
    {
      "entropy": 5.760820722579956,
      "epoch": 4.809927466573451,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029163107704941084,
      "loss": 5.2198,
      "mean_token_accuracy": 0.20484415888786317,
      "num_tokens": 111531704.0,
      "step": 55040
    },
    {
      "entropy": 5.747834491729736,
      "epoch": 4.8103644149261555,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029160000103386704,
      "loss": 5.1866,
      "mean_token_accuracy": 0.21097802072763444,
      "num_tokens": 111542015.0,
      "step": 55045
    },
    {
      "entropy": 5.673553371429444,
      "epoch": 4.810801363278861,
      "grad_norm": 1.125,
      "learning_rate": 0.00029156892469992657,
      "loss": 5.1376,
      "mean_token_accuracy": 0.22296708077192307,
      "num_tokens": 111551198.0,
      "step": 55050
    },
    {
      "entropy": 5.711730909347534,
      "epoch": 4.811238311631565,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002915378480481854,
      "loss": 5.149,
      "mean_token_accuracy": 0.21335457563400267,
      "num_tokens": 111561872.0,
      "step": 55055
    },
    {
      "entropy": 5.787834930419922,
      "epoch": 4.81167525998427,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002915067710792396,
      "loss": 5.2706,
      "mean_token_accuracy": 0.2068930059671402,
      "num_tokens": 111572686.0,
      "step": 55060
    },
    {
      "entropy": 5.738528633117676,
      "epoch": 4.812112208336974,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002914756937936854,
      "loss": 5.1896,
      "mean_token_accuracy": 0.21189526915550233,
      "num_tokens": 111582656.0,
      "step": 55065
    },
    {
      "entropy": 5.753657674789428,
      "epoch": 4.81254915668968,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002914446161921187,
      "loss": 5.2318,
      "mean_token_accuracy": 0.20735996067523957,
      "num_tokens": 111594182.0,
      "step": 55070
    },
    {
      "entropy": 5.772929239273071,
      "epoch": 4.812986105042384,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002914135382751356,
      "loss": 5.2211,
      "mean_token_accuracy": 0.2094511166214943,
      "num_tokens": 111605174.0,
      "step": 55075
    },
    {
      "entropy": 5.7994554996490475,
      "epoch": 4.813423053395089,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0002913824600433323,
      "loss": 5.2322,
      "mean_token_accuracy": 0.2126450091600418,
      "num_tokens": 111617489.0,
      "step": 55080
    },
    {
      "entropy": 5.834039640426636,
      "epoch": 4.813860001747793,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029135138149730485,
      "loss": 5.2981,
      "mean_token_accuracy": 0.20438664704561232,
      "num_tokens": 111627250.0,
      "step": 55085
    },
    {
      "entropy": 5.749123811721802,
      "epoch": 4.814296950100498,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029132030263764923,
      "loss": 5.1933,
      "mean_token_accuracy": 0.21170329600572585,
      "num_tokens": 111637726.0,
      "step": 55090
    },
    {
      "entropy": 5.722538614273072,
      "epoch": 4.814733898453203,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002912892234649617,
      "loss": 5.1368,
      "mean_token_accuracy": 0.21365794241428376,
      "num_tokens": 111647630.0,
      "step": 55095
    },
    {
      "entropy": 5.710267877578735,
      "epoch": 4.815170846805907,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002912581439798384,
      "loss": 5.1847,
      "mean_token_accuracy": 0.21251987516880036,
      "num_tokens": 111657428.0,
      "step": 55100
    },
    {
      "entropy": 5.849756717681885,
      "epoch": 4.815607795158613,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002912270641828753,
      "loss": 5.2745,
      "mean_token_accuracy": 0.20552761405706405,
      "num_tokens": 111668614.0,
      "step": 55105
    },
    {
      "entropy": 5.684843111038208,
      "epoch": 4.816044743511317,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029119598407466866,
      "loss": 5.1068,
      "mean_token_accuracy": 0.22194218188524245,
      "num_tokens": 111677506.0,
      "step": 55110
    },
    {
      "entropy": 5.69975323677063,
      "epoch": 4.816481691864022,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002911649036558145,
      "loss": 5.156,
      "mean_token_accuracy": 0.21548077166080476,
      "num_tokens": 111686316.0,
      "step": 55115
    },
    {
      "entropy": 5.712018823623657,
      "epoch": 4.816918640216726,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002911338229269091,
      "loss": 5.1725,
      "mean_token_accuracy": 0.21857229322195054,
      "num_tokens": 111696788.0,
      "step": 55120
    },
    {
      "entropy": 5.754030847549439,
      "epoch": 4.8173555885694315,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029110274188854843,
      "loss": 5.1128,
      "mean_token_accuracy": 0.21766848415136336,
      "num_tokens": 111706868.0,
      "step": 55125
    },
    {
      "entropy": 5.76751594543457,
      "epoch": 4.817792536922136,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002910716605413288,
      "loss": 5.2271,
      "mean_token_accuracy": 0.2130691349506378,
      "num_tokens": 111716401.0,
      "step": 55130
    },
    {
      "entropy": 5.707801818847656,
      "epoch": 4.81822948527484,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002910405788858462,
      "loss": 5.1615,
      "mean_token_accuracy": 0.22055586874485017,
      "num_tokens": 111725797.0,
      "step": 55135
    },
    {
      "entropy": 5.627292823791504,
      "epoch": 4.818666433627545,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000291009496922697,
      "loss": 5.1029,
      "mean_token_accuracy": 0.22034774720668793,
      "num_tokens": 111735906.0,
      "step": 55140
    },
    {
      "entropy": 5.634447336196899,
      "epoch": 4.81910338198025,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002909784146524773,
      "loss": 5.1204,
      "mean_token_accuracy": 0.2232227146625519,
      "num_tokens": 111745368.0,
      "step": 55145
    },
    {
      "entropy": 5.724129295349121,
      "epoch": 4.819540330332955,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029094733207578316,
      "loss": 5.2171,
      "mean_token_accuracy": 0.21545556783676148,
      "num_tokens": 111755196.0,
      "step": 55150
    },
    {
      "entropy": 5.743943071365356,
      "epoch": 4.819977278685659,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002909162491932108,
      "loss": 5.1846,
      "mean_token_accuracy": 0.21058716326951982,
      "num_tokens": 111765617.0,
      "step": 55155
    },
    {
      "entropy": 5.766711235046387,
      "epoch": 4.8204142270383645,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029088516600535644,
      "loss": 5.2302,
      "mean_token_accuracy": 0.21333785206079484,
      "num_tokens": 111776667.0,
      "step": 55160
    },
    {
      "entropy": 5.755307054519653,
      "epoch": 4.820851175391069,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002908540825128163,
      "loss": 5.1595,
      "mean_token_accuracy": 0.21822539567947388,
      "num_tokens": 111785558.0,
      "step": 55165
    },
    {
      "entropy": 5.755583333969116,
      "epoch": 4.8212881237437735,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002908229987161866,
      "loss": 5.2184,
      "mean_token_accuracy": 0.20884118229150772,
      "num_tokens": 111795406.0,
      "step": 55170
    },
    {
      "entropy": 5.753202629089356,
      "epoch": 4.821725072096478,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029079191461606343,
      "loss": 5.2533,
      "mean_token_accuracy": 0.20619907528162001,
      "num_tokens": 111806426.0,
      "step": 55175
    },
    {
      "entropy": 5.741737413406372,
      "epoch": 4.822162020449183,
      "grad_norm": 1.15625,
      "learning_rate": 0.000290760830213043,
      "loss": 5.2074,
      "mean_token_accuracy": 0.21512620151042938,
      "num_tokens": 111816140.0,
      "step": 55180
    },
    {
      "entropy": 5.786946392059326,
      "epoch": 4.822598968801888,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002907297455077217,
      "loss": 5.1975,
      "mean_token_accuracy": 0.2182660683989525,
      "num_tokens": 111826349.0,
      "step": 55185
    },
    {
      "entropy": 5.693618869781494,
      "epoch": 4.823035917154592,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002906986605006956,
      "loss": 5.0955,
      "mean_token_accuracy": 0.21951317340135573,
      "num_tokens": 111837164.0,
      "step": 55190
    },
    {
      "entropy": 5.637372398376465,
      "epoch": 4.823472865507297,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029066757519256105,
      "loss": 5.1185,
      "mean_token_accuracy": 0.21486842930316924,
      "num_tokens": 111847268.0,
      "step": 55195
    },
    {
      "entropy": 5.7450957775115965,
      "epoch": 4.823909813860002,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029063648958391403,
      "loss": 5.1838,
      "mean_token_accuracy": 0.21254131495952605,
      "num_tokens": 111856154.0,
      "step": 55200
    },
    {
      "entropy": 5.697121667861938,
      "epoch": 4.8243467622127065,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002906054036753511,
      "loss": 5.1532,
      "mean_token_accuracy": 0.201450677216053,
      "num_tokens": 111865903.0,
      "step": 55205
    },
    {
      "entropy": 5.772738075256347,
      "epoch": 4.824783710565411,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029057431746746826,
      "loss": 5.1777,
      "mean_token_accuracy": 0.2109357386827469,
      "num_tokens": 111875873.0,
      "step": 55210
    },
    {
      "entropy": 5.732186937332154,
      "epoch": 4.825220658918116,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029054323096086186,
      "loss": 5.1871,
      "mean_token_accuracy": 0.20780181437730788,
      "num_tokens": 111886874.0,
      "step": 55215
    },
    {
      "entropy": 5.721883249282837,
      "epoch": 4.825657607270821,
      "grad_norm": 1.640625,
      "learning_rate": 0.0002905121441561282,
      "loss": 5.2123,
      "mean_token_accuracy": 0.21943912357091905,
      "num_tokens": 111896889.0,
      "step": 55220
    },
    {
      "entropy": 5.731064176559448,
      "epoch": 4.826094555623525,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029048105705386346,
      "loss": 5.1594,
      "mean_token_accuracy": 0.2104778230190277,
      "num_tokens": 111906155.0,
      "step": 55225
    },
    {
      "entropy": 5.744929027557373,
      "epoch": 4.82653150397623,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029044996965466393,
      "loss": 5.1651,
      "mean_token_accuracy": 0.21520878225564957,
      "num_tokens": 111915244.0,
      "step": 55230
    },
    {
      "entropy": 5.7096446514129635,
      "epoch": 4.826968452328934,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029041888195912587,
      "loss": 5.222,
      "mean_token_accuracy": 0.2161304235458374,
      "num_tokens": 111925576.0,
      "step": 55235
    },
    {
      "entropy": 5.745758199691773,
      "epoch": 4.82740540068164,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002903877939678456,
      "loss": 5.1618,
      "mean_token_accuracy": 0.2201346203684807,
      "num_tokens": 111937015.0,
      "step": 55240
    },
    {
      "entropy": 5.769502830505371,
      "epoch": 4.827842349034344,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029035670568141936,
      "loss": 5.2095,
      "mean_token_accuracy": 0.21080504655838012,
      "num_tokens": 111946730.0,
      "step": 55245
    },
    {
      "entropy": 5.770426464080811,
      "epoch": 4.8282792973870485,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002903256171004434,
      "loss": 5.1492,
      "mean_token_accuracy": 0.21919335126876832,
      "num_tokens": 111956026.0,
      "step": 55250
    },
    {
      "entropy": 5.7038524627685545,
      "epoch": 4.828716245739754,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002902945282255142,
      "loss": 5.1952,
      "mean_token_accuracy": 0.21913581788539888,
      "num_tokens": 111966453.0,
      "step": 55255
    },
    {
      "entropy": 5.713424015045166,
      "epoch": 4.829153194092458,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002902634390572279,
      "loss": 5.1558,
      "mean_token_accuracy": 0.21493232548236846,
      "num_tokens": 111977318.0,
      "step": 55260
    },
    {
      "entropy": 5.746366834640503,
      "epoch": 4.829590142445163,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002902323495961808,
      "loss": 5.2256,
      "mean_token_accuracy": 0.2094702571630478,
      "num_tokens": 111987448.0,
      "step": 55265
    },
    {
      "entropy": 5.781856393814087,
      "epoch": 4.830027090797868,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029020125984296927,
      "loss": 5.2496,
      "mean_token_accuracy": 0.20759304165840148,
      "num_tokens": 111997222.0,
      "step": 55270
    },
    {
      "entropy": 5.79464750289917,
      "epoch": 4.830464039150573,
      "grad_norm": 1.0,
      "learning_rate": 0.0002901701697981896,
      "loss": 5.1917,
      "mean_token_accuracy": 0.21543731689453124,
      "num_tokens": 112007537.0,
      "step": 55275
    },
    {
      "entropy": 5.706589078903198,
      "epoch": 4.830900987503277,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002901390794624381,
      "loss": 5.1423,
      "mean_token_accuracy": 0.21366411447525024,
      "num_tokens": 112017511.0,
      "step": 55280
    },
    {
      "entropy": 5.672612857818604,
      "epoch": 4.831337935855982,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002901079888363112,
      "loss": 5.1159,
      "mean_token_accuracy": 0.21915527135133744,
      "num_tokens": 112027139.0,
      "step": 55285
    },
    {
      "entropy": 5.713265991210937,
      "epoch": 4.831774884208686,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002900768979204051,
      "loss": 5.1783,
      "mean_token_accuracy": 0.21701471954584123,
      "num_tokens": 112037643.0,
      "step": 55290
    },
    {
      "entropy": 5.720634269714355,
      "epoch": 4.832211832561391,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029004580671531624,
      "loss": 5.1684,
      "mean_token_accuracy": 0.21684591621160507,
      "num_tokens": 112048763.0,
      "step": 55295
    },
    {
      "entropy": 5.774409246444702,
      "epoch": 4.832648780914096,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002900147152216408,
      "loss": 5.2139,
      "mean_token_accuracy": 0.20699829757213592,
      "num_tokens": 112058883.0,
      "step": 55300
    },
    {
      "entropy": 5.708579778671265,
      "epoch": 4.8330857292668,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028998362343997543,
      "loss": 5.1707,
      "mean_token_accuracy": 0.21266497075557708,
      "num_tokens": 112069038.0,
      "step": 55305
    },
    {
      "entropy": 5.7538364887237545,
      "epoch": 4.833522677619506,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002899525313709162,
      "loss": 5.2145,
      "mean_token_accuracy": 0.21607211381196975,
      "num_tokens": 112078599.0,
      "step": 55310
    },
    {
      "entropy": 5.810165023803711,
      "epoch": 4.83395962597221,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028992143901505955,
      "loss": 5.2164,
      "mean_token_accuracy": 0.2080511674284935,
      "num_tokens": 112088671.0,
      "step": 55315
    },
    {
      "entropy": 5.7523521900177,
      "epoch": 4.834396574324915,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028989034637300206,
      "loss": 5.2016,
      "mean_token_accuracy": 0.21389421075582504,
      "num_tokens": 112098428.0,
      "step": 55320
    },
    {
      "entropy": 5.707657241821289,
      "epoch": 4.834833522677619,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028985925344533975,
      "loss": 5.1571,
      "mean_token_accuracy": 0.21934026181697847,
      "num_tokens": 112109271.0,
      "step": 55325
    },
    {
      "entropy": 5.697061586380005,
      "epoch": 4.8352704710303245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002898281602326692,
      "loss": 5.1253,
      "mean_token_accuracy": 0.21828444451093673,
      "num_tokens": 112118755.0,
      "step": 55330
    },
    {
      "entropy": 5.768978881835937,
      "epoch": 4.835707419383029,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002897970667355868,
      "loss": 5.2597,
      "mean_token_accuracy": 0.20364119559526445,
      "num_tokens": 112129628.0,
      "step": 55335
    },
    {
      "entropy": 5.745412445068359,
      "epoch": 4.836144367735733,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002897659729546889,
      "loss": 5.2205,
      "mean_token_accuracy": 0.21098894625902176,
      "num_tokens": 112140188.0,
      "step": 55340
    },
    {
      "entropy": 5.6584557056427,
      "epoch": 4.836581316088438,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002897348788905719,
      "loss": 5.1718,
      "mean_token_accuracy": 0.21966925263404846,
      "num_tokens": 112149807.0,
      "step": 55345
    },
    {
      "entropy": 5.681790351867676,
      "epoch": 4.837018264441143,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028970378454383223,
      "loss": 5.1933,
      "mean_token_accuracy": 0.21678835898637772,
      "num_tokens": 112160531.0,
      "step": 55350
    },
    {
      "entropy": 5.764501619338989,
      "epoch": 4.837455212793848,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028967268991506625,
      "loss": 5.1753,
      "mean_token_accuracy": 0.21829783618450166,
      "num_tokens": 112171173.0,
      "step": 55355
    },
    {
      "entropy": 5.756858825683594,
      "epoch": 4.837892161146552,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028964159500487034,
      "loss": 5.2111,
      "mean_token_accuracy": 0.21345387548208236,
      "num_tokens": 112181345.0,
      "step": 55360
    },
    {
      "entropy": 5.657019996643067,
      "epoch": 4.8383291094992575,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000289610499813841,
      "loss": 5.1401,
      "mean_token_accuracy": 0.22107028514146804,
      "num_tokens": 112190798.0,
      "step": 55365
    },
    {
      "entropy": 5.734310722351074,
      "epoch": 4.838766057851962,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002895794043425746,
      "loss": 5.146,
      "mean_token_accuracy": 0.21945756822824478,
      "num_tokens": 112199836.0,
      "step": 55370
    },
    {
      "entropy": 5.7040589332580565,
      "epoch": 4.8392030062046665,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028954830859166767,
      "loss": 5.1264,
      "mean_token_accuracy": 0.2202705129981041,
      "num_tokens": 112209885.0,
      "step": 55375
    },
    {
      "entropy": 5.742745685577392,
      "epoch": 4.839639954557371,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028951721256171653,
      "loss": 5.1823,
      "mean_token_accuracy": 0.21357880681753158,
      "num_tokens": 112219775.0,
      "step": 55380
    },
    {
      "entropy": 5.712052822113037,
      "epoch": 4.840076902910076,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028948611625331756,
      "loss": 5.1815,
      "mean_token_accuracy": 0.21717533618211746,
      "num_tokens": 112230354.0,
      "step": 55385
    },
    {
      "entropy": 5.751460409164428,
      "epoch": 4.840513851262781,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002894550196670674,
      "loss": 5.211,
      "mean_token_accuracy": 0.21093082427978516,
      "num_tokens": 112240821.0,
      "step": 55390
    },
    {
      "entropy": 5.797072792053223,
      "epoch": 4.840950799615485,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028942392280356227,
      "loss": 5.1906,
      "mean_token_accuracy": 0.2140906721353531,
      "num_tokens": 112250561.0,
      "step": 55395
    },
    {
      "entropy": 5.745711803436279,
      "epoch": 4.84138774796819,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028939282566339884,
      "loss": 5.1277,
      "mean_token_accuracy": 0.21935701221227646,
      "num_tokens": 112261829.0,
      "step": 55400
    },
    {
      "entropy": 5.730072546005249,
      "epoch": 4.841824696320895,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002893617282471734,
      "loss": 5.1804,
      "mean_token_accuracy": 0.2139231875538826,
      "num_tokens": 112272277.0,
      "step": 55405
    },
    {
      "entropy": 5.711677694320679,
      "epoch": 4.8422616446735995,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002893306305554825,
      "loss": 5.1652,
      "mean_token_accuracy": 0.2222071424126625,
      "num_tokens": 112283629.0,
      "step": 55410
    },
    {
      "entropy": 5.7431567192077635,
      "epoch": 4.842698593026304,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028929953258892264,
      "loss": 5.1454,
      "mean_token_accuracy": 0.21994926184415817,
      "num_tokens": 112293278.0,
      "step": 55415
    },
    {
      "entropy": 5.749947023391724,
      "epoch": 4.843135541379009,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028926843434809023,
      "loss": 5.1154,
      "mean_token_accuracy": 0.21697156876325607,
      "num_tokens": 112303293.0,
      "step": 55420
    },
    {
      "entropy": 5.6628447532653805,
      "epoch": 4.843572489731714,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002892373358335819,
      "loss": 5.1147,
      "mean_token_accuracy": 0.22281646430492402,
      "num_tokens": 112313659.0,
      "step": 55425
    },
    {
      "entropy": 5.805704832077026,
      "epoch": 4.844009438084418,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028920623704599386,
      "loss": 5.2927,
      "mean_token_accuracy": 0.20629142820835114,
      "num_tokens": 112324923.0,
      "step": 55430
    },
    {
      "entropy": 5.739860534667969,
      "epoch": 4.844446386437123,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002891751379859228,
      "loss": 5.1602,
      "mean_token_accuracy": 0.21265140324831008,
      "num_tokens": 112334803.0,
      "step": 55435
    },
    {
      "entropy": 5.6653954029083256,
      "epoch": 4.844883334789828,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002891440386539651,
      "loss": 5.1498,
      "mean_token_accuracy": 0.22313251942396164,
      "num_tokens": 112344909.0,
      "step": 55440
    },
    {
      "entropy": 5.7491248607635494,
      "epoch": 4.845320283142533,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028911293905071735,
      "loss": 5.141,
      "mean_token_accuracy": 0.2157704085111618,
      "num_tokens": 112355191.0,
      "step": 55445
    },
    {
      "entropy": 5.783435344696045,
      "epoch": 4.845757231495237,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028908183917677603,
      "loss": 5.1859,
      "mean_token_accuracy": 0.21390270292758942,
      "num_tokens": 112365055.0,
      "step": 55450
    },
    {
      "entropy": 5.736615133285523,
      "epoch": 4.8461941798479415,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002890507390327377,
      "loss": 5.2475,
      "mean_token_accuracy": 0.21087306439876558,
      "num_tokens": 112375775.0,
      "step": 55455
    },
    {
      "entropy": 5.762967920303344,
      "epoch": 4.846631128200647,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028901963861919876,
      "loss": 5.2113,
      "mean_token_accuracy": 0.20952051430940627,
      "num_tokens": 112385734.0,
      "step": 55460
    },
    {
      "entropy": 5.7011724472045895,
      "epoch": 4.847068076553351,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002889885379367559,
      "loss": 5.1015,
      "mean_token_accuracy": 0.21634086817502976,
      "num_tokens": 112396212.0,
      "step": 55465
    },
    {
      "entropy": 5.727505302429199,
      "epoch": 4.847505024906056,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002889574369860055,
      "loss": 5.2152,
      "mean_token_accuracy": 0.20853934735059737,
      "num_tokens": 112405982.0,
      "step": 55470
    },
    {
      "entropy": 5.752575445175171,
      "epoch": 4.847941973258761,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002889263357675442,
      "loss": 5.2269,
      "mean_token_accuracy": 0.2132108986377716,
      "num_tokens": 112415976.0,
      "step": 55475
    },
    {
      "entropy": 5.7381621360778805,
      "epoch": 4.848378921611466,
      "grad_norm": 1.125,
      "learning_rate": 0.0002888952342819684,
      "loss": 5.1989,
      "mean_token_accuracy": 0.21630951911211013,
      "num_tokens": 112426284.0,
      "step": 55480
    },
    {
      "entropy": 5.723618412017823,
      "epoch": 4.84881586996417,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028886413252987483,
      "loss": 5.0712,
      "mean_token_accuracy": 0.21814430207014085,
      "num_tokens": 112436467.0,
      "step": 55485
    },
    {
      "entropy": 5.711077737808227,
      "epoch": 4.849252818316875,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028883303051185975,
      "loss": 5.1463,
      "mean_token_accuracy": 0.2089372158050537,
      "num_tokens": 112446086.0,
      "step": 55490
    },
    {
      "entropy": 5.688257503509521,
      "epoch": 4.84968976666958,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028880192822852,
      "loss": 5.1949,
      "mean_token_accuracy": 0.2079111695289612,
      "num_tokens": 112456202.0,
      "step": 55495
    },
    {
      "entropy": 5.717661333084107,
      "epoch": 4.850126715022284,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002887708256804521,
      "loss": 5.1966,
      "mean_token_accuracy": 0.20909526348114013,
      "num_tokens": 112465836.0,
      "step": 55500
    },
    {
      "entropy": 5.725714826583863,
      "epoch": 4.850563663374989,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002887397228682525,
      "loss": 5.2027,
      "mean_token_accuracy": 0.2163031667470932,
      "num_tokens": 112475966.0,
      "step": 55505
    },
    {
      "entropy": 5.828962516784668,
      "epoch": 4.851000611727693,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002887086197925178,
      "loss": 5.3265,
      "mean_token_accuracy": 0.19713015109300613,
      "num_tokens": 112485232.0,
      "step": 55510
    },
    {
      "entropy": 5.709229803085327,
      "epoch": 4.851437560080399,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002886775164538446,
      "loss": 5.1607,
      "mean_token_accuracy": 0.21250830739736556,
      "num_tokens": 112495954.0,
      "step": 55515
    },
    {
      "entropy": 5.705896472930908,
      "epoch": 4.851874508433103,
      "grad_norm": 1.125,
      "learning_rate": 0.00028864641285282944,
      "loss": 5.1354,
      "mean_token_accuracy": 0.2191699415445328,
      "num_tokens": 112505492.0,
      "step": 55520
    },
    {
      "entropy": 5.68035740852356,
      "epoch": 4.852311456785808,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000288615308990069,
      "loss": 5.1542,
      "mean_token_accuracy": 0.219952230155468,
      "num_tokens": 112515216.0,
      "step": 55525
    },
    {
      "entropy": 5.701674699783325,
      "epoch": 4.852748405138513,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002885842048661598,
      "loss": 5.0948,
      "mean_token_accuracy": 0.2226253941655159,
      "num_tokens": 112524797.0,
      "step": 55530
    },
    {
      "entropy": 5.763986539840698,
      "epoch": 4.8531853534912175,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002885531004816983,
      "loss": 5.216,
      "mean_token_accuracy": 0.21653524041175842,
      "num_tokens": 112535057.0,
      "step": 55535
    },
    {
      "entropy": 5.798544692993164,
      "epoch": 4.853622301843922,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002885219958372813,
      "loss": 5.2167,
      "mean_token_accuracy": 0.21585002541542053,
      "num_tokens": 112545926.0,
      "step": 55540
    },
    {
      "entropy": 5.71905927658081,
      "epoch": 4.854059250196626,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002884908909335054,
      "loss": 5.1745,
      "mean_token_accuracy": 0.21385369151830674,
      "num_tokens": 112556367.0,
      "step": 55545
    },
    {
      "entropy": 5.800466442108155,
      "epoch": 4.854496198549332,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002884597857709671,
      "loss": 5.2299,
      "mean_token_accuracy": 0.21844495683908463,
      "num_tokens": 112566862.0,
      "step": 55550
    },
    {
      "entropy": 5.748452472686767,
      "epoch": 4.854933146902036,
      "grad_norm": 1.03125,
      "learning_rate": 0.000288428680350263,
      "loss": 5.2071,
      "mean_token_accuracy": 0.21247778385877608,
      "num_tokens": 112577306.0,
      "step": 55555
    },
    {
      "entropy": 5.778145503997803,
      "epoch": 4.855370095254741,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028839757467198987,
      "loss": 5.205,
      "mean_token_accuracy": 0.21530607491731643,
      "num_tokens": 112587865.0,
      "step": 55560
    },
    {
      "entropy": 5.747153329849243,
      "epoch": 4.855807043607445,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002883664687367442,
      "loss": 5.2024,
      "mean_token_accuracy": 0.21239830404520035,
      "num_tokens": 112597103.0,
      "step": 55565
    },
    {
      "entropy": 5.733795928955078,
      "epoch": 4.8562439919601506,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028833536254512264,
      "loss": 5.1266,
      "mean_token_accuracy": 0.21691207438707352,
      "num_tokens": 112606971.0,
      "step": 55570
    },
    {
      "entropy": 5.810974740982056,
      "epoch": 4.856680940312855,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002883042560977219,
      "loss": 5.2857,
      "mean_token_accuracy": 0.2093384876847267,
      "num_tokens": 112617588.0,
      "step": 55575
    },
    {
      "entropy": 5.781201648712158,
      "epoch": 4.8571178886655595,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028827314939513843,
      "loss": 5.1732,
      "mean_token_accuracy": 0.2198425367474556,
      "num_tokens": 112627854.0,
      "step": 55580
    },
    {
      "entropy": 5.731297731399536,
      "epoch": 4.857554837018265,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028824204243796907,
      "loss": 5.2344,
      "mean_token_accuracy": 0.213006092607975,
      "num_tokens": 112637969.0,
      "step": 55585
    },
    {
      "entropy": 5.771115827560425,
      "epoch": 4.857991785370969,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028821093522681044,
      "loss": 5.213,
      "mean_token_accuracy": 0.21111010760068893,
      "num_tokens": 112648482.0,
      "step": 55590
    },
    {
      "entropy": 5.728520345687866,
      "epoch": 4.858428733723674,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002881798277622591,
      "loss": 5.1709,
      "mean_token_accuracy": 0.2180999681353569,
      "num_tokens": 112658256.0,
      "step": 55595
    },
    {
      "entropy": 5.684829568862915,
      "epoch": 4.858865682076378,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002881487200449118,
      "loss": 5.0872,
      "mean_token_accuracy": 0.22789523750543594,
      "num_tokens": 112668301.0,
      "step": 55600
    },
    {
      "entropy": 5.746471357345581,
      "epoch": 4.859302630429084,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002881176120753651,
      "loss": 5.156,
      "mean_token_accuracy": 0.214575132727623,
      "num_tokens": 112679215.0,
      "step": 55605
    },
    {
      "entropy": 5.831026792526245,
      "epoch": 4.859739578781788,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028808650385421573,
      "loss": 5.2471,
      "mean_token_accuracy": 0.20663468837738036,
      "num_tokens": 112689304.0,
      "step": 55610
    },
    {
      "entropy": 5.709692096710205,
      "epoch": 4.8601765271344926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028805539538206043,
      "loss": 5.13,
      "mean_token_accuracy": 0.2154805615544319,
      "num_tokens": 112698726.0,
      "step": 55615
    },
    {
      "entropy": 5.736917495727539,
      "epoch": 4.860613475487197,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028802428665949584,
      "loss": 5.1941,
      "mean_token_accuracy": 0.20732951313257217,
      "num_tokens": 112708500.0,
      "step": 55620
    },
    {
      "entropy": 5.727482318878174,
      "epoch": 4.861050423839902,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028799317768711844,
      "loss": 5.2531,
      "mean_token_accuracy": 0.20725856572389603,
      "num_tokens": 112719195.0,
      "step": 55625
    },
    {
      "entropy": 5.772289657592774,
      "epoch": 4.861487372192607,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002879620684655252,
      "loss": 5.2237,
      "mean_token_accuracy": 0.2077828362584114,
      "num_tokens": 112728610.0,
      "step": 55630
    },
    {
      "entropy": 5.736398792266845,
      "epoch": 4.861924320545311,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002879309589953126,
      "loss": 5.1779,
      "mean_token_accuracy": 0.20648748725652694,
      "num_tokens": 112738818.0,
      "step": 55635
    },
    {
      "entropy": 5.760472822189331,
      "epoch": 4.862361268898017,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002878998492770776,
      "loss": 5.2334,
      "mean_token_accuracy": 0.20762248784303666,
      "num_tokens": 112748965.0,
      "step": 55640
    },
    {
      "entropy": 5.782203340530396,
      "epoch": 4.862798217250721,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028786873931141655,
      "loss": 5.2193,
      "mean_token_accuracy": 0.2108073890209198,
      "num_tokens": 112758638.0,
      "step": 55645
    },
    {
      "entropy": 5.72031888961792,
      "epoch": 4.863235165603426,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002878376290989264,
      "loss": 5.187,
      "mean_token_accuracy": 0.2152616113424301,
      "num_tokens": 112768891.0,
      "step": 55650
    },
    {
      "entropy": 5.749126672744751,
      "epoch": 4.86367211395613,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028780651864020385,
      "loss": 5.1488,
      "mean_token_accuracy": 0.21435364782810212,
      "num_tokens": 112779938.0,
      "step": 55655
    },
    {
      "entropy": 5.793727779388428,
      "epoch": 4.864109062308835,
      "grad_norm": 1.125,
      "learning_rate": 0.0002877754079358455,
      "loss": 5.2511,
      "mean_token_accuracy": 0.2082228034734726,
      "num_tokens": 112791601.0,
      "step": 55660
    },
    {
      "entropy": 5.7129088878631595,
      "epoch": 4.86454601066154,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002877442969864482,
      "loss": 5.1353,
      "mean_token_accuracy": 0.22258741706609725,
      "num_tokens": 112802502.0,
      "step": 55665
    },
    {
      "entropy": 5.705753946304322,
      "epoch": 4.864982959014244,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028771318579260855,
      "loss": 5.1353,
      "mean_token_accuracy": 0.22212677150964738,
      "num_tokens": 112812336.0,
      "step": 55670
    },
    {
      "entropy": 5.696966361999512,
      "epoch": 4.865419907366949,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002876820743549234,
      "loss": 5.1576,
      "mean_token_accuracy": 0.2193630501627922,
      "num_tokens": 112822572.0,
      "step": 55675
    },
    {
      "entropy": 5.7499494552612305,
      "epoch": 4.865856855719654,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002876509626739893,
      "loss": 5.2378,
      "mean_token_accuracy": 0.21019668132066727,
      "num_tokens": 112832523.0,
      "step": 55680
    },
    {
      "entropy": 5.80539345741272,
      "epoch": 4.866293804072359,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002876198507504032,
      "loss": 5.1938,
      "mean_token_accuracy": 0.2177655205130577,
      "num_tokens": 112842511.0,
      "step": 55685
    },
    {
      "entropy": 5.772046518325806,
      "epoch": 4.866730752425063,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028758873858476174,
      "loss": 5.1547,
      "mean_token_accuracy": 0.2173904374241829,
      "num_tokens": 112852207.0,
      "step": 55690
    },
    {
      "entropy": 5.83592529296875,
      "epoch": 4.8671677007777685,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028755762617766167,
      "loss": 5.2435,
      "mean_token_accuracy": 0.20232122391462326,
      "num_tokens": 112862574.0,
      "step": 55695
    },
    {
      "entropy": 5.739378547668457,
      "epoch": 4.867604649130473,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002875265135296997,
      "loss": 5.1726,
      "mean_token_accuracy": 0.21782728731632234,
      "num_tokens": 112872356.0,
      "step": 55700
    },
    {
      "entropy": 5.700423288345337,
      "epoch": 4.868041597483177,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028749540064147266,
      "loss": 5.1586,
      "mean_token_accuracy": 0.21925998777151107,
      "num_tokens": 112882958.0,
      "step": 55705
    },
    {
      "entropy": 5.741831684112549,
      "epoch": 4.868478545835882,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002874642875135773,
      "loss": 5.1564,
      "mean_token_accuracy": 0.21350433826446533,
      "num_tokens": 112892061.0,
      "step": 55710
    },
    {
      "entropy": 5.712700080871582,
      "epoch": 4.868915494188587,
      "grad_norm": 1.125,
      "learning_rate": 0.0002874331741466104,
      "loss": 5.1792,
      "mean_token_accuracy": 0.21304671317338944,
      "num_tokens": 112901940.0,
      "step": 55715
    },
    {
      "entropy": 5.770779705047607,
      "epoch": 4.869352442541292,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002874020605411686,
      "loss": 5.2197,
      "mean_token_accuracy": 0.21158721596002578,
      "num_tokens": 112911451.0,
      "step": 55720
    },
    {
      "entropy": 5.723216581344604,
      "epoch": 4.869789390893996,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002873709466978489,
      "loss": 5.14,
      "mean_token_accuracy": 0.2172623097896576,
      "num_tokens": 112921183.0,
      "step": 55725
    },
    {
      "entropy": 5.751719808578491,
      "epoch": 4.870226339246701,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000287339832617248,
      "loss": 5.2258,
      "mean_token_accuracy": 0.2091553419828415,
      "num_tokens": 112931658.0,
      "step": 55730
    },
    {
      "entropy": 5.726093673706055,
      "epoch": 4.870663287599406,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002873087182999626,
      "loss": 5.1206,
      "mean_token_accuracy": 0.21996334493160247,
      "num_tokens": 112942398.0,
      "step": 55735
    },
    {
      "entropy": 5.836611604690551,
      "epoch": 4.8711002359521105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028727760374658946,
      "loss": 5.3408,
      "mean_token_accuracy": 0.1995621904730797,
      "num_tokens": 112953012.0,
      "step": 55740
    },
    {
      "entropy": 5.710789346694947,
      "epoch": 4.871537184304815,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002872464889577255,
      "loss": 5.189,
      "mean_token_accuracy": 0.21160593181848525,
      "num_tokens": 112962740.0,
      "step": 55745
    },
    {
      "entropy": 5.803231239318848,
      "epoch": 4.87197413265752,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002872153739339674,
      "loss": 5.2516,
      "mean_token_accuracy": 0.20718563795089723,
      "num_tokens": 112973424.0,
      "step": 55750
    },
    {
      "entropy": 5.70398530960083,
      "epoch": 4.872411081010225,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028718425867591213,
      "loss": 5.0678,
      "mean_token_accuracy": 0.21257898211479187,
      "num_tokens": 112982582.0,
      "step": 55755
    },
    {
      "entropy": 5.711944437026977,
      "epoch": 4.872848029362929,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002871531431841563,
      "loss": 5.1249,
      "mean_token_accuracy": 0.21710333228111267,
      "num_tokens": 112992131.0,
      "step": 55760
    },
    {
      "entropy": 5.703410959243774,
      "epoch": 4.873284977715634,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002871220274592969,
      "loss": 5.1724,
      "mean_token_accuracy": 0.2145173266530037,
      "num_tokens": 113002176.0,
      "step": 55765
    },
    {
      "entropy": 5.737448406219483,
      "epoch": 4.873721926068339,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002870909115019306,
      "loss": 5.1407,
      "mean_token_accuracy": 0.22045733630657197,
      "num_tokens": 113011463.0,
      "step": 55770
    },
    {
      "entropy": 5.811934185028076,
      "epoch": 4.874158874421044,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002870597953126543,
      "loss": 5.2252,
      "mean_token_accuracy": 0.2089719444513321,
      "num_tokens": 113021012.0,
      "step": 55775
    },
    {
      "entropy": 5.811890316009522,
      "epoch": 4.874595822773748,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028702867889206484,
      "loss": 5.2489,
      "mean_token_accuracy": 0.20482703894376755,
      "num_tokens": 113030623.0,
      "step": 55780
    },
    {
      "entropy": 5.7963659286499025,
      "epoch": 4.8750327711264525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028699756224075893,
      "loss": 5.1945,
      "mean_token_accuracy": 0.21237918138504028,
      "num_tokens": 113040059.0,
      "step": 55785
    },
    {
      "entropy": 5.722334194183349,
      "epoch": 4.875469719479158,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002869664453593336,
      "loss": 5.1481,
      "mean_token_accuracy": 0.2142421394586563,
      "num_tokens": 113049102.0,
      "step": 55790
    },
    {
      "entropy": 5.699647903442383,
      "epoch": 4.875906667831862,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002869353282483855,
      "loss": 5.1108,
      "mean_token_accuracy": 0.22111679315567018,
      "num_tokens": 113059155.0,
      "step": 55795
    },
    {
      "entropy": 5.748638772964478,
      "epoch": 4.876343616184567,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028690421090851156,
      "loss": 5.1783,
      "mean_token_accuracy": 0.21989172846078872,
      "num_tokens": 113069716.0,
      "step": 55800
    },
    {
      "entropy": 5.747121715545655,
      "epoch": 4.876780564537272,
      "grad_norm": 1.125,
      "learning_rate": 0.0002868730933403087,
      "loss": 5.2203,
      "mean_token_accuracy": 0.21252707093954087,
      "num_tokens": 113078826.0,
      "step": 55805
    },
    {
      "entropy": 5.68862018585205,
      "epoch": 4.877217512889977,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028684197554437355,
      "loss": 5.1385,
      "mean_token_accuracy": 0.21602482348680496,
      "num_tokens": 113088626.0,
      "step": 55810
    },
    {
      "entropy": 5.780160284042358,
      "epoch": 4.877654461242681,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002868108575213032,
      "loss": 5.2839,
      "mean_token_accuracy": 0.20123777091503142,
      "num_tokens": 113097917.0,
      "step": 55815
    },
    {
      "entropy": 5.758562231063843,
      "epoch": 4.878091409595386,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028677973927169445,
      "loss": 5.189,
      "mean_token_accuracy": 0.2113079845905304,
      "num_tokens": 113109522.0,
      "step": 55820
    },
    {
      "entropy": 5.804228115081787,
      "epoch": 4.878528357948091,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000286748620796144,
      "loss": 5.2385,
      "mean_token_accuracy": 0.21113554686307906,
      "num_tokens": 113119063.0,
      "step": 55825
    },
    {
      "entropy": 5.733671569824219,
      "epoch": 4.878965306300795,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002867175020952489,
      "loss": 5.1812,
      "mean_token_accuracy": 0.21554885804653168,
      "num_tokens": 113130013.0,
      "step": 55830
    },
    {
      "entropy": 5.740290260314941,
      "epoch": 4.8794022546535,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028668638316960604,
      "loss": 5.2109,
      "mean_token_accuracy": 0.21595513224601745,
      "num_tokens": 113139900.0,
      "step": 55835
    },
    {
      "entropy": 5.807356643676758,
      "epoch": 4.879839203006204,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002866552640198121,
      "loss": 5.2638,
      "mean_token_accuracy": 0.20364703685045243,
      "num_tokens": 113149378.0,
      "step": 55840
    },
    {
      "entropy": 5.775939178466797,
      "epoch": 4.88027615135891,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002866241446464642,
      "loss": 5.2458,
      "mean_token_accuracy": 0.20813919305801393,
      "num_tokens": 113159637.0,
      "step": 55845
    },
    {
      "entropy": 5.767749547958374,
      "epoch": 4.880713099711614,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028659302505015913,
      "loss": 5.1534,
      "mean_token_accuracy": 0.21307235807180405,
      "num_tokens": 113169663.0,
      "step": 55850
    },
    {
      "entropy": 5.752093315124512,
      "epoch": 4.881150048064319,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028656190523149367,
      "loss": 5.1825,
      "mean_token_accuracy": 0.21296752691268922,
      "num_tokens": 113179492.0,
      "step": 55855
    },
    {
      "entropy": 5.70615520477295,
      "epoch": 4.881586996417024,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028653078519106485,
      "loss": 5.1351,
      "mean_token_accuracy": 0.22055363804101943,
      "num_tokens": 113190626.0,
      "step": 55860
    },
    {
      "entropy": 5.7368348121643065,
      "epoch": 4.8820239447697285,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002864996649294695,
      "loss": 5.1357,
      "mean_token_accuracy": 0.221841961145401,
      "num_tokens": 113200783.0,
      "step": 55865
    },
    {
      "entropy": 5.7454578399658205,
      "epoch": 4.882460893122433,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002864685444473046,
      "loss": 5.2375,
      "mean_token_accuracy": 0.2098264291882515,
      "num_tokens": 113211221.0,
      "step": 55870
    },
    {
      "entropy": 5.638657855987549,
      "epoch": 4.882897841475137,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000286437423745167,
      "loss": 5.0565,
      "mean_token_accuracy": 0.22598379254341125,
      "num_tokens": 113221170.0,
      "step": 55875
    },
    {
      "entropy": 5.618614625930786,
      "epoch": 4.883334789827842,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002864063028236536,
      "loss": 5.1203,
      "mean_token_accuracy": 0.2291001334786415,
      "num_tokens": 113231748.0,
      "step": 55880
    },
    {
      "entropy": 5.687842321395874,
      "epoch": 4.883771738180547,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028637518168336134,
      "loss": 5.1533,
      "mean_token_accuracy": 0.22508975118398666,
      "num_tokens": 113241801.0,
      "step": 55885
    },
    {
      "entropy": 5.782644033432007,
      "epoch": 4.884208686533252,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002863440603248871,
      "loss": 5.2596,
      "mean_token_accuracy": 0.21211432814598083,
      "num_tokens": 113251181.0,
      "step": 55890
    },
    {
      "entropy": 5.720224475860595,
      "epoch": 4.884645634885956,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002863129387488279,
      "loss": 5.1406,
      "mean_token_accuracy": 0.2171988919377327,
      "num_tokens": 113260249.0,
      "step": 55895
    },
    {
      "entropy": 5.773242235183716,
      "epoch": 4.8850825832386615,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002862818169557805,
      "loss": 5.1263,
      "mean_token_accuracy": 0.2124277248978615,
      "num_tokens": 113270481.0,
      "step": 55900
    },
    {
      "entropy": 5.622313690185547,
      "epoch": 4.885519531591366,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028625069494634206,
      "loss": 5.1454,
      "mean_token_accuracy": 0.21663544923067093,
      "num_tokens": 113280990.0,
      "step": 55905
    },
    {
      "entropy": 5.803869819641113,
      "epoch": 4.8859564799440705,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028621957272110935,
      "loss": 5.2315,
      "mean_token_accuracy": 0.20567741990089417,
      "num_tokens": 113291454.0,
      "step": 55910
    },
    {
      "entropy": 5.736435604095459,
      "epoch": 4.886393428296776,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002861884502806794,
      "loss": 5.1591,
      "mean_token_accuracy": 0.21216014176607131,
      "num_tokens": 113300825.0,
      "step": 55915
    },
    {
      "entropy": 5.789859342575073,
      "epoch": 4.88683037664948,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028615732762564904,
      "loss": 5.1566,
      "mean_token_accuracy": 0.22067820131778718,
      "num_tokens": 113311236.0,
      "step": 55920
    },
    {
      "entropy": 5.841843700408935,
      "epoch": 4.887267325002185,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028612620475661535,
      "loss": 5.2719,
      "mean_token_accuracy": 0.20751139670610427,
      "num_tokens": 113321072.0,
      "step": 55925
    },
    {
      "entropy": 5.773119783401489,
      "epoch": 4.887704273354889,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028609508167417517,
      "loss": 5.2318,
      "mean_token_accuracy": 0.21296197772026063,
      "num_tokens": 113331330.0,
      "step": 55930
    },
    {
      "entropy": 5.756232357025146,
      "epoch": 4.888141221707594,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002860639583789255,
      "loss": 5.2852,
      "mean_token_accuracy": 0.20309222638607025,
      "num_tokens": 113342290.0,
      "step": 55935
    },
    {
      "entropy": 5.750438690185547,
      "epoch": 4.888578170060299,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002860328348714634,
      "loss": 5.1483,
      "mean_token_accuracy": 0.21437667310237885,
      "num_tokens": 113351903.0,
      "step": 55940
    },
    {
      "entropy": 5.7361969470977785,
      "epoch": 4.8890151184130035,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002860017111523857,
      "loss": 5.1437,
      "mean_token_accuracy": 0.21657679975032806,
      "num_tokens": 113361658.0,
      "step": 55945
    },
    {
      "entropy": 5.731916713714599,
      "epoch": 4.889452066765708,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002859705872222894,
      "loss": 5.1934,
      "mean_token_accuracy": 0.21605057567358016,
      "num_tokens": 113372148.0,
      "step": 55950
    },
    {
      "entropy": 5.750705718994141,
      "epoch": 4.889889015118413,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028593946308177147,
      "loss": 5.2827,
      "mean_token_accuracy": 0.20859819054603576,
      "num_tokens": 113383775.0,
      "step": 55955
    },
    {
      "entropy": 5.7530395030975345,
      "epoch": 4.890325963471118,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028590833873142895,
      "loss": 5.1456,
      "mean_token_accuracy": 0.21787882298231126,
      "num_tokens": 113393854.0,
      "step": 55960
    },
    {
      "entropy": 5.75929856300354,
      "epoch": 4.890762911823822,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028587721417185877,
      "loss": 5.1685,
      "mean_token_accuracy": 0.21566808521747588,
      "num_tokens": 113404562.0,
      "step": 55965
    },
    {
      "entropy": 5.745816040039062,
      "epoch": 4.891199860176527,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002858460894036579,
      "loss": 5.2035,
      "mean_token_accuracy": 0.20912858247756957,
      "num_tokens": 113414082.0,
      "step": 55970
    },
    {
      "entropy": 5.758726167678833,
      "epoch": 4.891636808529232,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002858149644274234,
      "loss": 5.2011,
      "mean_token_accuracy": 0.2064880833029747,
      "num_tokens": 113424990.0,
      "step": 55975
    },
    {
      "entropy": 5.785908174514771,
      "epoch": 4.892073756881937,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002857838392437522,
      "loss": 5.2457,
      "mean_token_accuracy": 0.21137595027685166,
      "num_tokens": 113435850.0,
      "step": 55980
    },
    {
      "entropy": 5.776176881790161,
      "epoch": 4.892510705234641,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028575271385324123,
      "loss": 5.1955,
      "mean_token_accuracy": 0.21690904796123506,
      "num_tokens": 113445987.0,
      "step": 55985
    },
    {
      "entropy": 5.800772380828858,
      "epoch": 4.8929476535873455,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002857215882564877,
      "loss": 5.262,
      "mean_token_accuracy": 0.20704723745584488,
      "num_tokens": 113454879.0,
      "step": 55990
    },
    {
      "entropy": 5.745269823074341,
      "epoch": 4.893384601940051,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002856904624540884,
      "loss": 5.2184,
      "mean_token_accuracy": 0.20825385600328444,
      "num_tokens": 113464840.0,
      "step": 55995
    },
    {
      "entropy": 5.72880482673645,
      "epoch": 4.893821550292755,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028565933644664036,
      "loss": 5.2495,
      "mean_token_accuracy": 0.21033222675323487,
      "num_tokens": 113474279.0,
      "step": 56000
    },
    {
      "entropy": 5.718190431594849,
      "epoch": 4.89425849864546,
      "grad_norm": 1.125,
      "learning_rate": 0.0002856282102347408,
      "loss": 5.1464,
      "mean_token_accuracy": 0.2211580067873001,
      "num_tokens": 113484238.0,
      "step": 56005
    },
    {
      "entropy": 5.790044736862183,
      "epoch": 4.894695446998165,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028559708381898654,
      "loss": 5.1756,
      "mean_token_accuracy": 0.21346895396709442,
      "num_tokens": 113494961.0,
      "step": 56010
    },
    {
      "entropy": 5.770692157745361,
      "epoch": 4.89513239535087,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002855659571999747,
      "loss": 5.2936,
      "mean_token_accuracy": 0.2134101375937462,
      "num_tokens": 113505893.0,
      "step": 56015
    },
    {
      "entropy": 5.74025526046753,
      "epoch": 4.895569343703574,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028553483037830225,
      "loss": 5.1726,
      "mean_token_accuracy": 0.21413908004760743,
      "num_tokens": 113515919.0,
      "step": 56020
    },
    {
      "entropy": 5.695375585556031,
      "epoch": 4.896006292056279,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002855037033545662,
      "loss": 5.0989,
      "mean_token_accuracy": 0.22058930099010468,
      "num_tokens": 113525748.0,
      "step": 56025
    },
    {
      "entropy": 5.770010900497437,
      "epoch": 4.896443240408984,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002854725761293636,
      "loss": 5.2485,
      "mean_token_accuracy": 0.21337591260671615,
      "num_tokens": 113536803.0,
      "step": 56030
    },
    {
      "entropy": 5.720101642608642,
      "epoch": 4.896880188761688,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028544144870329153,
      "loss": 5.1191,
      "mean_token_accuracy": 0.21829664260149,
      "num_tokens": 113547099.0,
      "step": 56035
    },
    {
      "entropy": 5.743057107925415,
      "epoch": 4.897317137114393,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002854103210769471,
      "loss": 5.2199,
      "mean_token_accuracy": 0.2129770502448082,
      "num_tokens": 113557334.0,
      "step": 56040
    },
    {
      "entropy": 5.7601768493652346,
      "epoch": 4.897754085467097,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028537919325092714,
      "loss": 5.2179,
      "mean_token_accuracy": 0.2079107269644737,
      "num_tokens": 113566112.0,
      "step": 56045
    },
    {
      "entropy": 5.749208211898804,
      "epoch": 4.898191033819803,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028534806522582885,
      "loss": 5.2324,
      "mean_token_accuracy": 0.2053006574511528,
      "num_tokens": 113576069.0,
      "step": 56050
    },
    {
      "entropy": 5.725499153137207,
      "epoch": 4.898627982172507,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028531693700224926,
      "loss": 5.2133,
      "mean_token_accuracy": 0.21157745718955995,
      "num_tokens": 113585935.0,
      "step": 56055
    },
    {
      "entropy": 5.708007287979126,
      "epoch": 4.899064930525212,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028528580858078537,
      "loss": 5.1442,
      "mean_token_accuracy": 0.22339177429676055,
      "num_tokens": 113597413.0,
      "step": 56060
    },
    {
      "entropy": 5.773425912857055,
      "epoch": 4.899501878877917,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028525467996203425,
      "loss": 5.1818,
      "mean_token_accuracy": 0.21352264434099197,
      "num_tokens": 113607446.0,
      "step": 56065
    },
    {
      "entropy": 5.7560622215271,
      "epoch": 4.8999388272306215,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000285223551146593,
      "loss": 5.11,
      "mean_token_accuracy": 0.22052198946475982,
      "num_tokens": 113616847.0,
      "step": 56070
    },
    {
      "entropy": 5.77470908164978,
      "epoch": 4.900375775583326,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002851924221350588,
      "loss": 5.2391,
      "mean_token_accuracy": 0.2047356754541397,
      "num_tokens": 113626809.0,
      "step": 56075
    },
    {
      "entropy": 5.742395210266113,
      "epoch": 4.90081272393603,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002851612929280285,
      "loss": 5.1113,
      "mean_token_accuracy": 0.2188296064734459,
      "num_tokens": 113636695.0,
      "step": 56080
    },
    {
      "entropy": 5.761635160446167,
      "epoch": 4.901249672288736,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002851301635260993,
      "loss": 5.1798,
      "mean_token_accuracy": 0.2194493219256401,
      "num_tokens": 113646522.0,
      "step": 56085
    },
    {
      "entropy": 5.6508537292480465,
      "epoch": 4.90168662064144,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002850990339298683,
      "loss": 5.0894,
      "mean_token_accuracy": 0.22143255323171615,
      "num_tokens": 113656844.0,
      "step": 56090
    },
    {
      "entropy": 5.737072706222534,
      "epoch": 4.902123568994145,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002850679041399325,
      "loss": 5.1774,
      "mean_token_accuracy": 0.21892839521169663,
      "num_tokens": 113667466.0,
      "step": 56095
    },
    {
      "entropy": 5.738754558563232,
      "epoch": 4.902560517346849,
      "grad_norm": 1.078125,
      "learning_rate": 0.000285036774156889,
      "loss": 5.2006,
      "mean_token_accuracy": 0.21568211913108826,
      "num_tokens": 113677614.0,
      "step": 56100
    },
    {
      "entropy": 5.804470062255859,
      "epoch": 4.9029974656995545,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028500564398133496,
      "loss": 5.2065,
      "mean_token_accuracy": 0.21277347803115845,
      "num_tokens": 113687289.0,
      "step": 56105
    },
    {
      "entropy": 5.768615674972534,
      "epoch": 4.903434414052259,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028497451361386743,
      "loss": 5.2221,
      "mean_token_accuracy": 0.2102539971470833,
      "num_tokens": 113696695.0,
      "step": 56110
    },
    {
      "entropy": 5.718229341506958,
      "epoch": 4.9038713624049635,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028494338305508343,
      "loss": 5.1381,
      "mean_token_accuracy": 0.21699308156967162,
      "num_tokens": 113706952.0,
      "step": 56115
    },
    {
      "entropy": 5.715770959854126,
      "epoch": 4.904308310757669,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028491225230558014,
      "loss": 5.2416,
      "mean_token_accuracy": 0.2188661053776741,
      "num_tokens": 113717774.0,
      "step": 56120
    },
    {
      "entropy": 5.736741209030152,
      "epoch": 4.904745259110373,
      "grad_norm": 1.125,
      "learning_rate": 0.00028488112136595473,
      "loss": 5.1602,
      "mean_token_accuracy": 0.21653435826301576,
      "num_tokens": 113727479.0,
      "step": 56125
    },
    {
      "entropy": 5.720523643493652,
      "epoch": 4.905182207463078,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028484999023680417,
      "loss": 5.1214,
      "mean_token_accuracy": 0.21850937902927398,
      "num_tokens": 113738212.0,
      "step": 56130
    },
    {
      "entropy": 5.7906702041625975,
      "epoch": 4.905619155815782,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028481885891872574,
      "loss": 5.2472,
      "mean_token_accuracy": 0.21192692667245866,
      "num_tokens": 113748054.0,
      "step": 56135
    },
    {
      "entropy": 5.747129726409912,
      "epoch": 4.906056104168488,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028478772741231636,
      "loss": 5.1396,
      "mean_token_accuracy": 0.21840706765651702,
      "num_tokens": 113757756.0,
      "step": 56140
    },
    {
      "entropy": 5.706213045120239,
      "epoch": 4.906493052521192,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028475659571817325,
      "loss": 5.1367,
      "mean_token_accuracy": 0.2231571212410927,
      "num_tokens": 113767812.0,
      "step": 56145
    },
    {
      "entropy": 5.692752933502197,
      "epoch": 4.9069300008738965,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028472546383689353,
      "loss": 5.1779,
      "mean_token_accuracy": 0.2094934195280075,
      "num_tokens": 113779473.0,
      "step": 56150
    },
    {
      "entropy": 5.705496549606323,
      "epoch": 4.907366949226601,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028469433176907434,
      "loss": 5.14,
      "mean_token_accuracy": 0.2197796255350113,
      "num_tokens": 113788545.0,
      "step": 56155
    },
    {
      "entropy": 5.710284662246704,
      "epoch": 4.907803897579306,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002846631995153128,
      "loss": 5.1781,
      "mean_token_accuracy": 0.2126060888171196,
      "num_tokens": 113798713.0,
      "step": 56160
    },
    {
      "entropy": 5.804033946990967,
      "epoch": 4.908240845932011,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028463206707620597,
      "loss": 5.2077,
      "mean_token_accuracy": 0.21195212602615357,
      "num_tokens": 113809387.0,
      "step": 56165
    },
    {
      "entropy": 5.831253862380981,
      "epoch": 4.908677794284715,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002846009344523511,
      "loss": 5.1919,
      "mean_token_accuracy": 0.21438394635915756,
      "num_tokens": 113819867.0,
      "step": 56170
    },
    {
      "entropy": 5.719751501083374,
      "epoch": 4.909114742637421,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002845698016443452,
      "loss": 5.1952,
      "mean_token_accuracy": 0.21305272430181504,
      "num_tokens": 113830723.0,
      "step": 56175
    },
    {
      "entropy": 5.7082640647888185,
      "epoch": 4.909551690990125,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028453866865278557,
      "loss": 5.1702,
      "mean_token_accuracy": 0.21793736815452575,
      "num_tokens": 113839905.0,
      "step": 56180
    },
    {
      "entropy": 5.731372976303101,
      "epoch": 4.90998863934283,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028450753547826923,
      "loss": 5.153,
      "mean_token_accuracy": 0.21040772795677185,
      "num_tokens": 113848851.0,
      "step": 56185
    },
    {
      "entropy": 5.729773330688476,
      "epoch": 4.910425587695534,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028447640212139334,
      "loss": 5.2178,
      "mean_token_accuracy": 0.21685107052326202,
      "num_tokens": 113859157.0,
      "step": 56190
    },
    {
      "entropy": 5.703278589248657,
      "epoch": 4.910862536048239,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028444526858275514,
      "loss": 5.1343,
      "mean_token_accuracy": 0.219736747443676,
      "num_tokens": 113869083.0,
      "step": 56195
    },
    {
      "entropy": 5.783532381057739,
      "epoch": 4.911299484400944,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028441413486295173,
      "loss": 5.257,
      "mean_token_accuracy": 0.20964056104421616,
      "num_tokens": 113879665.0,
      "step": 56200
    },
    {
      "entropy": 5.798613834381103,
      "epoch": 4.911736432753648,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028438300096258036,
      "loss": 5.2423,
      "mean_token_accuracy": 0.20890228003263472,
      "num_tokens": 113889271.0,
      "step": 56205
    },
    {
      "entropy": 5.805928182601929,
      "epoch": 4.912173381106353,
      "grad_norm": 1.125,
      "learning_rate": 0.00028435186688223805,
      "loss": 5.2308,
      "mean_token_accuracy": 0.21433376967906953,
      "num_tokens": 113900504.0,
      "step": 56210
    },
    {
      "entropy": 5.756780767440796,
      "epoch": 4.912610329459058,
      "grad_norm": 1.03125,
      "learning_rate": 0.000284320732622522,
      "loss": 5.2231,
      "mean_token_accuracy": 0.20778333097696305,
      "num_tokens": 113910892.0,
      "step": 56215
    },
    {
      "entropy": 5.7645707607269285,
      "epoch": 4.913047277811763,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028428959818402945,
      "loss": 5.2314,
      "mean_token_accuracy": 0.21178022176027297,
      "num_tokens": 113921722.0,
      "step": 56220
    },
    {
      "entropy": 5.778825902938843,
      "epoch": 4.913484226164467,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028425846356735753,
      "loss": 5.1736,
      "mean_token_accuracy": 0.2119922935962677,
      "num_tokens": 113933183.0,
      "step": 56225
    },
    {
      "entropy": 5.678970670700073,
      "epoch": 4.9139211745171725,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002842273287731035,
      "loss": 5.176,
      "mean_token_accuracy": 0.2219328209757805,
      "num_tokens": 113943608.0,
      "step": 56230
    },
    {
      "entropy": 5.69344162940979,
      "epoch": 4.914358122869877,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028419619380186433,
      "loss": 5.1633,
      "mean_token_accuracy": 0.21574948877096176,
      "num_tokens": 113953065.0,
      "step": 56235
    },
    {
      "entropy": 5.5245523929595945,
      "epoch": 4.914795071222581,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002841650586542374,
      "loss": 4.9626,
      "mean_token_accuracy": 0.24044421166181565,
      "num_tokens": 113964188.0,
      "step": 56240
    },
    {
      "entropy": 5.697332572937012,
      "epoch": 4.915232019575286,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002841339233308198,
      "loss": 5.1617,
      "mean_token_accuracy": 0.2182936429977417,
      "num_tokens": 113973857.0,
      "step": 56245
    },
    {
      "entropy": 5.774263095855713,
      "epoch": 4.915668967927991,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028410278783220887,
      "loss": 5.2352,
      "mean_token_accuracy": 0.20570193380117416,
      "num_tokens": 113984158.0,
      "step": 56250
    },
    {
      "entropy": 5.757941818237304,
      "epoch": 4.916105916280696,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028407165215900166,
      "loss": 5.1986,
      "mean_token_accuracy": 0.2133970230817795,
      "num_tokens": 113994875.0,
      "step": 56255
    },
    {
      "entropy": 5.782089424133301,
      "epoch": 4.9165428646334,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002840405163117954,
      "loss": 5.2392,
      "mean_token_accuracy": 0.21225783079862595,
      "num_tokens": 114004947.0,
      "step": 56260
    },
    {
      "entropy": 5.784223651885986,
      "epoch": 4.916979812986105,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028400938029118726,
      "loss": 5.2033,
      "mean_token_accuracy": 0.21850354224443436,
      "num_tokens": 114015592.0,
      "step": 56265
    },
    {
      "entropy": 5.693281173706055,
      "epoch": 4.91741676133881,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002839782440977745,
      "loss": 5.1026,
      "mean_token_accuracy": 0.2220063790678978,
      "num_tokens": 114025538.0,
      "step": 56270
    },
    {
      "entropy": 5.741955184936524,
      "epoch": 4.9178537096915145,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002839471077321544,
      "loss": 5.1567,
      "mean_token_accuracy": 0.21623330116271972,
      "num_tokens": 114035196.0,
      "step": 56275
    },
    {
      "entropy": 5.71724681854248,
      "epoch": 4.918290658044219,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000283915971194924,
      "loss": 5.1387,
      "mean_token_accuracy": 0.21341091096401216,
      "num_tokens": 114045896.0,
      "step": 56280
    },
    {
      "entropy": 5.7306969165802,
      "epoch": 4.918727606396924,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028388483448668064,
      "loss": 5.2243,
      "mean_token_accuracy": 0.21151187717914582,
      "num_tokens": 114055854.0,
      "step": 56285
    },
    {
      "entropy": 5.740365839004516,
      "epoch": 4.919164554749629,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002838536976080214,
      "loss": 5.1829,
      "mean_token_accuracy": 0.2116010680794716,
      "num_tokens": 114066349.0,
      "step": 56290
    },
    {
      "entropy": 5.724569272994995,
      "epoch": 4.919601503102333,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002838225605595438,
      "loss": 5.2364,
      "mean_token_accuracy": 0.21227372586727142,
      "num_tokens": 114077374.0,
      "step": 56295
    },
    {
      "entropy": 5.706431436538696,
      "epoch": 4.920038451455038,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002837914233418447,
      "loss": 5.1364,
      "mean_token_accuracy": 0.2210673376917839,
      "num_tokens": 114088276.0,
      "step": 56300
    },
    {
      "entropy": 5.74147253036499,
      "epoch": 4.920475399807743,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028376028595552156,
      "loss": 5.2276,
      "mean_token_accuracy": 0.21757377535104752,
      "num_tokens": 114097983.0,
      "step": 56305
    },
    {
      "entropy": 5.689807510375976,
      "epoch": 4.9209123481604475,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002837291484011715,
      "loss": 5.122,
      "mean_token_accuracy": 0.22092814445495607,
      "num_tokens": 114107958.0,
      "step": 56310
    },
    {
      "entropy": 5.744474267959594,
      "epoch": 4.921349296513152,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002836980106793918,
      "loss": 5.1775,
      "mean_token_accuracy": 0.21149113178253173,
      "num_tokens": 114118426.0,
      "step": 56315
    },
    {
      "entropy": 5.715318727493286,
      "epoch": 4.9217862448658565,
      "grad_norm": 1.125,
      "learning_rate": 0.00028366687279077977,
      "loss": 5.1312,
      "mean_token_accuracy": 0.22509654909372329,
      "num_tokens": 114128014.0,
      "step": 56320
    },
    {
      "entropy": 5.760357332229614,
      "epoch": 4.922223193218562,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028363573473593254,
      "loss": 5.1583,
      "mean_token_accuracy": 0.21394690573215486,
      "num_tokens": 114137827.0,
      "step": 56325
    },
    {
      "entropy": 5.740752506256103,
      "epoch": 4.922660141571266,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028360459651544735,
      "loss": 5.2494,
      "mean_token_accuracy": 0.20764364004135133,
      "num_tokens": 114148171.0,
      "step": 56330
    },
    {
      "entropy": 5.760660409927368,
      "epoch": 4.923097089923971,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002835734581299216,
      "loss": 5.1801,
      "mean_token_accuracy": 0.2116239294409752,
      "num_tokens": 114157923.0,
      "step": 56335
    },
    {
      "entropy": 5.736221408843994,
      "epoch": 4.923534038276676,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002835423195799523,
      "loss": 5.1182,
      "mean_token_accuracy": 0.22066571712493896,
      "num_tokens": 114167471.0,
      "step": 56340
    },
    {
      "entropy": 5.789042997360229,
      "epoch": 4.923970986629381,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002835111808661369,
      "loss": 5.2446,
      "mean_token_accuracy": 0.2154172956943512,
      "num_tokens": 114177337.0,
      "step": 56345
    },
    {
      "entropy": 5.676362323760986,
      "epoch": 4.924407934982085,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028348004198907255,
      "loss": 5.1493,
      "mean_token_accuracy": 0.22061306834220887,
      "num_tokens": 114187019.0,
      "step": 56350
    },
    {
      "entropy": 5.642248153686523,
      "epoch": 4.9248448833347895,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028344890294935667,
      "loss": 5.1296,
      "mean_token_accuracy": 0.22533582597970964,
      "num_tokens": 114196553.0,
      "step": 56355
    },
    {
      "entropy": 5.7867512702941895,
      "epoch": 4.925281831687495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002834177637475863,
      "loss": 5.2341,
      "mean_token_accuracy": 0.2068001940846443,
      "num_tokens": 114207096.0,
      "step": 56360
    },
    {
      "entropy": 5.724665212631225,
      "epoch": 4.925718780040199,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002833866243843587,
      "loss": 5.1551,
      "mean_token_accuracy": 0.2160997435450554,
      "num_tokens": 114218294.0,
      "step": 56365
    },
    {
      "entropy": 5.781374835968018,
      "epoch": 4.926155728392904,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028335548486027145,
      "loss": 5.1485,
      "mean_token_accuracy": 0.21823150515556336,
      "num_tokens": 114228520.0,
      "step": 56370
    },
    {
      "entropy": 5.688702249526978,
      "epoch": 4.926592676745608,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002833243451759216,
      "loss": 5.1662,
      "mean_token_accuracy": 0.21762994527816773,
      "num_tokens": 114238491.0,
      "step": 56375
    },
    {
      "entropy": 5.7986681938171385,
      "epoch": 4.927029625098314,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028329320533190637,
      "loss": 5.2072,
      "mean_token_accuracy": 0.21328654140233994,
      "num_tokens": 114248040.0,
      "step": 56380
    },
    {
      "entropy": 5.735387802124023,
      "epoch": 4.927466573451018,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028326206532882313,
      "loss": 5.1716,
      "mean_token_accuracy": 0.21439576894044876,
      "num_tokens": 114258834.0,
      "step": 56385
    },
    {
      "entropy": 5.72076268196106,
      "epoch": 4.927903521803723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028323092516726917,
      "loss": 5.1855,
      "mean_token_accuracy": 0.21387782096862792,
      "num_tokens": 114268822.0,
      "step": 56390
    },
    {
      "entropy": 5.705960369110107,
      "epoch": 4.928340470156428,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002831997848478418,
      "loss": 5.1837,
      "mean_token_accuracy": 0.21474263966083526,
      "num_tokens": 114278844.0,
      "step": 56395
    },
    {
      "entropy": 5.70147385597229,
      "epoch": 4.928777418509132,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028316864437113814,
      "loss": 5.0662,
      "mean_token_accuracy": 0.23049772679805755,
      "num_tokens": 114287361.0,
      "step": 56400
    },
    {
      "entropy": 5.744932317733765,
      "epoch": 4.929214366861837,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028313750373775565,
      "loss": 5.1916,
      "mean_token_accuracy": 0.2220519945025444,
      "num_tokens": 114297971.0,
      "step": 56405
    },
    {
      "entropy": 5.729938983917236,
      "epoch": 4.929651315214541,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028310636294829163,
      "loss": 5.1776,
      "mean_token_accuracy": 0.21016976237297058,
      "num_tokens": 114308477.0,
      "step": 56410
    },
    {
      "entropy": 5.728144788742066,
      "epoch": 4.930088263567247,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002830752220033433,
      "loss": 5.1346,
      "mean_token_accuracy": 0.21590136736631393,
      "num_tokens": 114317453.0,
      "step": 56415
    },
    {
      "entropy": 5.784225273132324,
      "epoch": 4.930525211919951,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028304408090350794,
      "loss": 5.2345,
      "mean_token_accuracy": 0.20514058023691178,
      "num_tokens": 114326320.0,
      "step": 56420
    },
    {
      "entropy": 5.754023551940918,
      "epoch": 4.930962160272656,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028301293964938285,
      "loss": 5.1685,
      "mean_token_accuracy": 0.21619645208120347,
      "num_tokens": 114336552.0,
      "step": 56425
    },
    {
      "entropy": 5.764206171035767,
      "epoch": 4.93139910862536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002829817982415654,
      "loss": 5.1739,
      "mean_token_accuracy": 0.21224773079156875,
      "num_tokens": 114346711.0,
      "step": 56430
    },
    {
      "entropy": 5.797005891799927,
      "epoch": 4.9318360569780655,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000282950656680653,
      "loss": 5.2487,
      "mean_token_accuracy": 0.2121853157877922,
      "num_tokens": 114356000.0,
      "step": 56435
    },
    {
      "entropy": 5.764550352096558,
      "epoch": 4.93227300533077,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028291951496724275,
      "loss": 5.1789,
      "mean_token_accuracy": 0.21709613800048827,
      "num_tokens": 114366727.0,
      "step": 56440
    },
    {
      "entropy": 5.807846593856811,
      "epoch": 4.932709953683474,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028288837310193213,
      "loss": 5.1883,
      "mean_token_accuracy": 0.21171141415834427,
      "num_tokens": 114376880.0,
      "step": 56445
    },
    {
      "entropy": 5.813094758987427,
      "epoch": 4.93314690203618,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028285723108531827,
      "loss": 5.2162,
      "mean_token_accuracy": 0.20080817341804505,
      "num_tokens": 114386823.0,
      "step": 56450
    },
    {
      "entropy": 5.743796634674072,
      "epoch": 4.933583850388884,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002828260889179986,
      "loss": 5.2111,
      "mean_token_accuracy": 0.20964843928813934,
      "num_tokens": 114396335.0,
      "step": 56455
    },
    {
      "entropy": 5.770114755630493,
      "epoch": 4.934020798741589,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028279494660057056,
      "loss": 5.1671,
      "mean_token_accuracy": 0.20887496769428254,
      "num_tokens": 114407196.0,
      "step": 56460
    },
    {
      "entropy": 5.788804292678833,
      "epoch": 4.934457747094293,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002827638041336312,
      "loss": 5.1783,
      "mean_token_accuracy": 0.21169188618659973,
      "num_tokens": 114417226.0,
      "step": 56465
    },
    {
      "entropy": 5.768339967727661,
      "epoch": 4.9348946954469985,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002827326615177781,
      "loss": 5.1582,
      "mean_token_accuracy": 0.21694118678569793,
      "num_tokens": 114427197.0,
      "step": 56470
    },
    {
      "entropy": 5.720392274856567,
      "epoch": 4.935331643799703,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002827015187536085,
      "loss": 5.2144,
      "mean_token_accuracy": 0.2107618659734726,
      "num_tokens": 114438509.0,
      "step": 56475
    },
    {
      "entropy": 5.752692317962646,
      "epoch": 4.9357685921524075,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028267037584171973,
      "loss": 5.2049,
      "mean_token_accuracy": 0.21351480931043626,
      "num_tokens": 114448660.0,
      "step": 56480
    },
    {
      "entropy": 5.748010969161987,
      "epoch": 4.936205540505112,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002826392327827091,
      "loss": 5.1618,
      "mean_token_accuracy": 0.21799768060445784,
      "num_tokens": 114458869.0,
      "step": 56485
    },
    {
      "entropy": 5.69458498954773,
      "epoch": 4.936642488857817,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002826080895771741,
      "loss": 5.1182,
      "mean_token_accuracy": 0.2226152166724205,
      "num_tokens": 114468572.0,
      "step": 56490
    },
    {
      "entropy": 5.750216817855835,
      "epoch": 4.937079437210522,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002825769462257119,
      "loss": 5.2261,
      "mean_token_accuracy": 0.20866837948560715,
      "num_tokens": 114478692.0,
      "step": 56495
    },
    {
      "entropy": 5.837002372741699,
      "epoch": 4.937516385563226,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028254580272891983,
      "loss": 5.2926,
      "mean_token_accuracy": 0.20229814052581788,
      "num_tokens": 114490578.0,
      "step": 56500
    },
    {
      "entropy": 5.861393594741822,
      "epoch": 4.937953333915932,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002825146590873953,
      "loss": 5.2107,
      "mean_token_accuracy": 0.2078648865222931,
      "num_tokens": 114500979.0,
      "step": 56505
    },
    {
      "entropy": 5.768359804153443,
      "epoch": 4.938390282268636,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002824835153017358,
      "loss": 5.1829,
      "mean_token_accuracy": 0.21431590020656585,
      "num_tokens": 114510437.0,
      "step": 56510
    },
    {
      "entropy": 5.7093792915344235,
      "epoch": 4.9388272306213405,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028245237137253845,
      "loss": 5.1893,
      "mean_token_accuracy": 0.21930414140224458,
      "num_tokens": 114521458.0,
      "step": 56515
    },
    {
      "entropy": 5.667838430404663,
      "epoch": 4.939264178974045,
      "grad_norm": 1.125,
      "learning_rate": 0.0002824212273004007,
      "loss": 5.0532,
      "mean_token_accuracy": 0.22762400060892105,
      "num_tokens": 114531134.0,
      "step": 56520
    },
    {
      "entropy": 5.706734943389892,
      "epoch": 4.9397011273267495,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028239008308591996,
      "loss": 5.1373,
      "mean_token_accuracy": 0.2205590307712555,
      "num_tokens": 114541671.0,
      "step": 56525
    },
    {
      "entropy": 5.664160346984863,
      "epoch": 4.940138075679455,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002823589387296936,
      "loss": 5.1499,
      "mean_token_accuracy": 0.21688908636569976,
      "num_tokens": 114551454.0,
      "step": 56530
    },
    {
      "entropy": 5.738099765777588,
      "epoch": 4.940575024032159,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028232779423231893,
      "loss": 5.1706,
      "mean_token_accuracy": 0.20977382957935334,
      "num_tokens": 114561168.0,
      "step": 56535
    },
    {
      "entropy": 5.708716344833374,
      "epoch": 4.941011972384864,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002822966495943932,
      "loss": 5.12,
      "mean_token_accuracy": 0.2235735684633255,
      "num_tokens": 114570164.0,
      "step": 56540
    },
    {
      "entropy": 5.6076446056365965,
      "epoch": 4.941448920737569,
      "grad_norm": 0.953125,
      "learning_rate": 0.000282265504816514,
      "loss": 5.0548,
      "mean_token_accuracy": 0.22935943007469178,
      "num_tokens": 114582608.0,
      "step": 56545
    },
    {
      "entropy": 5.771885251998901,
      "epoch": 4.941885869090274,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028223435989927864,
      "loss": 5.2224,
      "mean_token_accuracy": 0.20742458701133729,
      "num_tokens": 114594768.0,
      "step": 56550
    },
    {
      "entropy": 5.810441970825195,
      "epoch": 4.942322817442978,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002822032148432844,
      "loss": 5.1706,
      "mean_token_accuracy": 0.21266455948352814,
      "num_tokens": 114605023.0,
      "step": 56555
    },
    {
      "entropy": 5.7728101253509525,
      "epoch": 4.942759765795683,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028217206964912877,
      "loss": 5.1924,
      "mean_token_accuracy": 0.21318964809179305,
      "num_tokens": 114615794.0,
      "step": 56560
    },
    {
      "entropy": 5.707527542114258,
      "epoch": 4.943196714148388,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028214092431740913,
      "loss": 5.1541,
      "mean_token_accuracy": 0.2170703738927841,
      "num_tokens": 114625389.0,
      "step": 56565
    },
    {
      "entropy": 5.7424072265625,
      "epoch": 4.943633662501092,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002821097788487227,
      "loss": 5.2087,
      "mean_token_accuracy": 0.21064504832029343,
      "num_tokens": 114635594.0,
      "step": 56570
    },
    {
      "entropy": 5.778677606582642,
      "epoch": 4.944070610853797,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028207863324366704,
      "loss": 5.1596,
      "mean_token_accuracy": 0.21820871233940126,
      "num_tokens": 114645730.0,
      "step": 56575
    },
    {
      "entropy": 5.786246490478516,
      "epoch": 4.944507559206501,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028204748750283953,
      "loss": 5.1311,
      "mean_token_accuracy": 0.21363750100135803,
      "num_tokens": 114655048.0,
      "step": 56580
    },
    {
      "entropy": 5.725740718841553,
      "epoch": 4.944944507559207,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028201634162683756,
      "loss": 5.2015,
      "mean_token_accuracy": 0.21523246318101882,
      "num_tokens": 114665965.0,
      "step": 56585
    },
    {
      "entropy": 5.663286304473877,
      "epoch": 4.945381455911911,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028198519561625833,
      "loss": 5.0889,
      "mean_token_accuracy": 0.2255419000983238,
      "num_tokens": 114676395.0,
      "step": 56590
    },
    {
      "entropy": 5.780646896362304,
      "epoch": 4.945818404264616,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002819540494716995,
      "loss": 5.2065,
      "mean_token_accuracy": 0.2084085077047348,
      "num_tokens": 114687302.0,
      "step": 56595
    },
    {
      "entropy": 5.705309629440308,
      "epoch": 4.946255352617321,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028192290319375827,
      "loss": 5.118,
      "mean_token_accuracy": 0.2221345692873001,
      "num_tokens": 114697252.0,
      "step": 56600
    },
    {
      "entropy": 5.717145919799805,
      "epoch": 4.946692300970025,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028189175678303217,
      "loss": 5.1557,
      "mean_token_accuracy": 0.21884154081344603,
      "num_tokens": 114706997.0,
      "step": 56605
    },
    {
      "entropy": 5.700933313369751,
      "epoch": 4.94712924932273,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028186061024011857,
      "loss": 5.171,
      "mean_token_accuracy": 0.211091448366642,
      "num_tokens": 114716452.0,
      "step": 56610
    },
    {
      "entropy": 5.722066259384155,
      "epoch": 4.947566197675435,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002818294635656149,
      "loss": 5.1988,
      "mean_token_accuracy": 0.21549075692892075,
      "num_tokens": 114726763.0,
      "step": 56615
    },
    {
      "entropy": 5.8294562816619875,
      "epoch": 4.94800314602814,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028179831676011845,
      "loss": 5.2526,
      "mean_token_accuracy": 0.21180299371480943,
      "num_tokens": 114736903.0,
      "step": 56620
    },
    {
      "entropy": 5.783427429199219,
      "epoch": 4.948440094380844,
      "grad_norm": 1.25,
      "learning_rate": 0.0002817671698242268,
      "loss": 5.2237,
      "mean_token_accuracy": 0.2125150665640831,
      "num_tokens": 114746125.0,
      "step": 56625
    },
    {
      "entropy": 5.734074401855469,
      "epoch": 4.948877042733549,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028173602275853734,
      "loss": 5.1541,
      "mean_token_accuracy": 0.2203684166073799,
      "num_tokens": 114755576.0,
      "step": 56630
    },
    {
      "entropy": 5.683131456375122,
      "epoch": 4.949313991086253,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002817048755636474,
      "loss": 5.1316,
      "mean_token_accuracy": 0.22245802730321884,
      "num_tokens": 114765982.0,
      "step": 56635
    },
    {
      "entropy": 5.703208684921265,
      "epoch": 4.9497509394389585,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028167372824015435,
      "loss": 5.1773,
      "mean_token_accuracy": 0.21385760456323624,
      "num_tokens": 114776434.0,
      "step": 56640
    },
    {
      "entropy": 5.758041954040527,
      "epoch": 4.950187887791663,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028164258078865575,
      "loss": 5.21,
      "mean_token_accuracy": 0.21092578917741775,
      "num_tokens": 114786613.0,
      "step": 56645
    },
    {
      "entropy": 5.743562412261963,
      "epoch": 4.950624836144367,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028161143320974897,
      "loss": 5.1429,
      "mean_token_accuracy": 0.21520970314741134,
      "num_tokens": 114796276.0,
      "step": 56650
    },
    {
      "entropy": 5.765290451049805,
      "epoch": 4.951061784497073,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002815802855040314,
      "loss": 5.1594,
      "mean_token_accuracy": 0.21988416314125062,
      "num_tokens": 114807535.0,
      "step": 56655
    },
    {
      "entropy": 5.693089199066162,
      "epoch": 4.951498732849777,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002815491376721006,
      "loss": 5.0882,
      "mean_token_accuracy": 0.2184676021337509,
      "num_tokens": 114817004.0,
      "step": 56660
    },
    {
      "entropy": 5.773602485656738,
      "epoch": 4.951935681202482,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028151798971455376,
      "loss": 5.3041,
      "mean_token_accuracy": 0.20360438376665116,
      "num_tokens": 114826980.0,
      "step": 56665
    },
    {
      "entropy": 5.704639005661011,
      "epoch": 4.952372629555186,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028148684163198856,
      "loss": 5.1511,
      "mean_token_accuracy": 0.21903940588235854,
      "num_tokens": 114837406.0,
      "step": 56670
    },
    {
      "entropy": 5.747394371032715,
      "epoch": 4.952809577907892,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002814556934250024,
      "loss": 5.1641,
      "mean_token_accuracy": 0.2192075416445732,
      "num_tokens": 114847405.0,
      "step": 56675
    },
    {
      "entropy": 5.770439720153808,
      "epoch": 4.953246526260596,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028142454509419256,
      "loss": 5.1701,
      "mean_token_accuracy": 0.2201479122042656,
      "num_tokens": 114857317.0,
      "step": 56680
    },
    {
      "entropy": 5.743255949020385,
      "epoch": 4.9536834746133005,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002813933966401566,
      "loss": 5.2026,
      "mean_token_accuracy": 0.2105194479227066,
      "num_tokens": 114866579.0,
      "step": 56685
    },
    {
      "entropy": 5.707648420333863,
      "epoch": 4.954120422966005,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002813622480634919,
      "loss": 5.2231,
      "mean_token_accuracy": 0.21125770658254622,
      "num_tokens": 114875638.0,
      "step": 56690
    },
    {
      "entropy": 5.7519324779510494,
      "epoch": 4.95455737131871,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000281331099364796,
      "loss": 5.1832,
      "mean_token_accuracy": 0.2150248870253563,
      "num_tokens": 114885496.0,
      "step": 56695
    },
    {
      "entropy": 5.763845682144165,
      "epoch": 4.954994319671415,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028129995054466635,
      "loss": 5.1137,
      "mean_token_accuracy": 0.21756340563297272,
      "num_tokens": 114894733.0,
      "step": 56700
    },
    {
      "entropy": 5.771236705780029,
      "epoch": 4.955431268024119,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002812688016037003,
      "loss": 5.2033,
      "mean_token_accuracy": 0.21236818730831147,
      "num_tokens": 114904991.0,
      "step": 56705
    },
    {
      "entropy": 5.718814182281494,
      "epoch": 4.955868216376825,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002812376525424953,
      "loss": 5.1951,
      "mean_token_accuracy": 0.21766597479581834,
      "num_tokens": 114915296.0,
      "step": 56710
    },
    {
      "entropy": 5.782224941253662,
      "epoch": 4.956305164729529,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002812065033616489,
      "loss": 5.2081,
      "mean_token_accuracy": 0.21054580956697463,
      "num_tokens": 114924614.0,
      "step": 56715
    },
    {
      "entropy": 5.676708221435547,
      "epoch": 4.956742113082234,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028117535406175847,
      "loss": 5.1391,
      "mean_token_accuracy": 0.22152451127767564,
      "num_tokens": 114934252.0,
      "step": 56720
    },
    {
      "entropy": 5.757295083999634,
      "epoch": 4.957179061434938,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002811442046434216,
      "loss": 5.2022,
      "mean_token_accuracy": 0.21221910566091537,
      "num_tokens": 114944163.0,
      "step": 56725
    },
    {
      "entropy": 5.733352327346802,
      "epoch": 4.957616009787643,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002811130551072356,
      "loss": 5.0767,
      "mean_token_accuracy": 0.21919770538806915,
      "num_tokens": 114953844.0,
      "step": 56730
    },
    {
      "entropy": 5.687009811401367,
      "epoch": 4.958052958140348,
      "grad_norm": 1.09375,
      "learning_rate": 0.000281081905453798,
      "loss": 5.0861,
      "mean_token_accuracy": 0.2260487973690033,
      "num_tokens": 114964395.0,
      "step": 56735
    },
    {
      "entropy": 5.7776342868804935,
      "epoch": 4.958489906493052,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002810507556837063,
      "loss": 5.2887,
      "mean_token_accuracy": 0.20675494968891145,
      "num_tokens": 114975479.0,
      "step": 56740
    },
    {
      "entropy": 5.7004969120025635,
      "epoch": 4.958926854845757,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002810196057975579,
      "loss": 5.135,
      "mean_token_accuracy": 0.22184014171361924,
      "num_tokens": 114984580.0,
      "step": 56745
    },
    {
      "entropy": 5.771249723434448,
      "epoch": 4.959363803198462,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002809884557959503,
      "loss": 5.1075,
      "mean_token_accuracy": 0.22228193879127503,
      "num_tokens": 114994623.0,
      "step": 56750
    },
    {
      "entropy": 5.733172225952148,
      "epoch": 4.959800751551167,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028095730567948105,
      "loss": 5.1418,
      "mean_token_accuracy": 0.2190784066915512,
      "num_tokens": 115003593.0,
      "step": 56755
    },
    {
      "entropy": 5.730103206634522,
      "epoch": 4.960237699903871,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002809261554487475,
      "loss": 5.2121,
      "mean_token_accuracy": 0.21062477082014083,
      "num_tokens": 115013781.0,
      "step": 56760
    },
    {
      "entropy": 5.752256870269775,
      "epoch": 4.9606746482565764,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028089500510434726,
      "loss": 5.2481,
      "mean_token_accuracy": 0.2067915081977844,
      "num_tokens": 115023392.0,
      "step": 56765
    },
    {
      "entropy": 5.710125255584717,
      "epoch": 4.961111596609281,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028086385464687764,
      "loss": 5.2156,
      "mean_token_accuracy": 0.21442446261644363,
      "num_tokens": 115033774.0,
      "step": 56770
    },
    {
      "entropy": 5.792310810089111,
      "epoch": 4.961548544961985,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002808327040769362,
      "loss": 5.1909,
      "mean_token_accuracy": 0.21181420534849166,
      "num_tokens": 115044536.0,
      "step": 56775
    },
    {
      "entropy": 5.756363296508789,
      "epoch": 4.96198549331469,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028080155339512043,
      "loss": 5.1142,
      "mean_token_accuracy": 0.21611283719539642,
      "num_tokens": 115055611.0,
      "step": 56780
    },
    {
      "entropy": 5.646006155014038,
      "epoch": 4.962422441667395,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002807704026020279,
      "loss": 5.1096,
      "mean_token_accuracy": 0.22966729551553727,
      "num_tokens": 115065082.0,
      "step": 56785
    },
    {
      "entropy": 5.740005970001221,
      "epoch": 4.9628593900201,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028073925169825605,
      "loss": 5.2068,
      "mean_token_accuracy": 0.20740694403648377,
      "num_tokens": 115074680.0,
      "step": 56790
    },
    {
      "entropy": 5.77784538269043,
      "epoch": 4.963296338372804,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002807081006844023,
      "loss": 5.215,
      "mean_token_accuracy": 0.21468712240457535,
      "num_tokens": 115083966.0,
      "step": 56795
    },
    {
      "entropy": 5.786445379257202,
      "epoch": 4.963733286725509,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028067694956106416,
      "loss": 5.1871,
      "mean_token_accuracy": 0.21077759116888045,
      "num_tokens": 115093962.0,
      "step": 56800
    },
    {
      "entropy": 5.692618322372437,
      "epoch": 4.964170235078214,
      "grad_norm": 1.125,
      "learning_rate": 0.0002806457983288392,
      "loss": 5.1312,
      "mean_token_accuracy": 0.21920888721942902,
      "num_tokens": 115104401.0,
      "step": 56805
    },
    {
      "entropy": 5.716006803512573,
      "epoch": 4.9646071834309184,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002806146469883248,
      "loss": 5.1785,
      "mean_token_accuracy": 0.2133341670036316,
      "num_tokens": 115114705.0,
      "step": 56810
    },
    {
      "entropy": 5.782796478271484,
      "epoch": 4.965044131783623,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002805834955401186,
      "loss": 5.2244,
      "mean_token_accuracy": 0.21394257843494416,
      "num_tokens": 115124499.0,
      "step": 56815
    },
    {
      "entropy": 5.714519739151001,
      "epoch": 4.965481080136328,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000280552343984818,
      "loss": 5.129,
      "mean_token_accuracy": 0.22031180113554,
      "num_tokens": 115134285.0,
      "step": 56820
    },
    {
      "entropy": 5.726344060897827,
      "epoch": 4.965918028489033,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002805211923230205,
      "loss": 5.2359,
      "mean_token_accuracy": 0.21121084094047546,
      "num_tokens": 115144984.0,
      "step": 56825
    },
    {
      "entropy": 5.800294828414917,
      "epoch": 4.966354976841737,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028049004055532374,
      "loss": 5.2119,
      "mean_token_accuracy": 0.21780136823654175,
      "num_tokens": 115154375.0,
      "step": 56830
    },
    {
      "entropy": 5.730953216552734,
      "epoch": 4.966791925194442,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028045888868232497,
      "loss": 5.1283,
      "mean_token_accuracy": 0.22190682888031005,
      "num_tokens": 115164433.0,
      "step": 56835
    },
    {
      "entropy": 5.72056770324707,
      "epoch": 4.967228873547147,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028042773670462203,
      "loss": 5.165,
      "mean_token_accuracy": 0.21702236980199813,
      "num_tokens": 115174235.0,
      "step": 56840
    },
    {
      "entropy": 5.685868883132935,
      "epoch": 4.9676658218998515,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028039658462281216,
      "loss": 5.1057,
      "mean_token_accuracy": 0.23155165761709212,
      "num_tokens": 115184149.0,
      "step": 56845
    },
    {
      "entropy": 5.6800782680511475,
      "epoch": 4.968102770252556,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028036543243749296,
      "loss": 5.1078,
      "mean_token_accuracy": 0.21905492395162582,
      "num_tokens": 115195093.0,
      "step": 56850
    },
    {
      "entropy": 5.719652462005615,
      "epoch": 4.9685397186052604,
      "grad_norm": 1.15625,
      "learning_rate": 0.000280334280149262,
      "loss": 5.1942,
      "mean_token_accuracy": 0.21983992159366608,
      "num_tokens": 115205972.0,
      "step": 56855
    },
    {
      "entropy": 5.7276195049285885,
      "epoch": 4.968976666957966,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028030312775871667,
      "loss": 5.1785,
      "mean_token_accuracy": 0.2170127362012863,
      "num_tokens": 115216260.0,
      "step": 56860
    },
    {
      "entropy": 5.704217004776001,
      "epoch": 4.96941361531067,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028027197526645463,
      "loss": 5.1219,
      "mean_token_accuracy": 0.21582517623901368,
      "num_tokens": 115225845.0,
      "step": 56865
    },
    {
      "entropy": 5.789033508300781,
      "epoch": 4.969850563663375,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002802408226730734,
      "loss": 5.2332,
      "mean_token_accuracy": 0.2022304505109787,
      "num_tokens": 115235951.0,
      "step": 56870
    },
    {
      "entropy": 5.724395036697388,
      "epoch": 4.97028751201608,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028020966997917035,
      "loss": 5.1736,
      "mean_token_accuracy": 0.21551339477300643,
      "num_tokens": 115246382.0,
      "step": 56875
    },
    {
      "entropy": 5.696929836273194,
      "epoch": 4.970724460368785,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002801785171853431,
      "loss": 5.093,
      "mean_token_accuracy": 0.22368303090333938,
      "num_tokens": 115256566.0,
      "step": 56880
    },
    {
      "entropy": 5.693652200698852,
      "epoch": 4.971161408721489,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028014736429218926,
      "loss": 5.1227,
      "mean_token_accuracy": 0.22098102569580078,
      "num_tokens": 115266506.0,
      "step": 56885
    },
    {
      "entropy": 5.737418746948242,
      "epoch": 4.9715983570741935,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028011621130030626,
      "loss": 5.1978,
      "mean_token_accuracy": 0.2091674566268921,
      "num_tokens": 115276671.0,
      "step": 56890
    },
    {
      "entropy": 5.7528845310211185,
      "epoch": 4.972035305426899,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002800850582102916,
      "loss": 5.203,
      "mean_token_accuracy": 0.2123747169971466,
      "num_tokens": 115285911.0,
      "step": 56895
    },
    {
      "entropy": 5.636626577377319,
      "epoch": 4.972472253779603,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028005390502274283,
      "loss": 5.0626,
      "mean_token_accuracy": 0.23333537876605986,
      "num_tokens": 115294931.0,
      "step": 56900
    },
    {
      "entropy": 5.691507816314697,
      "epoch": 4.972909202132308,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002800227517382576,
      "loss": 5.2085,
      "mean_token_accuracy": 0.20901839286088944,
      "num_tokens": 115304568.0,
      "step": 56905
    },
    {
      "entropy": 5.710832595825195,
      "epoch": 4.973346150485012,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002799915983574333,
      "loss": 5.1855,
      "mean_token_accuracy": 0.2162283256649971,
      "num_tokens": 115315067.0,
      "step": 56910
    },
    {
      "entropy": 5.7298603534698485,
      "epoch": 4.973783098837718,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002799604448808676,
      "loss": 5.1641,
      "mean_token_accuracy": 0.21687965989112853,
      "num_tokens": 115324963.0,
      "step": 56915
    },
    {
      "entropy": 5.7498589038848875,
      "epoch": 4.974220047190422,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002799292913091579,
      "loss": 5.2109,
      "mean_token_accuracy": 0.21175618171691896,
      "num_tokens": 115335671.0,
      "step": 56920
    },
    {
      "entropy": 5.759288501739502,
      "epoch": 4.974656995543127,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002798981376429018,
      "loss": 5.1368,
      "mean_token_accuracy": 0.2158793106675148,
      "num_tokens": 115345020.0,
      "step": 56925
    },
    {
      "entropy": 5.703660583496093,
      "epoch": 4.975093943895832,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002798669838826968,
      "loss": 5.1693,
      "mean_token_accuracy": 0.21734155118465423,
      "num_tokens": 115354584.0,
      "step": 56930
    },
    {
      "entropy": 5.767865705490112,
      "epoch": 4.975530892248536,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002798358300291407,
      "loss": 5.2131,
      "mean_token_accuracy": 0.204568549990654,
      "num_tokens": 115365451.0,
      "step": 56935
    },
    {
      "entropy": 5.783949661254883,
      "epoch": 4.975967840601241,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027980467608283065,
      "loss": 5.2231,
      "mean_token_accuracy": 0.2081458881497383,
      "num_tokens": 115375073.0,
      "step": 56940
    },
    {
      "entropy": 5.727159690856934,
      "epoch": 4.976404788953945,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027977352204436453,
      "loss": 5.1552,
      "mean_token_accuracy": 0.21458992063999177,
      "num_tokens": 115385564.0,
      "step": 56945
    },
    {
      "entropy": 5.756913805007935,
      "epoch": 4.976841737306651,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002797423679143396,
      "loss": 5.2583,
      "mean_token_accuracy": 0.2085264503955841,
      "num_tokens": 115397118.0,
      "step": 56950
    },
    {
      "entropy": 5.708488416671753,
      "epoch": 4.977278685659355,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002797112136933537,
      "loss": 5.1541,
      "mean_token_accuracy": 0.22137040942907332,
      "num_tokens": 115407681.0,
      "step": 56955
    },
    {
      "entropy": 5.720679140090942,
      "epoch": 4.97771563401206,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002796800593820041,
      "loss": 5.1182,
      "mean_token_accuracy": 0.21701388210058212,
      "num_tokens": 115417449.0,
      "step": 56960
    },
    {
      "entropy": 5.700461769104004,
      "epoch": 4.978152582364764,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027964890498088875,
      "loss": 5.0834,
      "mean_token_accuracy": 0.22064057290554046,
      "num_tokens": 115426932.0,
      "step": 56965
    },
    {
      "entropy": 5.805937433242798,
      "epoch": 4.9785895307174695,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002796177504906048,
      "loss": 5.3293,
      "mean_token_accuracy": 0.1994038552045822,
      "num_tokens": 115437381.0,
      "step": 56970
    },
    {
      "entropy": 5.718894290924072,
      "epoch": 4.979026479070174,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027958659591175,
      "loss": 5.2886,
      "mean_token_accuracy": 0.21079328060150146,
      "num_tokens": 115448452.0,
      "step": 56975
    },
    {
      "entropy": 5.788496255874634,
      "epoch": 4.979463427422878,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027955544124492196,
      "loss": 5.1889,
      "mean_token_accuracy": 0.20149825960397721,
      "num_tokens": 115458466.0,
      "step": 56980
    },
    {
      "entropy": 5.748788356781006,
      "epoch": 4.979900375775584,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002795242864907181,
      "loss": 5.228,
      "mean_token_accuracy": 0.20778121501207353,
      "num_tokens": 115468330.0,
      "step": 56985
    },
    {
      "entropy": 5.741438865661621,
      "epoch": 4.980337324128288,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00027949313164973605,
      "loss": 5.1577,
      "mean_token_accuracy": 0.2215763196349144,
      "num_tokens": 115478385.0,
      "step": 56990
    },
    {
      "entropy": 5.780423927307129,
      "epoch": 4.980774272480993,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002794619767225733,
      "loss": 5.2974,
      "mean_token_accuracy": 0.20569333881139756,
      "num_tokens": 115488907.0,
      "step": 56995
    },
    {
      "entropy": 5.6749004364013675,
      "epoch": 4.981211220833697,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002794308217098276,
      "loss": 5.1125,
      "mean_token_accuracy": 0.22525210976600646,
      "num_tokens": 115498399.0,
      "step": 57000
    },
    {
      "epoch": 4.981211220833697,
      "eval_entropy": 5.438720058153019,
      "eval_loss": 5.252114772796631,
      "eval_mean_token_accuracy": 0.2195909166982732,
      "eval_num_tokens": 115498399.0,
      "eval_runtime": 28.2719,
      "eval_samples_per_second": 1309.214,
      "eval_steps_per_second": 163.661,
      "step": 57000
    }
  ],
  "logging_steps": 5,
  "max_steps": 114420,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.72845675854208e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}